Dams and Development

Home , Aquaculture in Tonga, Bakolori Dam, International Rivers, Itaipu Dam, Kariba Dam, Philip Fearnside

DAMS AND DEVELOPMENT

A N E W F R A M E WO R K FOR DECISION-MAKING

THE REPORT OF THE WORLD COMMISSION ON DAMS

November 2000

Earthscan Publications Ltd, London and Sterling, VA First published in the UK and USA in 2000 by Earthscan Publications Ltd

A catalogue record for this book is available from the British Library

ISBN: 1-85383-798-9 paperback 1-85383-797-0 hardback

Printed and bound in the UK by Thanet Press, Margate, Kent Page design and layout by Page Arts, Cape Town Cover design by Page Arts, Cape Town Photo collages by Shane Smitsdorp, Cape Town

Cover photographs © Gallo Images. The Image Bank: Horowitz, Ross. I-Afrika: Miller, Eric.

Photo credits: Cape Argus: Schronen Johan, p226; Tromp, Dion, p66, 258; Yeld, John, p36. Cape Photo Library: Proust, Alain, p36; Stoffel-Wialliame, p72. Cosmi Corporation: p16, 19, 22, 49, 72, 80, 96, 108, 125, 134,157, 196, 250. I-Afrika: Bosch, Rodger, p14; Ingram Andrew, p96, 112; Miller, Eric, p2, 24, 96, 98, 104, 111, 114, 117, 129, 142, 153, 196, 201, 220; Pettersson, Anders, p243. IUCN -The World Conservation Union, p72, 85,87. Itaipú Binacional, p36, 68, 258. The Image Bank: Horowitz, Ross, p196, 206; Sund, Harald, p134, 149. United States Bureau of Reclamation, p36. World Bank, p96,100. World Wide Fund for Nature: Gunther, Michel, p72, 74,134,137; Harvey, Martin, p78. WWF-Canon; Burgler, Roel, p205, 212; Pratginestos, Juan, p72. Rautkari, Mauri; p145, Revesz, Tamás, p20, 210; Thorsell, James W, p212. Schafer, Kevin, p72, 92, 233; Torres, William H, p136.

Earthscan Publications Ltd 120 Pentonville Road, London, N1 9JN, UK Tel: +44 (0)20 7278 0433 Fax: +44 (0)20 7278 1142 Email: [email protected] http://www.earthscan.co.uk

22883 Quicksilver Drive, Sterling, VA 20166-2012, USA

Earthscan is an editorially independent subsidiary of Kogan Page Ltd and publishes in association with WWF-UK and the International Institute for Environment and Development

This book is printed on elemental chlorine-free paper Preface

Chair’s Preface Globalisation From Below

f politics is the art of the possible, Ithis document is a work of art. It redefines what is possible to all of us, for all of us, at a time when water pressure on governments has never

been more intense. Consider: on this blue planet, less than 2.5% of our water is fresh, less than 33% of fresh

water is fluid, less than 1.7% of fluid water runs in streams. And we have been stopping even these. We dammed

half our world’s rivers at unprecedented rates of one per hour, and at unprecedented scales of over 45 000

Professor Kader Asmal Chair, World Commission on Dams dams more than four storeys high.

The Report of the World Commission on Dams i Preface

As one who authorised the next stage of one Some see in our scarcity a harbinger of of the largest dams in the Southern Hemi- troubled waters to come. They believe water sphere I can argue that nations build large scarcity inevitably locks peoples, regions and dams for sound reasons. Dams store, use and nations in a fierce, competitive struggle in divert water for consumption, irrigation, which restless millions race to the bottom in cooling, transportation, construction, mills, fear and self-interest. And thus, they main- power and recreation. Dams remove water tain, when rivers cross borders within or from the Ganges, Amazon, Danube, Nile or between nations, water scarcity leads to Columbia to sustain cities on their banks. water stress which leads to water wars. For parting – or imparting – the waters, dams are our oldest tool. Yet are they our Our Commission, and through it, this Final only tool, or our best option? Report, contradicts that sentiment. We see water as an instrument, a catalyst for peace, The World Commission on Dams has that brings us together, neither to build undertaken a rigorous, independent and dams nor tear them down but to carefully inclusive global review, testing the waters to develop resources for the long term. help you answer that question with authority. But just as water scarcity drove previous Easier said than done? Not necessarily. The construction of dams, competition for water hard part here may lie in what can be ‘said’ has underscored the Commission’s work. As with intellectual honesty, vision and under- we seek water we face an escalating crisis, standing. Surprisingly such a statement has even of biblical proportions. In Ecclesiastes, rarely been attempted. Unlike every other recall the passage: aspect of our lives, large dams have long escaped deep and clear and impartial One generation passeth away, scrutiny into the process by which they and another generation cometh: emerge and are valued. This lapse is espe- but the earth abideth always... cially glaring when set against much smaller All rivers runneth to the sea, scale investments. We daily squeeze and yet the sea is not full... weigh fruits and vegetables to ensure we get a fair return at the market. We rigorously The words are beautiful, haunting and, test-drive and analyse the performance of suddenly, anachronistic. For they are not motor cars before and after paying a few true due to demands and dams during our thousand dollars for one. We conduct lives. Even degraded rivers seldom totally thorough due-diligence before purchase of run, but loiter in a chain of reservoirs. In either house or business. some years our mightiest rivers – Africa’s Nile, Asia’s Yellow, America’s Colorado, Yet this century we have collectively Australia’s Murray – do not reach the sea. bought, on average, one large dam per day, and there have been precious few, if any, Compounding that shortage, one in five comprehensive, independent analyses as to persons world-wide lacks access to safe why dams came about, how dams perform drinking water. Half the world lacks sanita- over time, and whether we are getting a fair tion; millions die from waterborne diseases. return from our $2 trillion investment. Farmers compete for water with booming but stressed cities. Towns drain aquifers that Until now. took centuries to fill. Saltwater pollutes groundwater miles from the sea. In China, Pioneer efforts are bound to be controver- Mexico and India water tables fall a metre a sial. And while the World Commission on year. In a few decades, as we seek a fifth Dams is, by design, strictly advisory, make more water for 3 billion new people, one in no mistake. Our genesis, work process and three of us may struggle to drink or bathe. implications of this Final Report are by

ii Dams and Development: A New Framework for Decision-Making Preface

nature quite political. Our mandate involves philosophical divides of the debate. That is the most precious element on earth, and its true strength. Indeed, the assumption that, of course, involves power: who wields that the number of people who write some- it, how to share it, which ways the state may thing is inversely related to how much it better balance it. says only goes so far. Hundreds of eloquent books and publications inform the dams and Some may feel this Report makes water use development debate from one side or the decisions even more difficult; by raising the other. Written by single authors, they say bar higher, as we do, a government must much but resolve little. The WCD Final exercise more energy and creativity to reach Report is meaningful, and will be remem- a sustainable result. But in truth we make bered, not just for what it says, but for the those decisions easier; for we show clearly fact that several hundred diverse men and which, how, where and why decisions can women were directly involved in saying it. It either work well or fail to deliver. is sculpted by the expertise of members of the Commission, many of whom have For that reason I assert that we are much devoted their entire lives to engineering, more than a ‘Dams Commission’. We are a environmental, social and institutional Commission to heal the deep and self- concerns of rivers and dams. All of us were inflicted wounds torn open wherever and sensitive to the needs of human develop- whenever far too few determine for far too ment as we listened to the aspirations, the many how best to develop or use water and pain and the anguish of individual people. energy resources. That is often the nature of power, and the motivation of those who Slow to speak, our Commission was quick to question it. Most recently governments, listen. Both sides of the debate gave their industry and aid agencies have been chal- perspectives: from dam officials with an lenged around the world for deciding the obligation to govern to dam affected people destiny of millions without including the with stories to tell. By airing facts we poor, or even popular majorities of countries reached a rapprochement that began in they believe to be helping. Gland, Switzerland and continued, non- stop, through hearings, consultations, case To confer legitimacy on such epochal studies, submissions and reviews covering decisions, real development must be people- roughly one thousand large dams. centred, while respecting the role of the state as mediating, and often representing, Through this process a shared understanding their interests. In the following pages we do and truth began to emerge, and with it the not endorse globalisation as led from above thin thread with which to sew the stitches by a few men. We do endorse globalisation of reconciliation. On this Commission, the as led from below by all, a new approach to first stitches came, perhaps, as a woman who global water policy and development. risks her life opposing a large dam threads the eye of the needle with an engineer who In this approach, we must deal with the past built his career designing them. Or when before we can chart a course for the future. the leader of one the world’s most powerful The integrity of our process determines the technology companies engaged with the integrity of this product, which raises a key leader of one of the world’s proud but point. I am proud to sign this work, and to dispossessed peoples. As Seattle, Toronto, guide this project, but the document you Washington, London and Berlin came hold is not, as are some Commission Re- unravelled by turbulent protests over ports, authored by its chair. globalisation and development, we quietly continued to apply stitch after stitch to sew It comes instead from many authors who a stronger, more resilient and colourful were originally separated by the cultural and tapestry.

The Report of the World Commission on Dams iii Preface

Our work is now over. With this document, Ministry of Water Affairs and Forestry under yours has begun. I wish it could be as simple Nelson Mandela. Five years of hands-on or catchy as popular manuals that offer work tempered that vision. Chairing this simplistic solutions for complex situations. Commission shattered it. We recognise all those arguments, and absorb them. But we push beyond Declara- Instead of my archetype I saw: dams built of tions, which urge: ‘Recognise human rights’ dirt and dams generating no electricity; or ‘Sustain natural resources’. dams praised by ecologists and dams despised by engineers; dams used for centuries by Telling me, a harried public official who indigenous peoples, dams boosting fisheries, must answer to 48 million restless, hungry dams causing deadly floods; dams changing and thirsty people, to ‘Ensure development river chemistry or increasing net greenhouse is sustainable and humane’ is like warning gas emissions. I saw dam benefits by-pass me, ‘Operate, but don’t inflict new wounds’. thirsty adjacent communities en route to the I know that. What I don’t know is how to city, dams exhaust and erode rich soils do it. To explain how to develop water in through water logging and salinity. I saw ways that do not exhaust either my constitu- dams displace no one, dams create wetlands ents or the resources we all depend on, we and work, dams cost thrice their budget, must go beyond platitudes. Our healing dams utterly abandoned and which had no must emerge not through anecdotes, but symbolic value. Then I saw politicians through a complex, coherent and cohesive approach rivers with ambitious, bureaucratic argument that shows clearly where we have schemes, opposed by local activists shouting, been, what happened, why we’re in conflict, ‘Save our beloved dam’. and how we can, with proper understanding, heal ourselves. No matter how much you know, or think you know, about dams, you cannot read the That involves first shedding misconcep- following report and keep your assumptions tions. Today’s demands are too complex, our intact. No matter how sceptical, you will technology too advanced, our constituency come away changed, I think, for the better. too diverse, our options too numerous to For the truth is no typical dams exist. allow just one solution. For example, imagine a typical dam. Yet the decisions that led to those dams share a great deal. Clear patterns have Perhaps you see a smooth, parabolic, con- emerged, and all parties have met. We have crete structure. It seems to generate cheap all reached agreement, established a healing electricity through turbines at its base. process that we hope will work, and set this Engineers worship it, ecologists curse it, manual before you. Read it carefully – indigenous tribes lose their culture to it. though not in one sitting – with an eye to Native fisheries plunge after construction, where it may apply to your own specific but floods decline as well. It pollutes neither needs and agenda. It is rigorous, without air nor water, provides water for nearby being rigid; it sees the State as an instru- towns, turns arid soils into rich farmland. ment of development yet recognises the People and animals were relocated, but the necessity for popular participation; it is economic returns made doing so cost-effective. dispassionate and advisory in tone, but The dam embodies ambitions of statesmen, authoritative in its practical application. but when politicians approach with their ambitious plans, apprehensive peoples hold It is said that if we do not master technolo- signs that say ‘Save our beloved river’. gy, technology will master us. In the past, our unrestrained reliance upon large dam That image was my own. It was what I technology weighed down upon us in all its envisioned when I first took over the unexamined mystery. It stood, like solid,

iv Dams and Development: A New Framework for Decision-Making Preface

divisive walls, between our left and right scarcity with confidence and assurance, banks, between the upstream and downstream knowing that water is not for fighting over. reaches of our rivers. The Commission’s work Water is for conserving. Water is for bath- is complete. And now, perhaps, technology ing. Water is for drinking. Water is for can instead be kept under our united and sharing. Water, through this report, can be democratic control, owned by all of us. In our catalyst for peace. that way we can meet the coming water

Professor Kader Asmal, Chair, World Commission on Dams

I would like to express on behalf of the Commission our particular appreciation to the following individuals They have along with many other friends, partners and contributors to the WCD process, played a vital role in enabling the Commision to fulfil its mandate.

Bruce Babbitt, Sadi Baron, Ger Bergkamp, Richard Bissell, Robert Bos, Peter Bosshard, Rodney Bridle, John Briscoe, Ian Curtis, Shripad Dharmadhikary, Bert Diphoorn, Osmar Vieira de Filho, Luis Garcia, Raymundo José S. Garrido, Pham Hong Giang, Liane Greeff, George Green, Biksham Gujja, Geir Y. Hermansen, Kaare Hoeg, Ann Jennervik, Olav Kjorven, Jean-Etienne Klimpt, Man- fred Konukiewitz, M.L Chanaphun Kridakorn, Maritta Koch- Weser, Nicholas Lapham, Donal O’Leary, Patrick McCully, David McDowell, Joseph Milewski, Reatile Mochebelele, Naoki Mori, Takehiro Nakamura, Peter van Niekerk, Raimundo Nonato do C. Silva, Tilak Ranaviraja, Frances Seymour, Aly Shady, Jaswant Singh, Jan Strömblad, Even Sund, Sardar Mohammed Tariq, Allan Taylor, Martin Ter Woort, Himanshu Thakkar, Klaus Töpfer, Dao Trong Tu, Mike J. Tumbare, Mumtaz Türfan, Michael Wiehen, James Wolfensohn, Mahmoud Abu Zeid, Tor Ziegler and Birgit Zimmerle.

The Report of the World Commission on Dams v Foreword

vi Dams and Development: A New Framework for Decision-Making Foreword

Commissioners’ Foreword

he World Commission on TDams (WCD) was born out of a small but significant IUCN- World Bank sponsored workshop in Gland, Switzerland in April 1997.

Representatives of diverse interests came together to discuss the highly controversial issues associated with

large dams. To the surprise of participants, deep-seated differences on the development benefits of large

dams did not prevent a consensus emerging. That consensus included the proposal for a World Commission

on Dams.

The Report of the World Commission on Dams vii Foreword

As Commissioners, we have been honoured opment, adopted by the UN General to serve on The WCD for the past two and a Assembly in 1986 argues that ‘development half years. Representing all sides of the is a comprehensive process aimed at the debate, we have worked as an autonomous constant improvement of the well-being of international team. Our task was to con- the entire population…’. Recently, vocal duct a rigorous, independent review of the condemnation of the globalisation process, development effectiveness of large dams, to pointing out that too many people are being Professor Kader Asmal assess alternatives and to propose practical left behind those forging ahead, has added Minister of Education, guidelines for future decision-making. support to this call for a better quality of life WCD Chair, South Africa for all of humanity, not just for some. The Since our work began in May 1998, we have United Nations Development Programme’s met on nine occasions. We have listened to Human Development Report 2000 has one anothers’ different viewpoints in a given us a timely reminder that the rights to genuine spirit of openness and desire to find security and basic freedoms, and to human a common understanding. All of us have development are two sides of the same coin found it a learning process, an enriching if and that when ‘human rights and human sometimes uncomfortable one. This consen- development advance together, they rein- sus document is the result of our work, but it force one another.’ Mr Lakshmi Chand Jain Industrial Development would be unfair to the process to suggest Services, that we have emerged completely of one As an international Commission, our WCD Vice-Chair, India mind. Individual differences still exist. process has been unique in taking on board However, we all agree on the fundamental a range of interests and opinions previously principles and values that underpin this held to be irreconcilable. We have exam- report and on the guidelines we offer for the ined evidence produced and opinions way forward. In the final drafting of our expressed by a wide range of stakeholders: report we have included “A Comment” by government agencies, project affected one Commissioner who was unable to people and non-governmental organisations, attend the final meeting. It presents that people’s movements, the dam construction Commissioner's additional views on the industry, the export credit agencies and Dr Judy Henderson findings and recommendations we have private investors, and the international Oxfam International, developed as a group of Commissioners. development community. Through this Australia work programme the Commission has added The WCD is delivering its product in a a new body of knowledge to existing data- rapidly changing international environ- bases and information on large dams, looking ment. Debates proliferate about how to at alternate ways of meeting irrigation, water conserve the world’s precious resource base supply, energy, flood control requirements and while meeting the needs of growing popula- processes of development planning. tions hungry for economic progress. Terms of investment, terms of trade, democratisa- How did we achieve this? First by creating a Mr Göran Lindahl tion, the role of the state, the role of civil 68 member Stakeholder Forum to act as a ABB Ltd, Sweden society, the obligation to protect threatened sounding board and advisory group for the ecosystems and preserve Planet Earth for WCD. The Forum has allowed us to reach future generations: all are part of the wider other partners in the debate and to sound context. Any policy on large infrastructure out those whose support for this report is projects – whether for dams, highways, essential to its effective use. power stations, or other mega-installations – has to be developed in this context. Second, by drawing on the wider stakeholder community for experts and analysts in At the same time, alternative perspectives developing the WCD Knowledge Base, and Prof Thayer Scudder California Institute on human rights and development are being for funds to support the Commission’s work. of Technology, USA more clearly expressed. The Right to Devel- To date, 53 public, private and civil society

viii Dams and Development: A New Framework for Decision-Making Foreword

organisations have pledged funds to the We are fully aware that this body of data WCD. This independent funding model is cannot and should not be seen as the ‘final unique for international commissions. verdict’ on the large dams story. The WCD has examined around 1 000 dams with Third, by undertaking a programme of four varying degrees of intensity. This is a small Regional Consultations in different parts of fraction of the more than 45 000 large dams the world that drew submissions from an world-wide. There has been little systematic collection of data about dam projects in the even wider network of interested parties. Ms Joji Cariño This provided a platform for all voices to be past, and without baseline data we cannot Tebtebba Foundation, heard. Altogether, 1 400 individuals from 59 arrive at definite conclusions about certain Philippines countries and from every type of stakeholder types of impacts. Albeit, the WCD report is group took part in these regional consulta- the first comprehensive global and inde- tions. The WCD also participated in two pendent review of the essential aspects of hearings on large dams organised for its the performance of dams. In many cases the benefit by NGOs in Southern Africa and impacts are still being played out, and will Europe. During its two-year lifetime the continue for many years to come. For this Commission has received 947 submissions reason it is important for the future manage- from over 80 countries. We have listed all of ment of dams to have continued and sys- them in a central database accessible via the tematic evaluation of their performance. Mr Donald Blackmore Murray-Darling Basin Internet. Commission, Australia In carrying out our review we have not tried Fourth, by initiating eight independent in- to judge individual dams. We have found depth Case Studies of specific large dam that the unprecedented expansion in large projects and two country studies (India and dam building over the past century, harness- China). Using a common methodology, the ing water for irrigation, domestic and case studies were conducted in a transparent industrial consumption, electricity genera- and participatory manner drawing inputs tion and flood control has clearly benefited from all stakeholders through an extensive many people globally. Nonetheless, this consultation process. We have used the positive contribution of large dams to Ms Medha Patkar findings from all these studies and submis- development has been marred in many cases Struggle to Save the sions to inform a central product of the by significant environmental and social Narmada River, India Report: the Global Review of Large Dams. impacts which, when viewed from today’s values, are unacceptable. Fifth, by undertaking 17 Thematic Reviews, employing experts from a wide range of We have sought to glean the lessons from disciplinary backgrounds, nationalities and the past in order to make recommendations institutional bases. These reviews fall into for the future through a prism of equitable five thematic categories: Social Issues, and sustainable development. Development based on five objectives: equity in resource Environmental Issues, Economic and Prof José Goldemberg Financial Issues, Options Assessment and allocation and in the spread of benefits; University of São Institutional Issues. Once again, we con- sustainability in the use of the world’s Paulo, Brazil ducted these reviews in a participatory diminishing resource-base; openness and manner, commissioning over 100 papers participation in decision-making processes; with full peer review. efficiency in the management of existing infrastructural developments; and accountabil- Lastly, by conducting a comprehensive ity towards present and future generations. global survey of 125 dams, which we used to ‘cross-check’, the findings of the individual In today’s rapidly globalising world the studies. The Cross-Check Survey has WCD process has pioneered a new path for global public policy making on issues of Ms Deborah Moore provided a basic set of data on trends to Environmental complement the knowledge base. equitable and sustainable development. This Defense, USA

The Report of the World Commission on Dams ix Foreword

Report is the test against which that process The life of the World Commission on will be measured. We hope that the policy Dams ends with the publication of this framework and practical guidelines for its Report. For the Commissioners this has implementation presented here will add been an exciting, challenging and enrich- significant value to existing norms and form ing process. It would not have been a basis for best practice in water and energy possible without the tireless commitment development. This is only a beginning, but of an extraordinary team of professionals we hope it is a dynamic beginning that in the Secretariat. But more, we are Mr Jan Veltrop Honorary President, others can take forward in the future. We indebted to hundreds of people around the ICOLD, USA also hope that the lessons learnt from our world who, mostly at their own expense, analysis of large dams will be seen as rele- have given us the bounty of their knowl- vant for other large infrastructural projects, edge, expertise and life’s experiences and that the framework of policy develop- though discussion papers, submissions and ment and implementation we have identi- presentations adding light and life to this fied will see wider application. report on large dams.

Mr Achim Steiner WCD Secretary-General, Germany Professor Kader Asmal Mr Donald Blackmore

Mr Lakshmi Chand Jain Ms Medha Patkar

Dr Judy Henderson Prof José Goldemberg

Mr Göran Lindahl Ms Deborah Moore

Prof Thayer Scudder Mr Jan Veltrop

Ms Joji Cariño Mr Achim Steiner

x Dams and Development: A New Framework for Decision-Making Table of Contents

Table of Contents

Chair’s Preface i Commissioners’ Foreword vii Table of Contents xi List of Tables xiii List of Figures xiii List of Boxes xiv Acronyms and Abbreviations xvii Acknowledgements xix Executive Summary xxvii

Chapter 1: Water, Development and Large Dams 1 Water and Development 3 Development and Large Dams 8 Large Dams as Instruments of Development 11 Problems Associated with Large Dams 15 Understanding the Large Dams Debate 17 Fulfilling the WCD Mandate – Process and Methodology 28

PART I: THE WCD GLOBAL REVIEW OF LARGE DAMS 35

Chapter 2: Technical, Financial and Economic Performance 37 Structure and Methodology 38 Construction Costs and Schedules 39 Irrigation Dams 42 Hydropower Dams 49 Water Supply Dams 56 Flood Control Dams 58 Multi-Purpose Dams 62 Physical Sustainability Issues 63 Findings and Lessons 68

The Report of the World Commission on Dams xi Table of Contents

Chapter 3: Ecosystems and Large Dams: Environmental Performance 73 Terresterial Ecosystems and Biodiversity 75 Greenhouse Gas Emissions 75 Downstream Aquatic Ecosystems and Biodiversity 77 Floodplain Ecosystems 83 Fisheries 84 Ecosystem Enhancement 86 Cumulative Impacts 88 Anticipating and Responding to Ecosystem Impacts 89 Findings and Lessons 92

Chapter 4: People and Large Dams: Social Performance 97 Socio-Economic Impacts through the Project and Planning Cycle 99 Displacement of People and Livelihoods 102 Indigenous Peoples 110 Downstream Livelihoods 112 Gender 114 Cultural Heritage 116 Human Health 118 Equity and the Distribution of Costs and Benefits 120 Findings and Lessons 129

Chapter 5: Options for Water and Energy Resources 135 Development Agriculture and Irrigation 137 Energy and Electricity 148 Water Supply 156 Integrated Flood Management 160 Findings and Lessons 163

Chapter 6: Decision-Making, Planning and Institutions 167 Decision-Making and the Political Economy of Large Dams 169 Role of Foreign Assistance 171 Planning and Evaluation 175 Compliance 185 Findings and Lessons 190

PART II: THE WAY FORWARD 195

Chapter 7: Enhancing Human Development: Rights, Risks and Negotiated Outcomes 197 From Global Review to Future Practise 198 Sustainable Human Development – A Global Framework 199 Trends and Challenges in Applying the New Development Framework 203 Rights and Risks – an Improved Tool for Decision-Making 206 Negotiated Agreements on the Basis of Rights and Risks 208 Conclusion 210

xii Dams and Development: A New Framework for Decision-Making Table of Contents

Chapter 8: Strategic Priorities – A New Policy Framework for the Development of Water and Energy Resources 213 Gaining Public Acceptance 215 Comprehensive Options Assessment 221 Addressing Existing Dams 225 Sustaining Rivers and Livelihoods 234 Recognising Entitlements and Sharing Benefits 240 Ensuring Compliance 244 Sharing Rivers for Peace, Development and Security 251

Chapter 9: Criteria and Guidelines – Applying the Strategic Priorities 259 Five Key Decision Points: The WCD Criteria 262 A Special Case: Dams in the Pipeline 276 A Set of Guidelines for Good Practice 278

Chapter 10: Beyond the Commission- An Agenda for Change 309 Strategic Entry Points for Follow-up 311 Taking the Initiative – Institutional Responses 313 Continuing the Dialogue 316 A Call to Action 319

List of Tables 1.1 Dams currently under construction 10 1.2 Estimated annual investment in dams in the 1990s 11 1.3 Population density of selected river basins 17 4.1 Illustration of the services and benefits generated by large dams in the WCD Case Studies 121 4.2 Profile of groups adversely affected by large dams: illustrations from WCD Case Studies 124 5.1 Complementary approaches to flood management 161 6.1 WCD Case Studies: options assessment 178 9.1 Valuation methods 289

List of Figures 1.1 Annual fresh water withdrawals as a percentage of total resources withdrawn (1996) 6 1.2 Annual fresh water withdrawals per capita average (1987-95) 6 1.3 Distribution of the world’s water 7 1.4 Selected water-stressed countries 7 1.5 Regional distribution of large dams at the end of the 20th century 8 1.6 Construction of dams by decade (1900-2000) 9 1.7 Dams constructed over time by region (1900-2000) 9 1.8 Distribution of existing large dams by region and purpose 12 1.9 Agricultural land irrigated from dams 13 1.10 World map showing the regional location of the case studies, country studies, cross-check survey dams, regional consultations, submissions and Forum members 31

The Report of the World Commission on Dams xiii Table of Contents

2.1 Cost overruns on large dams 39 2.2 Average cost overruns for large dams 40 2.3 Project schedule performance 42 2.4 Achievement of command area 43 2.5 Actual irrigated area compared to planned targets over time 43 2.6 Economic performance of multilateral-financed irrigation dams 47 2.7 Project averages for actual versus hydropower generation 50 2.8 Actual versus planned hydropower generation over time 51 2.9 WCD case study hydropower performance: capacity and power generation 52 2.10 Multilateral bank evaluation results on the economic performance for hydropower dams 54 2.11 Project averages of actual versus planned bulk water supply delivery 57 2.12 Actual versus planned bulk water supply delivery over time 57 2.13 Trends in dam safety assessments 65 2.14 Loss of active storage due to sedimentation 65 2.15 Loss of active storage due to sedimentation by reach of river 65 2.16 Waterlogging and salinity 67

3.1 Gross greenhouse emissions from reservoirs 76 3.2 Greenhouse gas emissions from natural habitats 76 3.3 Modification of annual regimes due to a hydropower dam, Colorado River at Lee’s Ferry, United States 79 3.4 Fluctuations of daily streamflow regime due to hydropower peaking operations, Colorado River at Lee’s Ferry, September 79 3.5 Decline in species numbers but increase in fisheries productivity, Tucurui (a&b) 86 3.6 Fragmentation in 225 large river basins 87 3.7 Anticipated and unanticipated ecosystem impacts 89

5.1 Schematic of electricity options 150

6.1 Development assistance for large dams, 1950-1999 171 6.2 Trends in provisions for participation and information disclosure 176 6.3 Trends in the implementation of economic and financial analyses 186 6.4 Trends in the implementation of environmental and social assessments 187

7.1 The WCD policy framework 202 7.2 From rights and risks to negotiated agreements: a framework for options assessment and project planning 208

9.1 WCD Criteria and Guidelines strengthen other decision support instruments 260 9.2 Five key points in planning and project development 263 9.3 Preference matrix for ranking options 285

List of Boxes 1.1 New paradigm for water use 3 1.2 Types of large dams 11 1.3 Changing physical attributes and impacts of large dams 15 1.4 Central issues in the dams debate: past and present 21

2.1 Efficiency in the use of irrigation water 46 2.2 Economic and financial performance of the Columbia Basin Project 47 2.3 Cost recovery for the Aslantas dam 48

xiv Dams and Development: A New Framework for Decision-Making Table of Contents

2.4 Optimising operations with the aid of a computerised decision-support system 53 2.5 Financial and economic performance of hydropower at Grand Coulee dam 55 2.6 Economic performance and cost recovery of hydropower at Tucurui dam 56 2.7 Flood protection in Japan 59 2.8 From flood control to flood management in the United States 61 2.9 Cost recovery in a multi-purpose scheme: Grand Coulee and the Columbia Basin Project 62 2.10 Dam safety in the United States 64

3.1 Mitigating and compensating for terrestrial impacts 75 3.2 Greenhouse gas emissions at Tucurui, Brazil 77 3.3 How one dam has affected two different species in opposite ways 78 3.4 Minimising impacts of changes in streamflow regime: environmental flow requirements 81 3.5 Mitigation measure: fish passes 82 3.6 Restoring ecosystem function through managed floods 84 3.7 Cumulative impact of dams: the Aral Sea 88 3.8 Ecosystem restoration through decommissioning in the United States 92

4.1 Bringing electricity to the favelas in São Paulo, Brazil 101 4.2 Economic, socio-cultural and health impacts of livelihood displacement 103 4.3 Missing numbers of affected people: Sardar Sarovar project, India, and Pak Mun dam, Thailand 104 4.4 Economic value of downstream floodplains, Hadejia-Nguru, Nigeria 113 4.5 The Aswan High dam: a milestone in the history of archaeology 117 4.6 Mercury and human health at Tucurui 119 4.7 Royalties to communities: a Brazilian law for hydropower benefit-sharing 127

5.1 Conjunctive management of salinity 139 5.2 Cultivation techniques can reduce irrigation water use 141 5.3 A local approach to integrated water management, Rajasthan, India 144 5.4 Rainwater harvesting for domestic and agricultural use in China 144 5.5 Wetland and flood plain agriculture 145 5.6 Rainwater harvesting is spreading to urban areas 158 5.7 Flood resilience 162

6.1 WCD Case Studies: political decisions to build large dams 170 6.2 WCD Case Studies and Submissions: foreign involvement in dam projects 173 6.3 Nordic influence in the Pangani Falls Redevelopment Project, Tanzania 174 6.4 Co-operation in shared river basins 174 6.5 Even late participation leads to a consensus resettlement plan: Salto Caixas dam, Brazil 177 6.6 Public participation and project acceptance: three scenarios from Austria 177 6.7 Environmental Impact Analysis (EIA): too little, too late 183 6.8 Licensing processes and duration 185 6.9 Allegations of corruption 187 6.10 Export Credit Agencies: competing for business versus common standards 189 6.11 WCD Case Studies: a compliance report card 190

7.1 Shared values and institutional practices – the UN Millenium Report 199 7.2 Human rights and human development 203 7.3 Voluntary risk takers and involuntary risk bearers 207 7.4 Good governance and the UN Millenium Report 209

The Report of the World Commission on Dams xv Table of Contents

9.1 Health impact assessment 284 9.2 Cultural heritage impact assessment 285 9.3 Ghazi-Barotha, Pakistan 291 9.4 Design and cost of environmental flows - Pollan dam, Ireland 295 9.5 Benefits of improving fish passes 296 9.6 Financial assurances and the Environment Protection Agency, Victoria, Australia 304 9.7 Suriname Central Nature Reserve 305 9.8 Mendoza Province, Argentina 305

10.1 Priorities for strengthening the knowledge base 317

A Comment – Medha Patkar 321

ANNEXES 323 I Bibliography 323 II Glossary 344 III WCD Work Programme – Approach and Methodology 349 IV Reports in the WCD Knowledge Base 359 V Dams, Water and Energy – A Statistical Profile 368 VI United Nations Declarations 383 VII Members of the World Commission on Dams 394 VIII A Profile of the WCD Secretariat 397

INDEX 399

xvi Dams and Development: A New Framework for Decision-Making Acknowledgements

Acknowledgements

The Forum Forum members can help to build ownership of Commission work. The WCD is also Coming from 68 institutions in 36 coun- conscious that reports in themselves have tries, members of the Forum reflect the little impact if they are not firmly rooted in diverse range of interests in the dams a process that enables all interest groups to debate. The Forum acts as a ‘sounding develop an understanding of and confidence board’ for the work of the Commission, and in the process itself. Membership of the helps maintain two-way communication Forum however does not imply endorsement with the various far-flung constituencies. of the Commission’s report and findings.

Affected People’s Groups LHWP, Lesotho Highlands Water Project, CODESEN, Co-ordination for the Senegal Lesotho River Basin, Senegal Ministry of Water Resources, China COICA, Federación de Indigenas del Estado National Water Commission, Mexico Bolívar, Venezuela Ministry of Mahaweli Development, Sri Lanka Grand Council of the Cree, Canada Ministry of Water Resources, India MAB, Movimento dos Atingidos por Barragens, Brazil International Associations NBA, Narmada Bachao Andolan, India ICID, International Commission for Irrigation Sungi Development Foundation, Pakistan and Drainage Cordillera People’s Alliance, Philippines ICOLD, International Commission on Large Dams Bilateral Agencies/Export Credit IHA, International Hydropower Association Guarantee Agencies IAIA, International Association for Impact BMZ, Federal Ministry for Economic Co- Assessments operation and Development, Germany NORAD, Norwegian Agency for Multilateral Agencies International Co-operation, Norway ADB, Asian Development Bank JBIC, Japan Bank for International Co- AfDB, African Development Bank operation, Japan FAO, United Nations Food and Agriculture SDC, Swiss Agency for Development and Co- Organisation operation, Switzerland IADB, Inter-American Development Bank Sida, Swedish International Development UNDP, United Nations Development Agency, Sweden Programme U.S. Export/Import Bank, USA UNEP, United Nations Environment Government Agencies Programme United States Bureau of Reclamation, USA WB, World Bank

The Report of the World Commission on Dams xix Acknowledgements

Non-Governmental Organisations Research Institutes/Resource Persons Berne Declaration, Switzerland Centro EULA, Ciudad Universitaria, Chile ENDA, Environmental Development Action, Tropica Environmental Consultants Ltd., Senegal Senegal Help the Volga River, Russia WRI, World Resources Institute, USA IRN, International Rivers Network, USA Water Research Institute, Israel ITDG, Intermediate Technology Development Winrock International, Nepal Group, United Kingdom Focus on the Global South, Thailand IUCN, The World Conservation Union, ISPH, Institute of Hydroelectric Studies and Switzerland Design, Romania Sobrevivencia-Friends of the Earth, Paraguay IWMI, International Water Management WWF, World Wide Fund for Nature, Institute, Sri Lanka Switzerland Worldwatch Institute, USA DAWN, Development Alternatives with Wuppertal Institute, Germany Women for a New Era, Fiji TI, Transparency International, Germany River Basin Authorities WEED, World Ecology, Environment and Confederación Hydrográfica del Ebro, Spain Development, Germany Mekong River Commission, Cambodia Swedish Society for Nature Conservation, Volta River Authority, Ghana Sweden Jordan Valley Authority, Jordan Wetlands International, Japan Utilities Private Sector Firms Eletrobras, Brazil Enron, USA Hydro-Québec, Canada Harza Engineering, USA Nepal Electricity Authority, Nepal Siemens, Germany Mini Hydro Division, Philippines ABB, Switzerland Electricité de France, France Saman Engineering Consultants, South Korea Engevix, Brazil

Partnerships and Co-operation

The Food and Transparency The United The Environmental International Agriculture International Nations Monitoring Group Commission on Organisation Environment Large Dams Program

International The World The International The World Health International Institute Association Archaeological Energy Agency Organisation for Sustainable for Impact Congress Development Assessment

The World Bank

xx Dams and Development: A New Framework for Decision-Making Acknowledgements

Financial contributors various foundations. According to the mandate of the Commission, all funding Financial support was received from 53 received by it was ‘untied’ – i.e. these funds contributors including governments, interna- were provided with no conditions attached tional agencies, the private sector, NGOs and to them.

People’s Republic of China – Ministry of Water Resources

Finland – Ministry of Foreign Affairs

Germany

United Nations Environment Programme Ireland – Ministry of Inter-American Foreign Affairs Develoment Bank

South Africa – Ministry of Water Affairs and Forestry The World Bank

UK Department for International Development

Japan – Ministry of Foreign Affairs Swiss Agency for Development and Cooperation Swedish International Development Agency

Denmark – Ministry of Foreign Affairs

Netherlands – Ministry of Foreign Affairs Goldman Environmental Foundation

C.S. Mott Norway – Ministry of Asian Development Foundation Foreign Affairs Bank

The Report of the World Commission on Dams xxi Acknowledgements

WCD Case Studies – Team Leaders, Team of Authors, Additional Authors and General Acknowledgements

Grand Coulee Dam, Columbia an, Ronald Lwamba, Hillary Masundire, River Basin, USA Dominic Mazvimavi, Ngonidzaishe Moyo, TL: Leonard Ortolano, Katherine Kao Cushing Herbert Mudenda, Fanuel Nangati, Daniel AA: Nicole T. Carter, Harza Engineering, Ndlela, Elias Nyakunu, Norman Reynolds, William Green, Carl Gotsch, Kris May, Tim John Santa Clara, Bennet Siamwiza, Steven Newton, Sophie Pierre, Josh Smienk, Michael Tembo GA: David Z. Mazvidza, Chris Soules, Marilyn Watkins GA: BC Hydro, Magadza, Steve Rothert Bonneville Power Administration, Bureau of Reclamation, Canadian First Nations, Columbia Tucurui Dam, Tocantins River, Basin Trust, Colville Trust, Spokane Tribe, Brazil United States Army Corps of Engineers TL: Emilio Lèbre La Rovere, Francisco Eduardo Mendes AA: Bertha Becker, Gilber- Tarbela Dam, Indus River Basin, to Canali, Rosa Carmina Couto, Paulo Diniz, Pakistan Iara Ferraz, Efrem Ferreira, José Alexandre Fortes, Maria das Graças da Silva, Marcia TL: Amir Muhammed Khan AA: Altaf Abro, Ismerio, Ana Lacorte, Renato Leme Lopes, Shahid Ahmed, Pervaiz Amir, Afzal Haq, Adriana Neves Luna, Sandra Macedo, Rosa Mehmooda S. Jilani, Riaz Ahmed Khan, Peter Acevedo Marin, Oscar de Moraes Cordeiro John Meynell, Javed Saleem Qamar, Riaz Netto, Sylvia Helena Padilha, Lúcio Flávio Hussain Qureshi, Riaz Nazir Tarar GA: Ministry Pinto, Eneas Salati, Maria Nazareth da Silva, of Water and Power, Water and Power Develop- Wanderli Pedro Tadei GA: Electronorte, ment Authority, Members of the Pakistan Agéncia Nacional de Energia Eléctrica, Network for Rivers, Dams and People, IUCN- Centro Agroecologico de Assessoria y Pakistan Office, World Bank-Pakistan Office Educación Popular, Osmar Vieira Filho, Aslantas Dam, Ceyhan River Marcos V. Freitas, Raimundo Nonato do C. Silva, Sadi Baron, Henri Acselrad, Philip Basin, Turkey Fearnside, Birgit Zimmerle, Jean Remy D. TL: Refik Çölasan AA: O. Türker Altan, Guimaraes Okan Arihan, Çigdem Baykal, Ali Çaglar, Ahmet Eltekin, Nadir Izgin, Riza Kanber, Pak Mun Dam, Mun-Mekong Suhandan Karauz, Haluk Kasnakoglu, Birsen River Basin, Thailand Gökçe, Zuhal Güler, Mete Kaan Kaynar, S¸ uha TA: Sakchai Amornsakchai, Philippe Annez, S¸atana, Bora Sürmeli, Zeliha Ünaldi, Erdal Sansanee Choowaew , Songkram Grachang- S¸ekeroglu, Tuluhan Yilmaz, Recep Yurtal GA: netara, Prasit Kunurat, Jaruwan Nippanon, Department of State Hydraulic Works (Anka- Roel Schouten, Pradit Sripapatrprasite, ra, Adana and Karamamaras), General Chayan Vaddhanaphuti, Chavalit Vidthay- Directorate of Rural Services, Water Users anon, Suphat Vongvisessomjai, Ek Watana, Associations of Ceyhan Aslantas Project, Dogal Wanpen Wirojanagud GA: Ammar Siamwalla, Hayati Koruma Dernegi, Ministry of Environ- Chainarong Sretthachau, Darayes Mehta, ment and Provincial Directorate of Health, Kitcha Polparsi, M. L. Chanaphun Kridakorn, Directorate of Agriculture, Forestry and Nation- Michai Veravaidhya, Prudhisan Jumbala, al Parks, World Bank -Turkey Office Sansern Wongcha-um, Vatana Meevasana, Kariba Dam, Zambezi River, Vipada Apinan, Zakir Hussain Zambia/Zimbabwe Glomma and Laagen Basin, TL: Alois Hungwe AA: Julius Chileshe, Norway Moses Chimbari, Dennis Chiwele, Paulman TL: Jostein Skurdal AA: Øystein Aas, Tor Chungu, Andrew Conybeare, Ezekiel Jonath- Arnesen, Per Christian Bøe, Åge Brabrand,

xxii Dams and Development: A New Framework for Decision-Making Acknowledgements

Jon Arne Eie, Bjørn P. Kaltenborn, Svein Erik Country Study: India Hagen, Karine Hertzberg, Trygve TA: Pranab Banerji, Ramaswamy R. Iyer, R. Hesthagen, Arne Linløkken, Dan Lundquist, Rangachari, Nirmal Sengupta, Shekhar Singh Pål Mellquist, Asbjørn Molle, Torbjørn GA: Som Pal, Z. Hasan, Raj Rajagopal, A. D. Østdahl, Trond Taugbøl, Jens Kristian Mohile, A. S. Desai, B. G. Verghese, K. R. Tingvold GA: Arne Erlandsen Øyvind Datye, M. C. Gupta, M. Gopalakrishnan, N. Fjeldseth, Geir Y. Hermansen, Thrond Berge C. Saxena, Pradeep K. Deb, Sunderlal Bahu- Larsen, Kurt Ole Linn, Pål Mellquist guna, P. S. Raghavan Orange River Pilot Study Country Study: China TL: WCD Secretariat AA: Azghar Adelzadeh, Andrew Ainslie, Geoff Antrobus, TA: John Boyle, Richard Fuggle, Habib Nicola Bergh, Bryan Davies, Chris de Wet, Khoury, Ismail Najjar, Sam Pillai, Bill Smith Tony Emmett, Muzi Muziya, Kyra Naudascher- Jankowski, Barry Nkomo, Maartin van Veelen Briefing Paper on Russia and GA: Thinus Basson, Fannie du Plessis, NIS countries Mirriam Kibi, Mike Meuller, Geraldine TA: Elena A. Barabanova, Nikolai I. Ko- Schoeman, Staff of Surplus Peoples’ Project ronkevich, Law and Environment Eurasia in Cape Town, Peter van Niekerk, Theo van Partnership (Central Asia), Lilia K. Malik, Robbroek Vladimir Smakhtin, Irina S. Zaitseva

Thematic Reviews - Lead Writers and Contributing Writers

I.1 Social impact of large dams: equity and Michel Larinier; Jeremy Lazenby; Gerd distributional issues: LW: William Marmulla; Don McAllister; Mathew Adams CW: Adrian Adams; Hugh McCartney; Steve Miranda; John Brody; Dominique Egre; Carmen Ferra- Morton; Dianne Murray; Mary Seddon; das; Pablo Gutman; Lyla Mehta; Joseph Leonard Sklar; David Smith; Caroline Milewski; Bina Srinivasan; Lubiao Zhang Sullivan; Rebecca Tharme I.2 Dams, indigenous people and vulnera- II.2 Dams and global change: ble ethnic minorities: LW: Marcus LW: Nigel Arnell; Mike Hulme; Luiz Colchester CW: Jaqueline Carino; Pinguelli Rosa; Marco Aurelio dos Jaroslave Colajacomo; Andrew Corbett; Santos CW: Albert Mumma Gabungan; Luke Hertlein; Manisha III.1 Economic, financial and distributional Marwaha; Lyla Mehta; Amrita Patward- analysis: LW: Alec Penman; Michelle han; Ande Somby; Maria Stankovitch Manion; Bruce McKenney; Robert I.3 Displacement, resettlement, rehabilita- Unsworth CW: Colin Green; Pablo tion, reparation and development: Gutman; Anneli Lagman; Anil Mar- LW: Leopoldo Bartolome; Chris de Wet; kandya; David Mullins; Kyra Nau- Harsh Mander; Vijay Nagaraj CW: dascher-Jankowski; Douglas Southgate Christine Danklmaier; Ravi Hemadri; III.2 International trends in project financing: Jun Jing; Scott Robinson LW: Per Ljung CW: Lily Donge; Chris II.1 Dams, ecosystem functions and envi- Head; Michael Kelly; Hilary Sunman ronmental restoration: LW: Mike IV.1 Electricity supply and demand manage- Acreman; Ger Bergkamp; Patrick Dugan; ment options: LW: Maritess Cabrera; Jeff McNeely CW: Asheline Appleton; Anton Eberhard; Michael Lazarus; Edward Barbier; Garry Bernacsek; Thierry Lefevre; Donal O’Leary; Chella Martin Birley; John Bizer; Cate Brown; Rajan CW: Glynn Morris; Roger Peters; Kenneth Campbell; John Craig; Nick Bjorn Svenson; Rona Wilkinson Davidson; Simon Delany; Charles IV.2 Irrigation options: LW: K. Sanmuga- DiLeva; Frank Farquharson; Nicholas nathan CW: Pablo Anguita Salas; Hodgson; Donald Jackson; Jackie King; S. Char; Keith Frausto; Alfred Heuper-

The Report of the World Commission on Dams xxiii Acknowledgements

man; Khalid Hussain; ICRISAT; Hector V.2 Environmental and social assessment Maletta; Dieter Prinz; Yehuda Shevah; for large dams: LW: Barry Sadler Anupam Singh; Laurence Smith; CW: Frank Vanclay; Iara Verocai Himanshu Thakkar V.3 River basins - institutional frameworks IV.3 Water supply options: LW: Colin Fenn; and management options: LW: Peter David Sutherland CW: Mary Dickinson; Millington CW: Len Abrams; Enrique John Gould; Allan Lambert; Jon Lane; Castelan Crespo; Fiona Curtin; Luis Guy Preston; Philip Turner Garcia; Raymundo Garrido; Ramaswamy IV.4 Flood control and management options: Iyer; Erik Mostert; Cecilia Tortajada; LW: Colin Green CW: Luis Berga; Anthony Turton; Aaron Wolf Patrick Hawker; Minoru Kuriki; Dennis V.4 Regulation, compliance and implemen- Parker; Sylvia Tunstall; Johannes van tation: LW: Angela Cropper; Mark Duivendijk; Herb Wiebe Halle; John Scanlon CW: Daniel IV.5 Operation, monitoring and decommis- Bradlow; Gabriel Eckstein; Balakrishnan sioning of dams: LW: Peder Hjorth; Rajagopal; Tom Rotherham; Lori Udall; Charles Howard; Kuniyoshi Takeuchi Michael Wiehen CW: K. Betts; Michael Falter; Enrique V.5 Participation, negotiation, and conflict Garcia; Peter Goodwin; Brian Haisman; management: LW: Bruce Stedman Joji Harada; V. Jauhari; Thomas Russo; CW: Tisha Greyling; Anne Randmer; Geoffrey Simms; James Westcoat; Vanchai Vatanasapt; Arch Isabel Viana Rodney White V.1 Planning approaches: LW: David Working Papers – Writers Nichols; Theo Stewart; David von Dams and Human Health: Martin Birley, Hippel CW: Daud Beg; Catherine Robert Bos, M’barack Diop, William Jobin, Fedorsky; Matthias Finger; P Unnikrishnan J. Karmacharya; Miguel Nucete; Don Dams and Cultural Heritage Management: Moore; Girish Sant Steven Brandt, Fekri Hassan

Contributors to the WCD Cross-Check Survey

Keizrul Abdullah, M.A. Abrougui, Rocha William Loker, Eleyterio Luz, Anil Mar- Afonso, K. Akapelwa, Antonio Altadill, S.C kandya, Isidro Lázaro Martín, Aboubacry Anand, Mike Anane, Valdemar Andrade, Carlo Mbodji, Patrick McCully, Brad McLane, Angelucci, Cindy Armstrong, Alexander Roland Mejias, Chantho Milattanapheng, Asarin, Leo Atakpu, Paul Aylward, Riad Baouab, Joseph Milewski, José Díaz Mora, Amir GR Basson, Mona Bechai, Edigson Perez Muhammed, Jorge Carreola Nava, Eden Bedoya, Hans Helmut Bernhart, Carlos Bertag- Napitupulu, Humphrey Ole Ncharo, Luz no, Isaac Bondet, Peter Bosshard, Rodney Nereida, Nguyen Anh Minh, Ali Noorzad, Bridle, Adelino Estevo Bucuane, Brian Davies, Magdelena Nunez de Cordero, David Okali, Cheickna Seydi A. Diawara, Shripad Dhama- Elizabeth Olsen, Torbjørn Østdahl, F.C dhikary, Foto Dhima, Bob Douthwaite, Tim Oweyegha-Afunaduula, Thomas Panella, Dunne, Gary Ellis, Richmond Evans-Appiah, Richard L. Pflueger, Hermien Pieterse, Julio Daryl Fields, Renzo Franzin, Alejandro Garcia, Pineda, Jose Polanco, Ambriosio Ramos, Alfonso García, Luis Garcia, Refik Ghalleb, Silvia Maria Ramos, Bernard Reverchon, Ben Marcus Gillespie, Handan Giray, M. Johan Rossouw, Paul Royet, Joseph Rückl, Gopalakrishnan, Liane Greef, Francis Grey, David Scivier, Jose Roberto Serrano, Thomas Jose Ramon Guifarro, Ronald Valverde Siepelmeyer, Kua Kia Soong, Bjorn Svenson, Guillen, Brian Haisman, Phil Hirsch, Ku Hsu, Gustavo Tamayo, Himanshu Thakkar, Suresh Dan Hulea, Alois Hungwe, K.L. Karmachar- Kumar Thapa, Dao Trong Tu, Jan Tosnar, ya, Shaheen Rafi Khan, David Kleiner, Elena Jaroslav Ungerman, Martin van Veelen, H.A. Kolpakova, Michael Kube, Minoru Kuriki, Wickramaratna, Brayton Willis, Ralph Witte- Hee Seung Lee, Knut Leitner, Melissa Loei, bolle, Chusak Wittayapak, Patricia Wouters

xxiv Dams and Development: A New Framework for Decision-Making Acknowledgements

Presenters at the WCD’s Regional Consultations

South Asia Baba, H. El Badraoui, Mohammed N. Bayoumi, Kamau Bobotti, Geoffrey Chavula, Marwa K B Chand, Giasuddin Ahmed Choudhury, K R Daoudy, Seydi Ahmed Diawara, Salif Diop, Datye, Shripadh Dharmadhikary, Aly Ercelawn, Stephanie Duvail, R. Evans-Appiah, Arif Drona Ghimire, Ramaswamy Iyer, S Karu- Gamal, Alioune Gassama, Munther J. Hadda- naratne, A R Karunawathie, Shaheen Rafi din, Olivier Hamerlynk, Mohammed Jellali, Khan, Tauhidul Anwar Khan, Ashish Kothari, E.A.K. Kalitsi, Elisabeth Khaka, Jacqueline Ki- Lakshman Mediwake, Alistair McKechnie, Zerbo, O.M. Letsela, Claire Limbwambwa, Bazak D K Mishra, Iswer Raj Onta, M G Padhye, Zakeyo Lungu, Bhekani Maphalala, Reatile Bikash Pandey, Tilak Ranaviraja, M S Reddy, Mochebelele, Anna Moepi, Frank Muramuzi, Saleem Samad, Girish Sant, S Selvarajah, P C Martin Musumba, Mohammed Nabil, Takehiro Senaratne, Sardar Muhammad Tariq, Himanshu Nakamura, Karim S Numayr, Hassan M. A. Thakkar, B G Verghese, D C Wijeratna, Osman, F.C. Oweyegha-Afunaduula, Levin Hemantha Withenage Özgen, Guy Preston, Karen Ross, Mahammadou Latin America Sacko, Motseao Senyane, Noxolo Olive Sephu- ma, A. Shalaby , David Smith, Robyn Stein, Aziz Ab’ Saber, Carlos Avogadro, Ismael Aguilar David Syantami Syankusule Barajas, Sadi Baron, Celio Bermann, Ricardo Canese, Jorge Cappato, José Porfirio Fontanelle de Carvalho, Margarita Rosa de Castro Illera, East and South East Asia Carlos Chen, Ivan Correa, Jorge Oscar Daneri, Keizrul Abdullah, Reiko Amano, Le Quy An, Luis Alberto Machado Fortunato, Cassio Viotti, Vu Hong Anh, Vipada Apinan, Michael Afonso Henriques Moreira Santos, Marcos Bristol, Le Thac Can, Harvey Demaine, Hans Aurelio de Freitas, Philip Fearnside, Fabio Freiderich, Arnaldo Tapao Gapuz, Shalmali Feldmann, Altino Ventura Filho, Hector Guttal, Nguyen Dinh Hoa, Pham Thi Mong Huertas, M. Kudlavicz, Jaime Millan, José Hoa, Susannah Hopkins Leisher, Hiroshi Rodrigues, Humberto Marengo, Euclides Pereira Hori, Michael Horowitz, Tran Minh Huan, Macuxi, Miguel Nucete, Cristian Opaso, Tomoo Inoue, Sung Kim, Joern Kristensen, Bonarge Pacheco, Elias Diaz Pena, Alan Poole, See-Jae Lee, Nguyen Duc Lien, Arthur H. Grethel Aguilar Rojas, Teodoro Sanchez, Mitchell, Masaru Nishida, Alastair M. North, Salomon Nahmad Sitton, Carlos Vainer Do Hong Phan, Grainne Ryder, Shyama Shepard, Kua Kia Soong, Chainarong Africa and the Middle East Sretthachau, Dao Trong Tu, A Rusfandi Adrian Adams, Tareq A. Ahmed, Cansen Usman, Mikhail Wakil, Wayne C. White, Akkaya, Leo Atakpu, Mohamed Lemine Ould Lubiao Zhang

The Report of the World Commission on Dams xxv Table of Contents

Acronyms and Abbreviations

ADB Asian Development Bank ILO International Labour AfDB African Development Bank Organization CBA cost-benefit analysis IRN International Rivers Network CHIA cultural heritage impact assessment IRP independent review panels CHP combined heat and power ISO International Organization for DRD Declaration on the Right to Standardization Development IUCN World Conservation Union DSM demand-side management LCA life cycle assessment DSS decision support system MCA multi-criteria analysis EA environmental assessment MIGA multilateral investment guarantee ECA export credit agency agency EFA environmental flow assessment MRDAP mitigation, resettlement and EFR environmental flow requirement development action plan EIA environmental impact assessment MW megawatts EIRR economic internal rate of return NGO non-governmental organisation EPA Environmental Protection Agency NORAD Norwegian Agency for FIRR financial internal rate of return International Co-operation FSC Forestry Stewardship Council O&M operation and maintenance GATT General Agreement on Tariffs and OECD Organisation for Economic Trade Co-operation and Development GDP gross domestic product OED Operations Evaluation Division GHG greenhouse gas/es (of the World Bank) HIA health impact assessment PCB polychlorinated biphenyls HLC high level committee IA impact assessment PIC prior and informed consent IADB Inter-American Development PV photovoltaic Bank R&D research and development ICID International Commission on SA strategic impact assessment Irrigation and Drainage SIA social impact assessment ICJ International Court of Justice SIDA Swedish International ICOLD International Commission on Development Agency Large Dams SRI socially responsible investing IEA International Energy Agency UNDP United Nations Development IFC International Finance Programme Corporation USAID US Agency for International IFI international finance institutions Development IHA International Hydropower WCD World Commission on Dams Association WRI World Resources Institute

The Report of the World Commission on Dams xvii Executive Summary

Executive Summary

he global debate about large Tdams is at once overwhelmingly complex and fundamentally simple. It is complex because the issues are not confined to

the design, construction and operation of dams themselves but embrace the range of social, environmental and

political choices on which the human aspiration to development and improved well-being depend. Dams

fundamentally alter rivers and the use of a natural resource, frequently entailing a reallocation of benefits from

local riparian users to new groups of beneficiaries at a regional or national

The Report of the World Commission on Dams xxvii Executive Summary

level. At the heart of the dams debate are ■ In too many cases an unacceptable and issues of equity, governance, justice and often unnecessary price has been paid to power – issues that underlie the many secure those benefits, especially in social intractable problems faced by humanity. and environmental terms, by people displaced, by communities downstream, by The dams debate is simple because behind taxpayers and by the natural environment. the array of facts and figures, of economic ■ Lack of equity in the distribution of statistics and engineering calculations, lie a benefits has called into question the number of basic and easily understood value of many dams in meeting water principles. If adhered to and routinely and energy development needs when applied, these principles would not only go a compared with the alternatives. long way towards responding to the contro- ■ By bringing to the table all those whose versy surrounding dams, but would markedly rights are involved and who bear the improve decision-making on water and risks associated with different options for energy resources, achieving better outcomes. water and energy resources development, In identifying these principles, the World the conditions for a positive resolution of Commission on Dams (WCD) has not had competing interests and conflicts are to look far; they are the same principles that created. emerge from the global commitments to ■ Negotiating outcomes will greatly human rights, development and sustainability. improve the development effectiveness of water and energy projects by eliminat- Our report tells a multifaceted story. But we ing unfavourable projects at an early draw from it some straightforward and stage, and by offering as a choice only practical advice to guide future decisions on those options that key stakeholders agree water and energy resources development. represent the best ones to meet the needs The report sets out to distil more than two in question. years of intense study, dialogue and reflection by the Commission, the WCD Secre- The direction we must take is clear. It is to tariat, the WCD Stakeholders’ Forum and break through the traditional boundaries of literally hundreds of individual experts and thinking and look at these issues from a affected people on every aspect of the dams different perspective. Our recommendations debate. It contains all the significant find- develop a rationale and framework that ings that result from this work and expresses responds to this critical need and offers everything that the Commission believes is scope for progress that no single perspective important to communicate to governments, can offer on its own. It will ensure that the private sector, civil society actors and decision-making on water and energy affected peoples – in short, to the entire development: spectrum of participants in the dams debate. ■ reflects a comprehensive approach to The evidence we present is compelling. We integrating social, environmental and feel confident that the material collected economic dimensions of development; and analysed by the Commission provides ■ creates greater levels of transparency and overwhelming support for the main messag- certainty for all involved; and es in the report. ■ increases levels of confidence in the ability of nations and communities to meet their We believe there can no longer be any future water and energy needs. justifiable doubt about the following: ■ Dams have made an important and There are no shortcuts to equitable and significant contribution to human sustainable development. The evidence of development, and the benefits derived success and failure we present in this report from them have been considerable. provides the best rationale why the ‘business

xxviii Dams and Development: A New Framework for Decision-Making Executive Summary

as usual’ scenario is neither a feasible nor a ing water quality and increasingly severe desirable option. limits to surface water extraction has brought sustainable water resources management to the top of the global development Water and Development – agenda. Although increasing competition The Changing Context for water suggests an expanding scope for conflict, it also provides an incentive for The key decisions are not about dams as new forms of co-operation and innovation. such, but about options for water and energy Dire scenarios for water demand must not development. They relate directly to one of overshadow the fact that development paths the greatest challenges facing the world in that meet and manage the demand exist and this new century – the need to rethink the are available for us to choose. History management of freshwater resources. A demonstrates that the path of co-operation number of global initiatives and reports has more often been followed than the path have documented the dramatic impact of of conflict. The same must be true for our human-induced water withdrawals from the future. world’s lakes, rivers and ground aquifers. Total annual freshwater withdrawals today During the 20th century, large dams emerged are estimated at 3 800 cubic kilometres – as one of the most significant and visible twice as much as just 50 years ago. tools for the management of water resources. The more than 45 000 large dams around The unfolding scenario for water use in the world have played an important role in many parts of the world is one of increasing helping communities and economies harness concern about access, equity and the re- water resources for food production, energy sponse to growing needs. This affects generation, flood control and domestic use. relations: Current estimates suggest that some 30– ■ within and between nations; 40% of irrigated land worldwide now relies on dams and that dams generate 19% of ■ between rural and urban populations; world electricity. ■ between upstream and downstream interests; From the 1930s to the 1970s, the construc- ■ between agricultural, industrial and tion of large dams became – in the eyes of domestic sectors; and many –synonymous with development and ■ between human needs and the require- economic progress. Viewed as symbols of ments of a healthy environment. modernisation and humanity’s ability to harness nature, dam construction accelerat- The challenge is not to mobilise so as to ed dramatically. This trend peaked in the compete successfully, but to co-operate in 1970s, when on average two or three large reconciling competing needs. It is to find dams were commissioned each day some- ways of sharing water resources equitably where in the world. and sustainably – ways that meet the needs of all people as well as those of the environ- While the immediate benefits were widely ment and economic development. These believed sufficient to justify the enormous needs are all intertwined, and our challenge investments made – total investment in is to resolve competing interests collectively. large dams worldwide is estimated at more Achieving equitable and sustainable solu- than $2 trillion – secondary and tertiary tions will be to the ultimate benefit of all. benefits were also often cited. These included food security considerations, local The imperative to supply growing popula- employment and skills development, rural tions and economies with water in a context electrification and the expansion of physical of depleting groundwater resources, declin- and social infrastructure such as roads and

The Report of the World Commission on Dams xxix Executive Summary

schools. The benefits were regarded as self- appropriate, for the planning, design, evident, while the construction and opera- appraisal, construction, operation, moni- tional costs tended to be limited to econom- toring and decommissioning of dams. ic and financial considerations that justified dams as a highly competitive option. The decision to proceed with a large dam, the way the decision was made, the opinions As experience accumulated and better and perspectives that were heard are at the information on the performance and conse- heart of the current debate about dams. This quences of dams became available, the full same question of choice – of decision- cost of large dams began to emerge as a making – also lay at the heart of the Com- serious public concern. Driven by informa- mission’s work. Our report is about improv- tion on the impacts of dams on people, river ing the way such decisions are made. basins and ecosystems, as well as their economic performance, opposition began to grow. Debate and controversy initially The WCD Global Review of focused on specific dams and their local Large Dams impacts. Gradually these locally driven conflicts evolved into a global debate about A large part of the Commission’s work the costs and benefits of dams. Global involved a broad and independent review of estimates of the magnitude of impacts the experience with large dams. The result- include some 40-80 million people displaced ing WCD Knowledge Base includes eight by dams while 60% of the world’s rivers detailed case studies of large dams, country have been affected by dams and diversions. reviews for India and China, a briefing paper The nature and magnitude of the impacts of for Russia and the Newly Independent dams on affected communities and on the States, a Cross-Check Survey of 125 exist- environment have now become established ing dams, 17 Thematic Review papers, as as key issues in the debate. well as the results of public consultations and more than 900 submissions made available to the Commission. This provided The World Commission on Dams was born the basis for the assessment of the technical, from this debate. Established in February financial, economic, environmental and 1998 through an unprecedented process of social performance of large dams, and the dialogue and negotiation involving repre- review of their alternatives. The review sentatives of the public, private and civil underlined the critical issues relating to society sectors, it began work in May of that governance and compliance that have come year under the Chairmanship of Professor to be associated with large dams. Kader Asmal, then South Africa’s Minister of Water Affairs and Forestry and later the Minister of Education. The Commission’s 12 The evaluation was based on the targets set members were chosen to reflect regional for large dams by their proponents – the diversity, expertise and stakeholder perspec- criteria that provided the basis for govern- tives. The WCD was created as an inde- ment approval. In reviewing this experience pendent body, with each member serving in the Commission has studied a broad spec- an individual capacity and none represent- trum of dams. Its analysis gave particular ing an institution or a country. attention to understanding the reasons why, how and where dams did not achieve their intended outcome or indeed produced The Commission’s two objectives were: unanticipated outcomes that explain the ■ to review the development effectiveness of issues underlying the dams debate. Present- large dams and assess alternatives for water ing this analysis does not overlook the resources and energy development; and substantial benefits derived from dams but ■ to develop internationally acceptable rather raises the question of why some dams criteria, guidelines and standards, where achieve their goals while others fail.

xxx Dams and Development: A New Framework for Decision-Making Executive Summary

Performance of large dams impacts, the poor quality and uncertainty of predictions, the difficulty of coping The knowledge base indicates that shortfalls with all impacts, and the only partial in technical, financial and economic implementation and success of mitiga- performance have occurred and are com- tion measures. pounded by significant social and environ- ■ Pervasive and systematic failure to assess mental impacts, the costs of which are often the range of potential negative impacts disproportionately borne by poor people, and implement adequate mitigation, indigenous peoples and other vulnerable resettlement and development pro- groups. Given the large capital investment grammes for the displaced, and the in large dams, the Commission was dis- failure to account for the consequences turbed to find that substantive evaluations of large dams for downstream livelihoods of completed projects are few in number, have led to the impoverishment and narrow in scope, poorly integrated across suffering of millions, giving rise to impact categories and scales, and inade- growing opposition to dams by affected quately linked to decisions on operations. communities worldwide. In assessing the large dams reviewed by the ■ Since the environmental and social costs Commission we found that: of large dams have been poorly accounted for in economic terms, the true ■ Large dams display a high degree of profitability of these schemes remains variability in delivering predicted water elusive. and electricity services – and related social benefits – with a considerable portion falling short of physical and Perhaps of most significance is the fact that economic targets, while others continue social groups bearing the social and environ- generating benefits after 30 to 40 years. mental costs and risks of large dams, especially the poor, vulnerable and future ■ Large dams have demonstrated a marked generations, are often not the same groups tendency towards schedule delays and that receive the water and electricity significant cost overruns. services, nor the social and economic ■ Large dams designed to deliver irrigation benefits from these. Applying a ‘balance- services have typically fallen short of sheet’ approach to assess the costs and physical targets, did not recover their benefits of large dams, where large inequities costs and have been less profitable in exist in the distribution of these costs and economic terms than expected. benefits, is seen as unacceptable given ■ Large hydropower dams tend to perform existing commitments to human rights and closer to, but still below, targets for sustainable development. power generation, generally meet their financial targets but demonstrate varia- Options for water and ble economic performance relative to electricity services targets, with a number of notable under- and over-performers. Today, a wide range of options for delivering ■ Large dams generally have a range of water and electricity services exists, al- extensive impacts on rivers, watersheds though in particular situations the cost and and aquatic ecosystems – these impacts feasibility of these options will vary in the are more negative than positive and, in face of constraints such as natural resource many cases, have led to irreversible loss endowments and site location. The Com- of species and ecosystems. mission found that: ■ Efforts to date to counter the ecosystem ■ Many of the non-dam options available impacts of large dams have met with today – including demand-side manage- limited success owing to the lack of ment, supply efficiency and new supply attention to anticipating and avoiding options – can improve or expand water

The Report of the World Commission on Dams xxxi Executive Summary

and energy services and meet evolving ness to political discourse in a country. development needs in all segments of However, the WCD Global Review docu- society. ments a frequent failure to recognise affect- ■ There is considerable scope for improved people and empower them to participate ing performance of both dam projects in the process. In some cases, the opportuni- and other options. ty for corruption provided by dams as large- scale infrastructure projects further distorted ■ Demand management, reducing con- decision-making. sumption, recycling and supply and end- use efficiency measures all have significant potential to reduce pressure on Once a proposed dam project passed prelim- water resources in all countries and inary technical and economic feasibility regions of the world. tests and attracted interest from financing agencies and political interests, the momen- ■ A number of supply-side options at all tum behind the project often prevailed over scales (ranging from small, distributed other considerations. Project planning and generation sources or localised water appraisal for large dams was confined collection and water-recovery systems to primarily to technical parameters and the regional-interconnection of power grids) narrow application of economic cost-benefit have emerged that – on their own or analyses. Historically, social and environ- collectively – can improve or expand the mental impacts were left outside the assess- delivery of water and energy services in a ment framework and the role of impact timely, cost-effective and publicly assessments in project selection remained acceptable manner. marginal, even into the 1990s. ■ Decentralised, small-scale options (micro hydro, home-scale solar electric systems, Conflicts over dams have heightened in the wind and biomass systems) based on last two decades due largely to the social and local renewable sources offer an impor- environmental impacts of dams that were tant near-term, and possibly long-term, either disregarded in the planning process or potential particularly in rural areas far unanticipated. However, it also stems from away from centralised supply networks. the failure by dam proponents and financing ■ Obstacles to the adoption of these agencies to fulfil commitments made, options range from market barriers to observe statutory regulations and abide by institutional, intellectual and financial internal guidelines. Whereas far-reaching barriers. A range of incentives – some improvements in policies, legal require- hidden – that favour conventional ments and assessment procedures have options limit the adoption rate of alter- occurred in particular countries and institu- natives. tions, in the 1990s it appears that business-as- usual too often prevailed. Further, past short- Decision-making, planning and comings and inequities remain unresolved, institutional arrangements and experience with appeals, dispute resolution and recourse mechanisms has been poor. The decision to build a dam is influenced by many variables beyond immediate technical considerations. As a development choice, Core Values for Decision- the selection of large dams often served as a Making focal point for the interests and aspirations of politicians, centralised government As the Global Review of dams makes clear, agencies, international aid donors and the improving development outcomes in the dam-building industry, and did not provide future requires a substantially expanded for a comprehensive evaluation of available basis for deciding on proposed water and alternatives. Involvement from civil society energy development projects – a basis that varied with the degree of debate and open- reflects a full knowledge and understanding

xxxii Dams and Development: A New Framework for Decision-Making Executive Summary

of the benefits, impacts and risks of large human rights accorded to all people by dam projects to all parties. It also requires virtue of their humanity. introducing new voices, perspectives and criteria into decision-making, as well as Rights, risks and negotiated processes that will build consensus around outcomes the decisions reached. This will fundamentally alter the way in which decisions are Reconciling competing needs and entitle- made and, we are convinced, improve the ments is the single most important factor in development effectiveness of future decisions. understanding the conflicts associated with development projects and programmes – The Commission grouped the core values particularly large-scale interventions such as that informed its understanding of these dams. The approach developed by the issues under five principal headings: Commission of recognising rights and ■ equity, assessing risks (particularly rights at risk) in the planning and project cycles offers a ■ efficiency, means to apply these core values to deci- ■ participatory decision-making, sion-making about water and energy re- ■ sustainability, and source management. Clarifying the rights context for a proposed project is an essential ■ accountability. step in identifying those legitimate claims and entitlements that may be affected by These five are more than simply issues – the project or its alternatives. It is also a they are the values that run through the pre-condition for effective identification of entire report. They provide the essential legitimate stakeholder groups that are tests that must be applied to decisions entitled to a formal role in the consultative relating to water and energy development. If process, and eventually in negotiating the report advances these values significantly project-specific agreements relating, for we will emerge at our destination – improved example, to benefit sharing, resettlement decision-making processes that deliver im- and compensation. proved outcomes for all stakeholders. The assessment of risk adds an important The debate about dams is a debate about the dimension to understanding how, and to very meaning, purpose and pathways for what extent, a project may impact on achieving development. This suggests that people’s rights. In the past, many groups decision-making on water and energy have not had an opportunity to participate management will align itself with the in decisions that imply major risks for their emerging global commitment to sustainable lives and livelihoods, thus denying them a human development and on the equitable stake in the development decision-making distribution of costs and benefits. The process commensurate with their exposure emergence of a globally accepted framework to risk. Indeed, many have had risks imposed of norms rests on the adoption of the on them involuntarily. Risks must be identi- Universal Declaration of Human Rights in fied and addressed explicitly. This will require 1948 and related covenants and conven- the notion of risk to be extended beyond tions thereafter. These later resolutions governments or developers to include both include the Declaration on the Right to those affected by a project and the environ- Development adopted by the UN General ment as a public good. Involuntary risk bearers Assembly in 1986, and the Rio Principles must be engaged by risk takers in a transparent agreed to at the UN Conference on Envi- process to negotiate equitable outcomes. ronment and Development in 1992. The core values that inform the Commission’s An approach based on the recognition of shared understanding are aligned with this rights and assessment of risks can lay the consensus and rest on the fundamental basis for greatly improved and significantly

The Report of the World Commission on Dams xxxiii Executive Summary

more legitimate decision-making on water rise key principles and actions that the and energy resource development. It offers Commission proposes all actors should an effective way to determine who has a adopt and implement. legitimate place at the negotiation table and what issues need to be included on the 1. Gaining Public Acceptance agenda. Only decision-making processes Public acceptance of key decisions is essen- based on the pursuit of negotiated outcomes, tial for equitable and sustainable water and conducted in an open and transparent energy resources development. Acceptance manner and inclusive of all legitimate actors emerges from recognising rights, addressing involved in the issue are likely to resolve risks, and safeguarding the entitlements of the complex issues surrounding water, dams all groups of affected people, particularly and development. indigenous and tribal peoples, women and other vulnerable groups. Decision making processes and mechanisms are used that Recommendations for a enable informed participation by all groups New Policy Framework of people, and result in the demonstrable acceptance of key decisions. Where projects Researching and analysing the history of affect indigenous and tribal peoples, such water resources management, the emergence processes are guided by their free, prior and of large dams, their impacts and perform- informed consent. ance, and the resultant dams debate led the Commission to view the controversy sur- 2. Comprehensive Options rounding dams within a broader normative Assessment framework. This framework, within which Alternatives to dams do often exist. To the dams debate clearly resides, builds upon explore these alternatives, needs for water, international recognition of human rights, food and energy are assessed and objectives the right to development and the right to a clearly defined. The appropriate develop- healthy environment. ment response is identified from a range of possible options. The selection is based on a Within this framework the Commission has comprehensive and participatory assessment developed seven strategic priorities and of the full range of policy, institutional and related policy principles. It has translated technical options. In the assessment process these priorities and principles into a set of social and environmental aspects have the corresponding criteria and guidelines for key same significance as economic and financial decision points in the planning and project factors. The options assessment process cycles. continues through all stages of planning, project development and operations. Together, they provide guidance on translating this framework into practice. They help 3. Addressing Existing Dams us move from a traditional, top-down, Opportunities exist to optimise benefits technology-focused approach to advocate from many existing dams, address outstand- significant innovations in assessing options, ing social issues and strengthen environ- managing existing dams – including process- mental mitigation and restoration measures. es for assessing reparations and environmen- Dams and the context in which they oper- tal restoration, gaining public acceptance ate are not seen as static over time. Benefits and negotiating and sharing benefits. and impacts may be transformed by changes in water use priorities, physical and land use The seven strategic priorities each supported changes in the river basin, technological by a set of policy principles, provide a developments, and changes in public policy principled and practical way forward for expressed in environment, safety, economic decision-making. Presented here as expres- and technical regulations. Management and sions of an achieved outcome, they summa- operation practices must adapt continuously to

xxxiv Dams and Development: A New Framework for Decision-Making Executive Summary

changing circumstances over the project’s life guidelines, and project-specific negotiated and must address outstanding social issues. agreements is secured at all critical stages in project planning and implementation. A set 4. Sustaining Rivers and Livelihoods of mutually reinforcing incentives and Rivers, watersheds and aquatic ecosystems mechanisms is required for social, environ- are the biological engines of the planet. mental and technical measures. These They are the basis for life and the liveli- should involve an appropriate mix of hoods of local communities. Dams transform regulatory and non-regulatory measures, landscapes and create risks of irreversible incorporating incentives and sanctions. impacts. Understanding, protecting and Regulatory and compliance frameworks use restoring ecosystems at river basin level is incentives and sanctions to ensure effective- essential to foster equitable human develop- ness where flexibility is needed to accommo- ment and the welfare of all species. Options date changing circumstances. assessment and decision-making around river development prioritises the avoidance 7. Sharing Rivers for Peace, of impacts, followed by the minimisation Development and Security and mitigation of harm to the health and Storage and diversion of water on trans- integrity of the river system. Avoiding boundary rivers has been a source of consid- impacts through good site selection and erable tension between countries and within project design is a priority. Releasing tailor- countries. As specific interventions for made environmental flows can help main- diverting water, dams require constructive tain downstream ecosystems and the com- co-operation. Consequently, the use and munities that depend on them. management of resources increasingly becomes the subject of agreement between 5. Recognising Entitlements and States to promote mutual self-interest for Sharing Benefits regional co-operation and peaceful collabo- Joint negotiations with adversely affected ration. This leads to a shift in focus from the people result in mutually agreed and legally narrow approach of allocating a finite enforceable mitigation and development resource to the sharing of rivers and their provisions. These recognise entitlements associated benefits in which States are that improve livelihoods and quality of life, innovative in defining the scope of issues for and affected people are beneficiaries of the discussion. External financing agencies project. Successful mitigation, resettlement support the principles of good faith negotia- and development are fundamental commit- tions between riparian States. ments and responsibilities of the State and the developer. They bear the onus to satisfy If we are to achieve equitable and sustaina- all affected people that moving from their ble outcomes, free of the divisive conflicts of current context and resources will improve the past, future decision-making about water their livelihoods. Accountability of respon- and energy resource projects will need to sible parties to agreed mitigation, resettle- reflect and integrate these strategic priorities ment and development provisions is ensured and their associated policy principles in the through legal means, such as contracts, and planning and project cycles. through accessible legal recourse at the national and international level. From Policy to Practice – 6. Ensuring Compliance The Planning and Project Ensuring public trust and confidence requires that the governments, developers, Cycle regulators and operators meet all commit- The Commission’s recommendations can ments made for the planning, implementa- best be implemented by focusing on the key tion and operation of dams. Compliance stages in decision-making on projects that with applicable regulations, criteria and influence the final outcome and where

The Report of the World Commission on Dams xxxv Executive Summary

compliance with regulatory requirements ■ the promotion of open and meaningful can be verified. Among the multitude of participation at all stages of planning decisions to be taken, the Commission has and implementation, leading to negotiat- identified five key decision points. The first ed outcomes; two relate to water and energy planning, ■ developers to accept accountability leading to decisions on a preferred develop- through contractual commitments for ment plan: effectively mitigating social and environ- ■ Needs assessment: validating the needs mental impacts; for water and energy services; and ■ improving compliance through inde- ■ Selecting alternatives: identifying the pendent review; and preferred development plan from among ■ dam owners to apply lessons learned from the full range of options. past experiences through regular monitoring and adapting to changing needs Where a dam emerges from this process as a and contexts. preferred development alternative, three further critical decision points occur: The Commission offers its criteria and ■ Project preparation: verifying that guidelines to help governments, developers agreements are in place before tender of and owners meet emerging societal expecta- the construction contract; tions when faced with the complex issues ■ Project implementation: confirming associated with dam projects. Adopting this compliance before commissioning; and framework will allow states to take informed and appropriate decisions, thereby raising ■ Project operation: adapting to changing the level of public acceptance and improv- contexts. ing development outcomes.

Social, environmental, governance and compliance aspects have been undervalued Beyond the Commission – in decision-making in the past. It is here An Agenda for Change that the Commission has developed criteria and guidelines to innovate and improve on The Commission’s report identifies the key the body of knowledge on good practices elements of the debate on water and energy and add value to guidelines already in resources management and the role of dams common use. Seen in conjunction with in this debate. It summarises the lessons existing decision-support instruments, the learned from our Global Review of experi- Commission’s criteria and guidelines provide ence with large dams. It elaborates the a new direction for appropriate and sustaina- development framework within which the ble development. controversies and underlying issues can be understood and addressed and proposes a Bringing about this change will require: decision-making process anchored in a rights-and-risks approach and based on ■ planners to identify stakeholders through negotiated outcomes. It offers a set of a process that recognises rights and strategic priorities, principles, criteria and assesses risks; guidelines to address the issues around ■ States to invest more at an earlier stage existing dams and to use in exploring new to screen out inappropriate projects and water and energy development options. facilitate integration across sectors within the context of the river basin; The report is not intended as a blueprint. ■ consultants and agencies to ensure We recommend that it be used as the outcomes from feasibility studies are starting point for discussions, debates, socially and environmentally acceptable; internal reviews and reassessments of what

xxxvi Dams and Development: A New Framework for Decision-Making Executive Summary

may be established procedures and for an together if the issues surrounding water and assessment of how these can evolve to energy resources development are to be address a changed reality. In looking at the resolved. It is a process with multiple heirs future, the Commission proposes a number and no clear arbiter. We must move forward of entry points to help organisations identify together or we will fail. The Commission immediate actions they might take in was given an exceptional opportunity and it response to the Commission’s report. has delivered a result reflecting our collec- Specific proposals are included for: tive learning process and understanding. If ■ national governments and line minis- our report does not win widespread support tries; among participants in the dams debate, it is unlikely that there will be another such ■ civil society organisations; opportunity for a long time. ■ the private sector; ■ bilateral aid agencies and multilateral We believe that our report is a milestone in development banks; the evolution of dams as a development ■ export credit agencies; option. We have: ■ inter-governmental organisations; ■ conducted the first comprehensive global ■ professional associations; and and independent review of the perform- ■ academic and research bodies. ance of essential aspects of dams and their contribution to development. We Engaging through these entry points will have done this through an inclusive initiate permanent changes to advance the process that has brought all significant principles, criteria and guidelines we set out. players into the debate; ■ shifted the centre of gravity in the dams The trust required to enable the different debate to one focused on investing in actors to work together must still be consoli- options assessment, evaluating opportu- dated. Early and resolute action to address nities to improve performance and issues arising from the past will go a long address legacies of existing dams, and way towards building that trust in the achieving an equitable sharing of bene- future. So, too, will an assurance to coun- fits in sustainable water resources devel- tries still at an early stage of economic opment; and development that the dams option will not ■ demonstrated that the future for water be foreclosed before they have had a chance and energy resources development lies to examine their water and energy develop- with participatory decision-making, ment choices within the context of their using a rights-and-risks approach that development process. will raise the importance of the social and environmental dimensions of dams The experience of the Commission demon- to a level once reserved for the economic strates that common ground can be found dimension. without compromising individual values or losing a sense of purpose. But it also demon- We have told our story. What happens next strates that all concerned parties must stay is up to you.

The Report of the World Commission on Dams xxxvii Water, Development and Large Dams

Chapter 1: Water, Development and Large Dams

ver 45 000 times in the last Ocentury, people took the decision to build a dam. Dams were built to provide water for irrigated agriculture, domestic or industrial use,

to generate hydropower or help control floods. But dams also altered and diverted river flows, affecting existing

rights and access to water, and resulting in significant impacts on livelihoods and the environment. Decisions to

build dams are being contested increasingly as human knowledge and experience expand, as we develop new

technologies, and as decision-making becomes more open, inclusive and transparent.

The Report of the World Commission on Dams 1 Chapter 1

At the heart of the current human needs long before the dawn of debate on dams is the way civilisation? choices are made, and the different opinions and At one level, the water issue justifies the perspectives that are ex- priority accorded it because demand for pressed – or denied expres- water is rapidly outstripping available supply sion – in the process. The in large parts of the world. As populations same question of decision- grow, and economic development leads to making is at the centre of higher consumption, demand for water the Commission’s area of increases sharply putting intense pressure on concern. available stocks. This can cause growing social tensions, or even lead to outright Dams are a means to an end, conflict. While commonly expressed predic- not an end in themselves. tions that future wars will largely be fought What is that end? How over water may be exaggerated, nobody central are the challenges doubts that access to both surface and ground that large dams set out to water is an increasingly contentious issue. meet? And how well can dams meet them? Where co-operation gives way to unfair competition between different water uses or The World Commission on between communities and countries, a range Dams considers that the end of any dam of new issues come to the fore. These issues project must be the sustainable improve- relate to the distribution of power and ment of human welfare. This means a influence within societies and between significant advance of human development countries. They concern the relative weight on a basis that is economically viable, of the different factors that make up the socially equitable, and environmentally decision-making mix. And they relate to sustainable. If a large dam is the best way to how choices are made between available achieve this goal it deserves our support. options. Where other options offer better solutions we should favour them over large dams. The issues surrounding dams are the issues Thus the debate around surrounding water, and how water-related dams challenges our view of decisions are made. There is little public The World Commission on how we develop and manage controversy about the choice between an Dams considers that the end our water resources. embankment dam or a gravity dam, or of any dam project must be whether to use earth, concrete or rock-fill, the sustainable improvement Water has attracted the possibly even over financing the develop- of human welfare. attention of political leaders ment. The issues all relate to what the dam at the highest level and has will do to river flow, to rights of access to triggered a series of strategic global initia- water and river resources, to whether it will tives such as the World Commission on uproot existing human settlements, disrupt Water.1 The most powerful players in the culture and sources of livelihood of local development have placed water at the top of communities and deplete and degrade their agendas. Why this sudden attention to environmental resources. Conflicts over a resource that was, after all, central to dams are more than conflicts over water.

2 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

They are conflicts over human development Water and Development and life itself. Today, around 3 800 km3 of fresh water is withdrawn annually from the world’s lakes, If water is life, rivers are its arteries. Dams rivers and aquifers.2 This is twice the regulate or divert the flow through these volume extracted 50 years ago. arteries, affecting the life-blood of humanity. The fact that they aim to do so in the Today, around 3 800 km3 A growing population and a interests of humankind only makes the of fresh water is rising level of economic activity decision to build a large dam more sensitive, withdrawn annually from both increase human demand for one that will trigger a range of apprehen- the world’s lakes, rivers water and water-related services. sions, hopes and fears, both rational and and aquifers. This is twice Development, technological irrational. the volume extracted 50 change, income distribution and years ago. life-styles all affect the level of This Chapter examines the context of the water demand. debate on large dams that led to the establishment of the Commission. It begins with How much water do we need? a broad look at the past and present ‘drivers’ of the demand for water, and the role of World population has passed 6 billion. large dams in meeting this demand. Then it Although the annual increase probably presents general patterns and trends in the peaked at about 87 million around 1990, the development of large dams during the 20th high proportion of young people in most century, along with a brief description of the developing countries means that global main purposes for building dams. The population will continue to increase signifi- chapter also introduces the scale and cantly well into this century. significance of the benefits, costs and Recent projections suggest a peak of be- impacts of large dams, described in more tween 7.3 billion and 10.7 billion around detail in Chapters 2 to 6. 2050 before total population begins to stabilise or fall.3 Predictions cannot be It goes on to look at the central issues in the precise, because other dimensions of devel- large dams debate – the benefits and adverse opment such as access to health, education, impacts – and the positions taken by differ- income, birth control and other services ent constituencies on the past effectiveness influence the pace of population growth. of dams, and what may constitute good practice in future decision-making. Chapter 7 Despite the massive investment in water picks up many of these themes and also resource management and particularly in situates the dams debate in the emerging global development framework. Box 1.1 New paradigm for water use

The final section of the chapter traces the Successfully meeting human demands for water in the next century will increasingly depend on non-structural solutions and a completely new approach establishment of the Commission and to planning and management. The most important single goal of this new highlights the mandate given to it in paradigm is to rethink water use with the objective of increasing the productive use of water. Two approaches are needed, increasing the efficiency with which response to the debate on large dams. It current needs are met, and increasing the efficiency with which water is allocated then lays out the methodology adopted and among different uses. Source: Gleick, 1998 the process followed by the Commission.

The Report of the World Commission on Dams 3 Chapter 1

areas, and contributing to rapidly growing demand for additional services. In heavily populated countries like China, India and Indonesia many question the sustainability of the high rates of urbanisation in mega-cities.

Economic growth and development

World economic activity has grown approximately five-fold since 1950 at a rate of about 4% per year.5 The regional balance is changing, with significant growth in Asia over the past 25 years. At present, OECD countries continue to account for the largest share, amounting to 55% of world production at purchasing power parity, nearly 80% at market prices. 6 dams, billions of children, women and men in rural areas lack access to the most basic Economic growth has two implications for water and sanitation services. Although water demand. The first is that increased problems of access are worst in rural areas, rapid urbanisation is also increasing the economic activity will increase the demand demand for water-related services. In 1995, for water-related services – regardless of 46% of the world’s population lived in urban whether the demand is satisfied by more areas. If current trends persist (and they may efficient use of the existing supply, or by accelerate), that figure could reach 60% by increasing the level of supply. The second is the year 2030 and over 70% by 2050.4 Most that both the development brought about by of this growth will take place in developing economic growth and the technological countries where an estimated 25 to 50% of changes that accompany it will lead to urban inhabitants live in impoverished structural changes in the pattern of goods slums and squatter settlements. Lack of and services that society produces, and in access to water in both rural and urban areas the way these services are provided. The is not just a question of supply. It is partly due water demand per unit of Gross Domestic to inequitable access to existing supplies. Product (GDP) will depend on how these two components of economic growth are Urbanisation implies an increasing concen- combined. Countries with the same product tration on water and energy demand in per capita but different production charac- mega-cities, a switch to different lifestyles teristics – for example, with large scale and consumption patterns, and a loss of irrigated agriculture or water intensive productive agricultural land through urban industries – may consume three or four expansion. It is a widely held view that lack times more water per dollar of GDP. This is of attention to development in rural areas is evident when comparing the United States fuelling unsustainable forms of urban and Canada with Germany or France, or growth, shifting poverty from rural to urban India with China.

4 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

Development and technological change can Drastically lower average figures for domes- also save water per dollar of output. Between tic consumption in developing countries 1950 and 1990 the world economy grew by a reflect not only different life styles and factor of five while world water withdrawals lower incomes, but also a huge backlog of only grew by a factor of 2.7 The last fifty unsatisfied demand. The lower average also years have seen a worldwide reduction in masks extremely high consumption among the amount of water per dollar of non- better-off urban households and acute agricultural production as a result of im- deprivation among rural and urban poor. proved technology, more recycling, enforcement of environmental standards, higher Competing uses of water water prices, and industries moving away Water analysts foresee increased competi- from natural resource intensive activities.8 tion among water users in meeting the However, the water management practices growing demand. They predict that competi- and technologies that enable such advances tion will increase among the three largest are not widely available or promoted and are water users in global terms. Agriculture often absent where they are most needed. accounts for about 67% of withdrawals, Despite the increasing number of options industry uses 19% and munici- available, the total number of people pal and domestic uses account without access to clean water is growing. for 9%.10 Analysts foresee that In 1990 over a billion Income distribution and life-styles these uses will continue to draw people had access to less Economic activity and development affect from the water needed to than 50 litres of water a income, income distribution and lifestyles. sustain natural systems. In dry day. These in turn affect the demand for water climates, evaporation from through changes in the level and composi- large reservoirs, estimated at tion of household consumption in areas such close to 5% of total water withdrawals, may as diet, the use of household appliances and also be a significant consumptive use of standards of sanitation. water.11

How much water is needed for one more Regional trends vary widely as shown in person? Although climate and culture Figures 1.1 and 1.2. Despite increasing influence what constitutes an appropriate urbanisation in Africa, Asia and Latin level of domestic water consumption, America, agriculture is the dominant water several international agencies and experts user in these regions, accounting for approx- have proposed 50 litres per person per day imately 85% of all water used. In all regions (or just over 18.25 m3 a year) as an amount of the world except Oceania, domestic or that covers basic human water requirements household water consumption accounts for for drinking, sanitation, bathing and food less than 20% of water use.12 In Africa, preparation. In 1990 over a billion people Central America and Asia, this is nearer to were below that level.9 5%. In the more developed regions of Europe and North America, industry is the On the other hand, households in devel- major water consumer. The water use oped countries and better-off households in breakdown in a specific country influences developing cities use from 4 to 14 times the where water demand management opportu- threshold of 50 litres per person a day. nities exist.

The Report of the World Commission on Dams 5 Chapter 1

Figure 1.1: Annual fresh water withdrawals as a Low income Lower middle Upper middle High income countries income countries income countries countries percentage of total resources withdrawn (1996)

% for agriculture % for domestic use % for industry

Source: World Bank 1999a.

For many water-stressed countries that are to increased competition for water, unless industrialising, and that have large urban more efficient water use in both sectors populations, the present water crisis often accompanies the transition from agricultural revolves around industrial and domestic to industrial based economies. 13 water supply and sanitation. This is evident in the significant trend over the past few There are additional challenges. To meet decades towards diverting water from food requirements, water used in agriculture agricultural to municipal and industrial uses. may have to increase 15 to 20% by 2025 As this is happening in the context of an even with improvements in irrigation overall increase in withdrawals, it will lead efficiency and agronomic potential.14 In addition to increasing food production in the face of water stress, distribution, equita- Figure 1.2: Annual fresh water withdrawals per capita average (1987-95) ble access, purchasing power and poverty are central issues in meeting food demand.

Beyond competing human demands, water OECD for nature is an essential consideration. The Industrialised countries fresh water ecosystems that provide the Eastern Europe & CIS livelihoods of the world’s riverine communi- Latin America & Caribbean ties and many other goods and services to South Asia our societies depend on water. Arresting, SE Asia & the Pacific and where possible reversing, the accelerat- East Asia ing trend to increasing degradation of many Arab States of the world’s watersheds caused by human Sub-Saharan Africa activity have emerged as an urgent priority. Least developed countries All developing countries World Availability and quality of water 0 200 400 600 800 1 000 1 200 Fresh water withdrawals per capita (m3) Rainfall and other sources of freshwater (rivers, lakes, groundwater) are unevenly distributed around the world and are not Source: UNDP, 1999.

6 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

always located where human water demand Figure 1.3: Distribution of the arises. Pollution threatens surface and Other Brazil world’s water ground water sources and may make them 35% 17% unfit for many uses, or require expensive Russia treatment. Pollution is especially serious 11% where ground water resources are overexploited and suffer from reduced natural Canada rates of recharge due to deforestation, land 7% use changes and urbanisation. India 5% China 7%

Fully one third of the countries in water- Bangladesh USA 6% Indonesia 6% stressed regions of the world are expected to 6% face severe water shortages this century, and within these regions there are great dispari- Source: Gleick, 1998. ties in access to fresh water. Not surprisingly, a significant number of less developed Intergovernmental Panel on Climate countries, including regions of India and Change (IPCC), strongly suggest that global China, are facing severe shortages.15 With warming trends could significantly increase population growth, the number of countries the variability of weather patterns.18 Re- in this category is increasing, and by 2025 sults could include a decline in rainfall in there will be approximately 6.5 times as many people – a total of 3.5 billion – living Figure 1.4 : Selected water stressed countries in water-stressed countries.16 Figures 1.3 and 1.4 show the distribution of world fresh water resources and selected water stressed countries. The uneven distribution of water Saudi Arabia supply means that countries may have water Yemen surplus and water deficit regions. Egypt Israel Not only surface water is under pressure. Korea The growing rate of extraction of fresh Iraq water from rivers and lakes is matched by Madagascar increasing extraction of ground water, with Spain many aquifers now seriously depleted. The Iran volume of ground water withdrawal, prima- Morocco rily for irrigation but also for municipal and Pakistan industrial use, exceeds long-term recharge Germany Italy rates. In many parts of India, Pakistan and South Africa China, the water table is sinking at the rate Poland of one to two meters a year.17 0 20 40 60 80 100 120 140 160 180 % Climate change can also affect the seasonal Water stress: withdrawals over 25% of annual distribution of rains and water availability. water resources Studies and modelling exercises reviewed by the Commission, including those by the Source: Raskin et al, 1995.

The Report of the World Commission on Dams 7 Chapter 1

arid and semi-arid regions, and an increase later. At that time, dams were built in the in seasonal variations in rainfall around the Mediterranean region, China and Meso globe over the next 50 to 100 years. America. Remains of earth embankment dams built for diverting water to large Development and Large community reservoirs can still be found in Dams Sri Lanka and Israel.19 The Dujiang irrigation project, which supplied 800 000 hec- River basins are renowned as the cradles of tares in China, is 2 200 years old.20 Dams civilisation and cultural heritage. Ancient and aqueducts built by the Romans to and modern communities alike have de- supply drinking water and sewer systems for pended on rivers for livelihood, commerce, towns still exist today. habitat and the sustaining ecological functions they provide. Throughout history The first use of dams for hydropower genera- alterations to rivers – natural or human tion was around 1890. By 1900, several generated – have affected riverine commu- hundred large dams had been built in nities in one way or another. different parts of the world, mostly for water supply and irrigation. The earliest evidence of river engineering is the ruins of irrigation canals over eight Dams in the 20th century thousand years old in Mesopotamia. Re- mains of water storage dams found in The last century saw a rapid increase in Jordan, Egypt and other parts of the Middle large dam building. By 1949 about 5 000 East date back to at least 3000 BC. large dams had been constructed worldwide, three-quarters of them in industrialised Historical records suggest that the use of countries. By the end of the 20th century, dams for irrigation and water supply became there were over 45 000 large dams in over more widespread about a thousand years 140 countries.21

Figure 1.5: Regional distribution of China large dams at the end of the 20th Asia century North and Central America Western Europe

Africa

Eastern Europe

South America Austral-Asia

0 5 000 10 000 15 000 20 000 25 000 Number of dams

Source: WCD estimates based on ICOLD,1998 and other sources (See Annex V).

8 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

The period of economic growth following Figure 1.6: Construction of dams by decade (1900-2000) the Second World War saw a phenomenal rise in the global dam construction rate, lasting well into the 1970s and 1980s. At its 6 000 peak, nearly 5 000 large dams were built worldwide in the period from 1970 to 1975. 5 000 The decline in the pace of dam building 4 000 over the past two decades has been equally

dramatic, especially in North America and 3 000 Europe where most technically attractive sites are already developed. The average 2 000 Number of dams large dam today is about 35 years old. 1 000 The top five dam-building countries ac- 0 count for nearly 80% of all large dams 1900s 1910s 1920s 1930s 1940s 1950s 1960s 1970s 1980s worldwide. China alone has built around e 1900 after 1990 22 000 large dams, or close to half the befor world’s total number. Before 1949 it had only 22 large dams. Other countries among the top Source: ICOLD, 1998. Note: Information excludes dams in China. five dam building nations include the United States with over 6 390 large dams; India with over 4 000; and Spain and Japan with between 1 000 and 1 200 large dams each.

Figure 1.7: Dams constructed over time by region (1900-2000)

8 000 Asia 7 000 North America

6 000 Europe

5 000

4 000

3 000 Number of dams

2 000 Africa 1 000 South America 0 Austral-Asia

e 1900 1900s 1910s 1920s 1930s 1940s 1950s 1960s 1970s 1980s after 1990 befor Time

Source: ICOLD, 1998. Note: Information excludes the time-trend of dams in China.

The Report of the World Commission on Dams 9 Chapter 1

Figure 1.5 shows the proportional distribu- Decommissioning of large tion of large dams in different regions of the dams world. Approximately two thirds of the The end of the 20th century saw the emer- world’s existing large dams are in developing gence of another trend relating to large countries. Figures 1.6 and 1.7 show the dams – decommissioning dams that no cumulative and time-line trends in the longer serve a useful purpose, are too expen- construction of large dams in the last sive to maintain safely, or have unaccepta- century. The time-trend figures exclude data ble levels of impacts in today’s view. Mo- on most of the dams in China and therefore mentum for river restoration is accelerating differ slightly from the trends described in in many countries, especially in the United the text. States where nearly 500 dams, mainly Current regional focus for large dam relatively old, small dams have been decom- construction missioned. Since 1998, the decommission- Construction of large dams peaked in the ing rate for large dams has overtaken the 1970s in Europe and North America. Today rate of construction in the United States. most activity in these regions is focused on Experience in North America and in Europe the management of existing dams, including shows that decommissioning dams has rehabilitation, renovation and optimising enabled the restoration of fisheries and the operation of dams for multiple func- riverine ecological processes. However, dam tions. An estimated 1 700 large dams have removals without proper studies and mitiga- been under construction in other parts of tion actions cause public concerns and the world in the last few years. Of this total, environmental problems. These include 40% are reportedly being built in India (see negative impacts on downstream aquatic life Table 1.1 and Annex V for details). due to a sudden flush of the sediments accumulated in the reservoir. Where there Since average construction periods generally has been industrial or mining activity range from 5 to 10 years, this indicates a upstream, these sediments may be contami- worldwide annual average of some 160 to nated with toxic substances. Another 320 new large dams per year. option is to open the floodgates of dams where this is possible, either as a form of decommissioning, or to allow natural river Table 1.1: Dams currently under construction Country Number of dams Purpose flows and unrestricted fish passage at critical India Varies from 695 to 960 depending Irrigation, multipurpose times of the year. While decommissioning on the source of information efforts in the United States and France have China 280 Flood control, irrigation, received public support thus far, there may hydropower including pumped storage, be local opposition where changes in the Turkey 209 Irrigation, hydropower, flow and water levels affect services previ- water supply ously provided by the dam, or where devel- South Korea 132 Irrigation, hydropower, flood management, water opment has taken place around the reservoir supply and downstream. Japan 90 Mainly flood control Iran 48 (above 60m) Irrigation, multipurpose There is comparatively little experience Sources: ICOLD, 1997; International Journal of Hydropower and Dams, 2000; WCD India Case with the removal of larger dams. The bigger Study, 2000; Japan Dam Almanac, 1999, National Register of Large Dams for India cited in WCD India Case Study. the dam, the more problems decommission-

10 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

ing or removal are likely to face, and the Table 1.2: Estimated annual investment in dams in the 1990s more expensive they are likely to be. More ($US billion per year) Developing Developed studies are needed to address the costs, benefits Total ccsountries ccsountries and impacts of decommissioning as the stock Damsforhydropower 12-18 7-10 19-28 of dams ages and choices must be made Damsforirrigation 8-11 between refurbishing and decommissioning. Damsforwatersupply 1.5 3-5 13-18 Damsforfloodcontrol 0.5-1.0 Large Dams as Instruments Total 22-31 10-15 32-46 Source:WCDThematicReviewIII.2,FinancingTrends.Notethatthesefiguresincludegenerating of Development installationsforhydropowerdamsbutnotcanalsandpipeddistributionsystemsservedbyirrigationand watersupplydams. Dams have been promoted as an important means of meeting perceived needs for water times the dry season flow. In these settings and energy services and as long-term, strategic dams have been constructed to capture and investments with the ability to deliver store water during wet seasons for multiple benefits. Some of these additional release during dry seasons. Dams are promoted as an benefits are typical of all large public infra- important way to meet structure projects, while others are unique to Large dams require significant water and energy needs dams and specific to particular projects. financial investments. Estimates suggest a worldwide investment and support economic development. Regional development, job creation and of at least two trillion US dollars fostering an industry base with export in the construction of large dams capability are most often cited as additional over the last century. During the 1990s, an considerations for building large dams. estimated $32-46 billion was spent annually Other goals include creating income from on large dams, four-fifths of it in developing 22 export earnings, either through direct sales countries. Of the $22-31 billion invested of electricity, or by selling cash crops or in dams each year in developing countries, processed products from electricity intensive about four-fifths was financed directly by the industry such as aluminium refining. public sector.

Today the world’s large dams regulate, store Water-rich countries such as Canada, and divert water from rivers for agricultural Norway, Brazil and parts of Russia have developed large dams for hydropower generation where suitable sites were availa- Box 1.2: Types of large dams ble. Governments in semi-arid countries There are various definitions of large dams. The International Commission on Large such as South Africa, Australia and Spain Dams (ICOLD), established in 1928, defines a large dam as a dam with a height of have tended to build dams with large storage 15m or more from the foundation. If dams are between 5-15m high and have a reservoir volume of more than 3 million m3, they are also classified as large dams. capacity to match water demand with stored Using this definition, there are over 45 000 large dams around the world. supply, and for security against the risk of The two main categories of large dams are reservoir type storage projects and drought. For example, in Spain – one of the run-of-river dams that often have no storage reservoir and may have limited daily pondage. Within these general classifications there is considerable top five dam-building countries – rainfall is diversity in scale, design, operation and potential for adverse impacts. highly variable between seasons and from Reservoir projects impound water behind the dam for seasonal, annual and, in year to year. some cases, multi-annual storage and regulation of the river. Run-of-river dams (weirs and barrages, and run-of-river diversion dams) create a hydraulic head in the river to divert some portion of the river flows to a canal or In East and Southeast Asia, during the power station. monsoon season, rivers swell to over 10

The Report of the World Commission on Dams 11 Chapter 1

production, human and industrial use in has a relatively even spread of large dam towns and cities, electricity generation, and functions. All other potential purposes, flood control. Dams have been constructed including recreation and navigation, were to a lesser extent to improve river transpor- found in less than 5% of projects. Overall, tation and, once created for other purposes, the proportional share of irrigation dams the reservoirs of many large dams have been and multi purpose dams has been increasing used for recreation, tourism, and aquaculture. over the last 20 years, while the share of hydropower dams has been decreasing. Figure 1.8 shows that about one third of large dams serve two or more purposes. Irrigation water supplied from Recent trends have favoured multi-purpose large dams dams. As the figure shows, there is considerable regional variation in the functions Irrigation is the single largest consumptive use served by large dams and these functions of fresh water in the world today. It is linked to have also changed over time. food production and food security. About one fifth of the world’s agricultural land is irrigated, The majority of large dams in Africa and and irrigated agriculture accounts for about Asia are for irrigation, though large dams are 40% of the world’s agricultural production.23 more often than not multi-purpose. There is The total area irrigated expanded dramatically growing interest in dams for flood protec- during the first years of the green revolution in tion and in pumped storage dams for power the 1960s, increasing yields and bringing down generation to meet peak demand in Asia. food prices. From 1970 to 1982, global growth Single-purpose hydropower dams are most in the irrigated area slowed to 2% a year. In common in Europe and South America, the post green revolution period between whereas single-purpose water supply projects 1982 and 1994 it declined to an annual dominate in Austral-Asia. North America average of 1.3%.24

Figure 1.8: Distribution of existing large 100% 19% 30% 25% 24% 14% 23% dams by region 90% and purpose 2% 80% 1% 3% 6% 4% 70% 7% 19% 2% 20% 2% 60% 24% 26% 31% 50% 44% 40% 11% 13% 65% 16% 30% 10% 52% 15% 25% 20% 11% 19% 10% 1% 2% 2% 3% 0% 13% 17% Africa North South Asia Austral-Asia Europe America America Flood Control Irrigation Water Supply Hydropower Other Single Purpose Multipurpose

Source: Adapted from ICOLD, 1998 (See Annex V).

12 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

Half the world’s large dams were built exclu- with the balance coming mainly from ground sively or primarily for irrigation, and an water sources. There is some controversy in estimated 30 to 40% of the 268 million calculating the percentage of food production hectares of irrigated lands worldwide rely on attributable to dams, and particularly on the dams. Discounting conjunctive use of ground methods used to account for conjunctive use water and surface water, dams are estimated to of surface and ground water.26 contribute to at most 12-16% of world food production. Ground water irrigation typically Unsustainable irrigation practices have has higher yields, for a given amount of water, affected more than a fifth of the world’s than surface water irrigation because of better irrigated area in arid and semi-arid regions. As control of the resource at farm level.25 a result, soil salinity and waterlogging either make agriculture impossible, or limit yields Four countries – China, India, United States and the types of crop that can be grown. In and Pakistan account for more than 50% of other regions, over-use of tubewells has the world’s total irrigated area. As Figure 1.9 depleted ground water aquifers, lowering water shows, the scale and significance of large dams tables and making extraction increasingly for irrigation varies significantly from country expensive and especially difficult for small- to country in terms of the percentage of holders. The absence of effective policies on agricultural land irrigated, and the proportion conjunctive use of ground water and surface of the irrigation water supplied from large water resources is one of the most important dams. Dams supply the water for almost 100% concerns. of irrigated production in Egypt – most coming from the Aswan High Dam – while in Nepal Water for industrial use and and Bangladesh dams provide only 1% of urban centres irrigation water. In the two countries with the largest irrigated areas – India and China – Globally, urban water consumption accounts official statistics suggest large dams supply for 7% of total freshwater withdrawals from approximately 30 to 35% of irrigation water, rivers, and 22% from lakes.27 Many reservoirs

Figure 1.9: Agricultural land 100 Egypt irrigated from 90 Uzbekistan Pakistan dams 80 Japan 70 Vietnam 60 China India 50 Korea – Republic of 40 Nepal

30 Bangladesh Spain Mexico Thailand

% of arable land irrigated 20 Turkey Malaysia 10 France Russia South Africa Australia 0 0 10 20 30 40 50 60 70 80 90 100 % of irrigated area with water sourced from dams

Source: WCD Thematic Review IV.2 Irrigation Options, Section 1.3.

The Report of the World Commission on Dams 13 Chapter 1

were built to provide a reliable countries and over 50% in 63 countries. supply of water to meet rapidly About a third of the countries in the world growing urban and industrial currently rely on hydropower for more than needs, especially in drought- half of their electricity needs. Five countries prone regions where natural – Canada, United States, Brazil, China and ground water sources and existing Russia – account for more than half the lakes or rivers were considered world’s hydropower generation. Between inadequate to meet all needs. 1973 and 1996 hydropower generation in Globally, about 12% of large dams are non-OECD countries grew from 29 to 50% designated as water supply dams. About percent of world production, with Latin 60% of these dams are in North America America increasing its share by the greatest and Europe. amount in that period.30

The extent to which cities rely on dams and Hydropower has been perceived and pro- reservoirs for urban and industrial water moted as a comparatively clean, low-cost, varies greatly even within countries. In the renewable source of energy that relies on Saxony region of Germany, reservoirs proven technology. Except for reservoir provide 40% of the water supplied to two evaporation, it is a non-consumptive use of million people, while Los Angeles derives water. Once built, hydropower, like all 55% of its water supply from local ground renewable sources, is considered to have low water resources and 37% from a system of operating costs and a long life, particularly reservoirs and pipelines that bring water for run-of-river projects and reservoir from more distant locations. Ho Chi Minh projects where sedimentation is no concern. City in Vietnam gets 89% of its water from In the past, hydropower was especially surface sources, whereas Hanoi gets 100% attractive to governments with limited fossil from ground water.28 fuel resources, who would otherwise have had to import fossil fuels to sustain power Electricity generation for the generation. At the global scale, current national grid levels of hydropower generation offset 4.4 million barrels of oil-equivalent (thermal Electricity generation is an important reason electric generation) a day, roughly 6% of the for building large dams in many countries, world’s oil production. either as the primary purpose, or as an additional function where a dam is built for Protecting against floods other purposes. Over the last 22 years, global electricity produc- While natural floods have many beneficial Hydropower provides tion has more than doubled, functions they also pose a threat to life, 19% of world’s total though access is highly skewed health, livelihoods, and property. They electricity supply with 24 between and within countries.29 remain among the world’s most frequent and countries depending on damaging disasters. Floods affected the lives, it for more than 90% of Hydropower currently provides on average, of 65 million people between their supply. 19% of the world’s total electric- 1972 and 1996, more than any other type of ity supply and is used in over disaster, including war, drought and fam- 150 countries. It represents more than 90% ine31 . During the same period, an estimated of the total national electricity supply in 24 3.3 million people were left homeless every

14 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

year as a result of floods. In terms of its region- Physical transformation of al significance, flood-related economic losses rivers in Asia exceeded those in North America and Large dams have fragmented and trans- Europe between 1987 and 1996.32 formed the world’s rivers. The World Resources Institute (WRI) found About 13% of all large dams in the world – that at least one large dam in more than 75 countries – have a flood modifies 46% of the world’s 106 management function.33 Although dams Large dams have primary watersheds.35 The extent have historically been extensively used as a fragmented and to which river flows have been defence against floods, recent approaches transformed the world’s changed varies around the world. see flood protection as more than building rivers, modifying 46% of The United States and the defences against rising waters. For instance, primary watersheds. European Union regulate the reporting on the 1998 flood, Chinese flow of 60–65% of the rivers in their territo- authorities acknowledged that its severity ries, though the amount varies from basin to was partly due to long-term environmental basin. Spain’s 53 km3 of storage behind large degradation and heavy logging throughout dams regulates 40% of its river flow, varying the affected watersheds.34 There are also from 71% in the Ebro river basin, to 11% in cases where dams have created or worsened the basins on the Galicia coast.36 In Asia, floods due to dam breaks, poor reservoir just under half the rivers that are regulated operation and changed downstream sedi- have more than one large dam.37 mentation patterns that reduce river channel capacity. The modification of river flows on transboundary rivers has particular implications. The last two decades have seen a thorough There are 261 watersheds that cross the re-evaluation of what constitutes the political boundaries of two or more coun- appropriate mix of prevention, defence and tries38 . These basins cover about 45% of the mitigation against flood disasters. As a earth’s land surface, account for about 80% of result, the focus on controlling floodwaters, global river flow and affect about 40% of the dominant in the 1950s–1960s, has lost ground to more environmentally based and integrated approaches. The reasons include frustration at floods occurring despite Box 1.3: Changing physical attributes and impacts of large dams structures being in place, the high cost of The volume of the reservoir relative to the annual river flow is important in engineering solutions, and a better under- relation to the purpose of the dam and how it is operated. It is also a major factor in the scale of the ecological effects. The surface area of the reservoir, or standing of how natural systems work. the area flooded, points to the potential resettlement impact. The majority of large dams have reservoir surface areas of 0-1 km2 (more than 60%). This includes run-of-river dams with no reservoir. A small percentage of dams (2%) Problems Associated with have reservoir areas greater than 100 km2. During the first half of the 20th Century, the average height, reservoir volume Large Dams and reservoir area increased in all regions. The average height of new dams was 30-34 meters from 1940 through 1990, but has increased to about 45 meters in While dams have contributed to economic the 1990’s, largely due to trends in Asia. The average reservoir area dramatically growth in the 20th century, the services they increased from 1945 into the 1960’s to 50 km2; declined through the 1980’s to and average of 17 km2; and has again increased in the 1990’s to around 23 km2. provide have come at a cost. This section Source: WCD analysis of ICOLD, 1998 gives an overview of the problems associated with large dams.

The Report of the World Commission on Dams 15 Chapter 1

world’s population.39 Issues and fish species, and provide many ecosystem The construction of large concerns facing riparian countries functions ranging from nutrient recycling dams has led to the in these basins range from water and water purification to soil replenishment displacement of some 40 quality to flow volumes. and flood control. At least 20% of the to 80 million people world’s more than 9 000 fresh water fish worldwide. The aggregate storage capacity species have become extinct, threatened or of large dams, based on dam endangered in recent years.43 design, is about 6 000 km3.40 Assuming that half the design storage is achieved in reality, Fish are a critical source of animal protein for the aggregate real storage capacity of large more that 1 billion people. In Africa, fish dams globally is similar to total freshwater protein is 21% of the total animal protein in withdrawals estimated at around 3 800 km3.41 the diet, and in Asia it is 28%. While rivers supply about 6% of fish protein consumed by An estimated 0.5–1% of the total fresh humans globally, it is often 100% of the supply water storage capacity of existing dams is for many inland riverine communities. lost each year to sedimentation in both large and small reservoirs worldwide.42 This Ecosystem transformations do not only means that 25% of the world’s existing fresh occur in the upper, lower and mid-reaches of water storage capacity may be lost in the watersheds, they also impact on river next 25 to 50 years in the absence of meas- estuaries, which are frequently complex ures to control sedimentation. This loss would ecosystems. Closing the mouths of major mostly be in developing countries and regions, rivers, salt intrusion, destruction of man- which have higher sedimentation rates. groves and loss of wetlands are among the many issues at stake. Riverine ecosystems impacts Growing threats to the ecological integrity Social consequences of large of the world’s watersheds come dams from rising populations, water While many have benefited from the pollution, deforestation, with- services large dams provide, their construc- drawals of water for irrigation tion and operation have led to many signifi- and municipal water supply and cant, negative social and human impacts. the regulation of water flows The adversely affected populations include resulting from the construction directly displaced families, host communi- of large dams. Among the many ties where families are resettled, and riverine factors leading to the degrada- communities, especially those downstream tion of watershed ecosystems, of dams, whose livelihood and access to dams are the main physical resources are affected in varying degrees by threat, fragmenting and trans- altered river flows and ecosystem fragmenta- forming aquatic and terrestrial tion. More broadly, whole societies have lost ecosystems with a range of effects access to natural resources and cultural that vary in duration, scale and heritage that were submerged by reservoirs degree of reversibility. or rivers transformed by dams.

The watersheds of the world are The construction of large dams has led to the habitat of 40% of the world’s the displacement of some 40 to 80 million

16 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

people worldwide, as shown in Chapter 4.44 World Bank funded projects The last three decades Many of them have not been resettled or involving displacement, dams especially have seen received adequate compensation, if any.45 and reservoirs accounted for dramatic and wide- Between 1986 and 1993, an estimated 63% of people displaced.52 reaching changes in 4 million people were displaced annually by These figures are only estimates perceptions of an average of 300 large dams starting and do not include people development and construction each year.46 These figures only displaced by other aspects of the concepts of give an idea of the size of the problem, since projects such as canals, power- interdependence with there are enormous variations from case to houses, project infrastructure other people and with case. and associated compensatory nature. measures such as bio-reserves. The scale and range of social issues encountered in the river basins altered by large dam These are not the only problem areas and construction vary from region to region. Chapters 2 to 6 examine other reasons for The number of people deriving their liveli- dissatisfaction. hood directly from the river and its ecosystem, and the overall population density in Understanding the Large the river basin, gives an indication of the potential impacts. As table 1.3 illustrates, Dams Debate population density can vary significantly. The huge growth of dam building in the 20th century took place against a backdrop of The world’s two most populous countries – tremendous political, economic and techni- China and India, have built around 57% of cal transformations – while the world’s the world’s large dams – and account for the population grew from 1.65 billion in 1900, largest number of people displaced. In the to 6 billion by the end of the century. The late 1980s China officially recognised some last three decades especially have seen 10.2 million people as ‘reservoir resettlers’, dramatic and wide-reaching changes in though other sources suggest the figure may perceptions of development and concepts of be substantially higher. 47 Large dams and interdependence with other people and with reservoirs already built in the Yangtze basin nature. This dynamic of change is also alone have led to the relocation of at least redefining the roles that governments, civil 10 million people.48 In India, estimates of the society and the private sector play in water total number of people displaced due to large dams vary from 16 to 38 million people.49 Table 1.3: Population density of selected river basins Population Resettlement caused by large dams has been Country River basin ddyensity a significant part of total resettlement for all ()(person/km2) China Yangtze 224 public infrastructure development. In Brazil Tocantins/Amazon 6 China, large dams are estimated to have India Ganges 375 displaced an estimated 27% of all people Norway GlommaandLaagen 26 displaced by development projects (the total SouthAfrica OrangeRiverBasin 12 includes people displaced by urban expan- Thailand Mun/Mekong 78 50 sion, roads and bridges). In India the UnitedStates Columbia/Snake 9 figure is 77% (this total excludes people ZimbabweandZambiaZambezi 35 51 displaced by urban development). Among Source:Revengaetal,1998.

The Report of the World Commission on Dams 17 Chapter 1

and energy resource develop- Records from the 17th century tell of Scot- As dam building ment planning. And while the tish fishermen trying to destroy a newly accelerated after the wider transformations have completed weir. In the 1910s, conservation- 1950’s, opposition to spurred the debate on large ist John Muir unsuccessfully lobbied public dams became more dams, that debate itself has opinion and the US Congress against the widespread, vocal and become a major catalyst of building of O’Shaugnessy dam in Yosemite organised. change. National Park in California.

We begin by looking at the genesis of Populations affected or threatened by dams opposition to large dams from the perspec- have fiercely resisted dam building through- tive of social and environmental move- out the last century. Because they were often ments, before considering the main issues in isolated, without help from outside sympa- the debate that the meeting in Gland Switzer- thisers, affected people’s resistance to dams land handed down to the Commission. often went unnoticed internationally and, in some cases, the states concerned used Genesis of the opposition to intimidation and violence to suppress it. dams Eight people died and over 30 were injured In the previous section we saw that dam when colonial government forces fired on building has a long history. Conflicts too Tonga people resisting removal to make way have a long history, though it is only in for the Kariba dam reservoir. But interna- recent years that they have come to com- tional attention focused on the mission to mand wider attention. Conflicts over water rescue wildlife stranded in the impound- 53 and dams are probably as ancient as dam ment area. At around the same time in building itself. In medieval England, boat Mexico, the Papaloapan River Commission owners opposed millers blocking rivers to set fire to the houses of indigenous Mazatecs create millponds to turn their water wheels. who refused to move for the Miguel Aleman dam. In 1978 police killed four people when they fired at an anti-resettlement rally at Chandil dam in the state of Bihar in India. In Nigeria in April 1980, police fired at people blocking roads in protest against the Bakolori dam.54 And in 1985, 376 Maya Achi Indians, most of them women and children, were murdered in the course of clearing the area to be submerged by the Chixoy dam in Guatemala55 .

As dam building accelerated after the 1950’s, opposition to dams became more widespread, vocal and organised. Conserva- tionists in northern countries, especially in the United States, led the first notable successes for campaigns against large dams.56 There, conservationists were able to stop

18 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

the 175 metre-high Echo Park dam on a priority target of dam critics, as it tributary of the Colorado River in the 1950s is often the first and single largest and two dams planned for the main stem of financier of large dams. In 1982, the Colorado River in the Grand Canyon in the Bank adopted an internal the following decade. A series of new laws directive on indigenous peoples. (notably the 1969 National Environmental Revisions of policies on resettle- Protection Act, and the 1974 Endangered ment and environmental assess- Species Act), together with growing public ment are other important mile- concern about environmental conservation, stones. More recently, in 1993, public subsidies and budget deficits served the World Bank established an increasingly as grounds for halting expensive appeals mechanism, the Inspec- dams, canals and channelisation projects tion Panel. This allows citizens through the 1970s and early 1980s. adversely affected by Bank funded projects to file claims regarding Over the past thirty years, the alliance of violations of its policies, procedures northern activist groups (environmental and and loan agreements. At the same human rights groups) with NGOs and time the Bank promoted more affected groups’ associations in the South flexible approaches to information has resulted in more vigorous and more co- disclosure.58 ordinated opposition to dams worldwide. In many cases, the strength of these coalitions Often, these policy reforms have been has had a major impact on dam-related closely related to the impacts of resistance planning and policy and at the level of from affected groups and international individual dams. As a result of these con- NGO campaigns around individual projects. certed pressures the planning process, which In 1973-77, the resistance of indigenous until the 1970s was the restricted preserve of peoples to four dams along the Chico River government agencies, engineers and econo- in the Philippines led the World mists, began to include environmental Bank to withdraw from the impact assessments and some public reviews. project and resulted in the The last few decades By the late 1980s environmentalists and government postponing it indefi- have seen a number of sociologists began to play a more important nitely.59 Other important mile- significant policy steps role in the planning process, and by the stones include the World Bank’s in response to civil mid-1990s the involvement of affected withdrawal from the Sardar society demands and peoples and NGOs in the process became Sarovar project in India in 1993, changing values. more significant.57 and then from Arun III in Nepal in 1995 (although in the latter case public It is only fair to note that popular action has pressure may not have been the deciding also supported dams. Farmers in Madrid factor)60 . Sardar Sarovar, still on the agenda recently marched to demand more water despite the World Bank withdrawal, contin- and more dams for irrigation. ues to be the focus of local opposition and international support. More recent examples The last few decades have seen a number of of campaigns include the Three Gorges significant policy steps in response to civil dams in China, the Pak Mun in Thailand, society demands and changing values. The Ilisu in Turkey, Ralco in Chile, Epupa in World Bank offers a good illustration. It is a Namibia, the Lesotho Highlands Water

The Report of the World Commission on Dams 19 Chapter 1

Project involving Lesotho hospitals, conservation areas, shopping and South Africa, and complexes, highways, parking areas, and Nam Theun II in Laos. many more.

As earlier sections of this If dams are not the only infrastructure chapter showed, the pace projects subject to rising criticism, why do of dam construction has they seem to be often at the centre of slowed significantly in controversy, dispute, and even violent recent years. This is partly confrontation? There are many reasons because industrialised largely related to the scale and scope of the countries have used most of dams and the impacts. their attractive sites, and ■ Large dams represent major investments, because of other factors, such as shifts in and in some cases may be the largest dam financing from public to private single investment in a country. These sources, matched with the decline in donor investments are essentially irreversible funding and increasing costs of large dams. and often highly politically charged. However, it also reflects the effectiveness of ■ Large dams are generally justified by anti-dam strategies developed by environ- national or regional macro-economic mental and human rights activist groups benefits while their physical impacts are worldwide. locally concentrated, mostly affecting those within the confines of the river One frequently asked question is why the valleys, and along the river reaches. The debate focuses so centrally on dams. The mismatch of benefits and costs translates issues being debated, such as environmental easily into confrontational attitudes. sustainability, equitable development, ■ Resettlement for large dams tends to be transparent and participatory decision- on a larger scale than resettlement for making also apply to other large infrastruc- other types of physical infrastructure. ture projects and can only be addressed in Roads and thermal power stations can be the context of broad societal change. sited on marginal land whereas dams generally flood rich and fertile agricul- The view that environmental and human ture land. rights groups have singled out large dams as their main target is misleading. One assess- ■ Those resettled from dam or reservoir ment found that, of the 36 World Bank sites very often lose not only their homes supported projects that NGO activist groups but also their livelihoods. Relocation in have targeted with some success, only 12 are rural settings where good land is already dam projects, compared to 14 forest and occupied can be problematic. natural resource management projects, five ■ Large dams affect critical, life sustaining mines or industrial management projects, resources, such as land, fisheries and the and two urban infrastructure projects.61 In quality and allocation of freshwater, an fact, large dams, like many other industrial, increasingly scarce and coveted resource. commercial, and governmental facilities are ■ Lack of adequate and accepted solutions to increasingly contentious and difficult to site the social and environmental impacts of and build, as are hazardous waste decontam- large facilities has resulted in increased ination facilities, solid waste landfills, social mobilisation around these issues.

20 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

■ The lower than anticipated economic perspective of ‘nation building’ and resource performance of many projects. allocation. To its proponents, the answer to the question of past performance is self- These factors aside, the perceived injustice evident. Dams have generally performed in the distribution of the benefits and well as an integral part of water and energy impacts, and the increased concern about resource development strategies in over 140 the environmental implications indicate nations and, with a few exceptions, dams that the debates, controversies and conflicts have provided an indispensable range of surrounding large dams are not about dams water and energy services. alone. They are part of a wider debate about development, a debate where diverging views Beyond this rough characterisation, it is on the use of natural resources and public possible to disaggregate the debate along a financial resources confront each other. few central themes. The way the debate is structured largely determines how it is Disaggregating the debate conducted at national and international levels. Clearly the dams debate has many layers and many interconnected issues. The debate Performance: costs and benefits is partly about what occurred in the past and Performance is often measured in terms of continues to occur today, and partly about whether the project delivered the benefits what may happen in the future if more dams that were the basis for approval and funding are built – or are not built. The extent to of the project. But there which the debate is driven by social or is more to judging per- Box 1.4: Central issues in the dams environmental concerns, or by broader formance than this. An debate: past and present development considerations, varies from assessment of overall country-to-country. The dams debate in the Performance: costs and benefits performance, or perform- Environmental impacts and sustainability United States, where the rate of decommis- ance over the full life- Social impacts and equity sioning dams is greater than the rate of Economics and finance cycle of a dam, often Governance and participation construction, is perhaps as intense as, but yields many different Wider development impacts of dams qualitatively different from, the debate in Alternatives to dams results. So too with costs, Cross-cutting issues India which, along with China, is now much depends on how building more dams than any other country. completely costs are internalised, and who The two principal poles in the dams debate, bears particular costs compared to how the much in evidence at the Gland Workshop benefits are shared. There is no easy formula in 1997, help to define the range of views for calculating the costs and benefits to yield on past experience with large dams. The a quick and easy judgement on the overall first focuses on the gap between the prom- balance. ised benefits of a dam and what happens on the ground. The review of global experience Environmental impacts and with dams set out in Chapters 2-6 confirms sustainability that these concerns are justified. Dams have At one level, the debate revolves around often not met their targets. how to measure the scale of the impacts, whether these impacts can be avoided or The other pole looks at the challenges of successfully mitigated, and whether they are water and energy development from a reversible or irreversible. A more fundamen-

The Report of the World Commission on Dams 21 Chapter 1

tal controversy centres on how Dealing with the legacy left by forced environmental considerations are relocation under both authoritarian and valued against immediate human democratic regimes and the need to allocate development needs. For commu- responsibility for redress are other issues in nities who depend directly on the debate. An issue that has caused a great river resources for their liveli- deal of tension in the past concerns the hoods, the environment is the basis on which trade-offs, such as the basis for meeting their needs. An potential benefits to many at the cost of example is the value placed on hardship for a few, are invoked and decided. biological diversity, or on the ecosystem functions that may be Economics and finance lost when the river flow is Controversy also surrounds the limits and regulated. This debate becomes the ability of methods for economic assess- especially heated where other ment to fully capture and reflect the various options are available. Other social and environmental impacts and controversies concern wider values. Governments and financial institu- regional or global environmental tions continue to use traditional economic impacts, for example whether and financial analysis – rate of return, dams emit greenhouse gases or discount rates, sensitivity tests and the reduce acid rain, under what exclusion or inclusion of indirect costs – as conditions, and to what extent relative to primary decision criteria. The debate is how the alternatives. This aspect of the debate adequately these are applied in practice and extends to whether large dams should be how they are balanced against other devel- included in climate change protocols, and opment objectives or criteria. Related issues whether and how dams should be treated in include the cost recovery levels for all types future carbon emission trading schemes. of dam projects, the implications for subsidised use of water and the equity dimensions Social impacts and equity of these subsidies. This includes both the scale of the impacts and the distribution of costs, benefits and Governance and participation impacts, including those borne by relocated The principal considerations related to families, host communities where families these issues centre on the transparency and are resettled, and riverine communities openness of options assessment, and how affected by the change in river flows and planning and decision-making processes are access to resources. Social issues go beyond conducted. Other issues relate to the meth- equity in the distribution of benefits and ods used to reconcile local or community- impacts and relate to fundamental rights. led planning and consensus-building proc- They include: esses with more traditional and centralised ■ the burden placed on indigenous peoples planning approaches, access to information, and ethnic minorities and the degree of and the dominance of single agencies in recognition of their distinctive status; planning with multiple responsibilities for ■ the impact on gender and basic human designing, construction and operating large rights; and infrastructure projects. At the heart of the ■ the loss of livelihood and health impacts debate on these issues is the degree of in rural areas. involvement of affected people and wider

22 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

groups of stakeholders in needs assessment incentives for different options The degree of even- and project-level decision-making. Imple- that provide the same service. handedness applied in mentation creates its own set of governance Cross-cutting issues considering alternatives challenges, including whether agreed to large dams is, standards are followed for social and envi- A range of cross-cutting issues perhaps, one of the most ronmental mitigation, compensation and turn on the role and influence of contested issues. enhancement. various public and private sector interests in the planning and Wider development impacts of dams decision making process. This includes the Many of the controversial issues go beyond roles and influence of industry groups the impact of the project itself and touch ranging from domestic and international upon wider regional or national develop- consultants, to developers, contractors and ment choices. Examples include the propor- suppliers, and extends to the financial tion of the development budget allocated to service providers. The financing role is large dams as opposed to other uses of public especially critical and includes the multilat- funds, the impact of an investment in a dam eral and bilateral development banks, on the country’s debt burden, and competi- insurers and export credit agencies, as well tiveness considerations linked to subsidies. as the commercial banks. Issues raised in the There are also more positive considerations, debate range from harmonising standards for including the potential of dams to contrib- financing dam construction to steps to ute to export earnings. address corrupt practices that can distort Alternatives to dams decision-making. There are numerous other cross-cutting issues such as the transboundary The degree of even-handedness applied in implications of dams on shared considering alternatives to large dams is, rivers. perhaps, one of the most contested issues. It Since the United Nations raises the question of whether dams are These examples illustrate the Charter (1945) and The selected over other options that may meet terrain, scope and complexity of Universal Declaration on the water development or energy objectives the debate, and how it has Human Rights (1948), a at lower cost, or that may offer more sustain- become intertwined with wider globally accepted able and equitable development benefits. development concerns. development framework This aspect of the debate extends to wheth- setting out universal er, and on what basis, dams should be Economic development during goals, norms, and considered complementary to, or mutually th standards has been exclusive of other options of different scales the first half of the 20 century gradually emerging. and types. Whether to give primacy to was dominated by an approach options such as demand-side management or that emphasised harnessing and improving the efficiency of existing supply appropriating water and other natural assets, and under what conditions, are also resources for economic activities. Since the debated. The options debate connects with United Nations Charter (1945) and The the political economy of decision-making, Universal Declaration on Human Rights and therefore to the distribution of power (1948), a globally accepted development and influence within societies. This includes framework setting out universal goals, how choices are made between available norms, and standards has been gradually options, and the extent to which market or emerging. These declarations have been other institutional factors create barriers and augmented over time by the Convention

The Report of the World Commission on Dams 23 Chapter 1

Concerning the Protection and Integration ensure that basic human rights and aspira- of Indigenous and Other Tribal and Semi- tions are fulfilled for their citizens. Compli- Tribal Populations in Independent Coun- ance with existing laws and policies is tries (1959), the International Covenant on another important issue. Economic, Social, and Cultural Rights, (1966), the UN Declaration on the Right to Is there a way forward? Development (1986), and the Rio Declara- All parties involved in the debate have their tion on Environment and Development own views on what needs to be done to (1992), among others. Together they form address the problems that they have identi- the current framework for sustainable fied. Some of the main suggestions made by human development. dam critics and proponents in various declarations and statements are as follows. This framework has been widely ratified by the world’s nations, Dam critics point to: although differences emerge in ■ the need for more sustainable and setting priorities and emphases, appropriate alternatives to dams; and in fulfilling, implementing, and resolving conflicts between ■ the imperative for improved transparen- competing rights. The realisation cy, accountability and public participa- of these rights presumes recognition in the planning of water and energy tion of conflicts between com- projects; peting rights and the setting up ■ the importance of prior project approval of mechanisms for negotiation by potentially affected groups; and conflict resolution. We ■ the need for protecting and promoting discuss the development frame- the rights of potentially affected peoples, work as it relates to the WCD’s and for setting in place measures to analysis and methodologies later reduce inequities; and in this chapter, and Chapter 7 ■ the necessity of reparation measures to presents an approach for design- address the legacy of unfulfilled commit- ing mechanisms to resolve and ments and unresolved problems. negotiate competing rights. Dam proponents underline: Where international laws, norms and ■ the evolution and change in practices aspirations are reflected in national legal over time; and policy frameworks, a number of obstacles often prevent their translation into ■ the recognised need for social and reality. These obstacles include lack of environmental concerns to be elevated political will or incentive, especially where to the same level as safety concerns. the democratisation process has not reached (They believe these environmental and a level where citizens and civil society social costs must be avoided, or mini- organisations are strong enough to influence mised and properly mitigated when they decision-making. Another major obstacle, occur.); especially facing poor countries, relates to ■ the importance of ensuring that affected the lack of capacity (financial and human people are better off as a result of dam resources but also institutional capacity) to development, and of considering them as

24 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

shareholders, partners, and therefore debate – is the lack of mutual trust between project beneficiaries; the key parties involved. For opponents, ■ the imperative of participatory decision- dams promoters too often agree to new making; policies and guidelines, but fail to comply with them. ■ the need to promote the principles of equity, efficiency and economic viability; and Toward the establishment of the WCD and its mandate ■ the importance of balancing the need for development with the requirement of The WCD can trace its roots back to the ensuring environmental sustainability. origins of the dams debate. In addition to the debate, two other related factors painted This broad taxonomy indicates that there the backdrop to its conception. The first is are many areas for potential convergence, the accelerating shift in accepted notions of especially regarding what needs to be done the appropriate relationship between the in the future. All agree on the need: State and its citizens. The second is the ■ to take environmental and social costs of increased recognition of the negative dams more seriously; environmental and social outcomes experienced with large ■ for more systematic consultation with One of the central dams. The result has been affected people; problems – and certainly growing controversy and conse- ■ one of the main causes to ensure that affected people are better off quent delays in development of stalemate in the as a result of the dam development; and decisions. ■ to ensure that the costs and benefits of debate – is the lack of dams are shared more equitably. In Chapter 7, the section on mutual trust between development trends explores the the key parties involved. However, unresolved issues still separate shift in public support away from positions on a number of financial, econom- centralised decisions often taken on the ic, social, and environmental issues. The basis of abstract notions of the greater most intractable include: common good. Instead the public is focusing ■ the extent to which alternatives to dams more sharply on notions of equity in the are viable for achieving various develop- distribution of the costs and benefits of ment goals, and whether alternatives are development and participation in decision- complementary or mutually exclusive; making. ■ the extent to which adverse environ- Growing disaffection by many groups in mental and social impacts are acceptable; society with the way decisions are taken, ■ the degree to which adverse environ- with the distribution of costs and benefits, mental and social impacts can be avoid- and with compliance and recourse mecha- ed or mitigated; and nisms has provoked growing resistance to ■ the degree to which local consent should large dams and escalating tension and govern development decisions in the conflicts around individual dam projects. future. While much of this conflict has centred on the developing world, it nevertheless strikes One of the central problems – and certainly at the dams industry as a whole. Because one of the main causes of stalemate in the many of the most appropriate dams sites in

The Report of the World Commission on Dams 25 Chapter 1

the developed world have already been communities or the environment, as no exploited, the future of the industry lies actions or investments were considered principally in the developing countries, attractive given the ongoing conflict. A new whether the industry itself is based there or way had to be found. in the developed countries. Linked to this are considerations about the image of The need of both dams proponents and international companies, their standing in opponents to negotiate a new, agreed basis the community, their relationship with for assessing options and for planning, shareholders and their reputation for corpo- deciding, implementing and operating them rate social responsibility. – created the conditions for setting up the WCD and giving it a mandate. New approaches are not always well received by developing country governments. Beyond these general considerations, several They often see them as a case of developed specific milestones mark the road leading to countries, having benefited from cutting the establishment of the WCD. These corners themselves, turning to insist that include: developing countries meet ■ The 1992 Morse report.62 This was an higher standards. However, to independent review of the Sardar Sarov- obtain international financing – By the early 1990s, it ar project, commissioned by the World public or private – developing was becoming clear that Bank as a result of growing controversy countries find themselves the cost of controversy over Bank funded projects and criticism having to comply with new could seriously affect of these projects at grassroots level and approaches, norms, and policies future prospects for internationally. Instead of tempering the as a condition of financing or dams and stall efforts to controversy, the Morse report fuelled partnership. finance other non-dam deep criticism of the World Bank’s water and energy internal decision-making. development projects to The net effect of opposition and ■ The Manibeli Declaration, signed in serve rural or urban the controversy over outcomes June 1994 by 326 activist groups and communities. has been to increase the level of risk associated with projects – NGO coalitions from 44 countries, especially those that fail to calling among other things for a morato- recognise the need for a change in the way rium on World Bank funded large dams things are done. This has increased delays until a comprehensive, independent on dam projects in cases where controversial review of all Bank funded dam projects 63 elements are contested in the courts. Both had been conducted. these factors directly or indirectly increase ■ The 1996 report of the World Bank’s the costs of dams. Operations Evaluation Department (OED).64 An internal review of the By the early 1990s, it was becoming clear performance and impacts of a sample of that the cost of controversy could seriously 50 Bank funded large dams. This desk affect future prospects for dams and stall study observed that 90% of dams re- efforts to finance other non-dam water and viewed met the Bank’s standards for energy development projects to serve rural resettlement at the time they were built, or urban communities. The stalemate did but 75% failed to meet the Bank’s most not benefit governments, dam builders, recent standards. Another important

26 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

finding was that proper mitigation of the champions and the critics of large The Gland workshop adverse environmental and social dams in Gland, Switzerland in brought together 39 impacts of most of the dams reviewed April 1997. While the Gland participants representing would have been feasible without com- workshop was focused on bring- governments, the private promising the economic feasibility of the ing a range of opinion around the sector, international projects. The report and the process by table to discuss the implications financial institutions, civil which it was prepared were highly of the World Bank/OED review society organisations, criticised by the NGO community. of 50 Bank funded dams, it found and affected people in a ■ sufficient common ground to set The soul-searching was not confined to balance later mirrored in in motion the process that led to the development finance community. both the WCD and the the formation of the WCD. Professional water and energy develop- Stakeholders Forum. ment associations also began assessing the causes of the growing controversy The Gland workshop brought and reaching conclusions on what together 39 participants representing needed to be done. The International governments, the private sector, interna- Commission On Large Dams (ICOLD) tional financial institutions, civil society published its Position Paper on Dams and organisations, and affected people in a the Environment in 1997.65 Similarly the balance later mirrored in both the WCD International Commission on Irrigation and the Stakeholders Forum. In addition to and Drainage (ICID) initiated a process assessing the OED report, they addressed that resulted in a major statement, The three issues: Role of Dams for Irrigation, Drainage and ■ Critical advances needed in knowledge Flood Control, in 2000.66 and practice in relation to energy and ■ The NGO community was also active in water resources management. gathering case material on experience ■ Methodologies and approaches required with large dams and drawing conclusions to achieve these advances. from it. Building on Goldsmith and Hildyard’s 1984 report, Silenced Rivers by ■ Proposals for a follow-up process involv- Patrick McCully of the International ing all key players. Rivers Network, published in 1996, depicts a particularly bleak record of the The workshop participants identified key social and environmental impacts of issues relating to the social, environmental, dams and their underlying political technical, and financial aspects of dams that dimensions. had to be addressed in reviewing the role of dams and their alternatives in sustainable By 1997, suspicion and mistrust between development. They also formed an Interim proponents and critics of large dams threat- Working Group composed of workshop ened to dominate and undermine wider participants and entrusted this group to discussion needed to reach agreements on establish the WCD. This task proved to be ways to improve access to water and energy long and complex, in part because of the services. In response to this the World Bank decision of the working group to consult all and the World Conservation Union of the key stakeholder groups at each step, (IUCN), a global union of more than 800 and also because of the time needed to build governments, government agencies and confidence in the good faith of all the NGOs – sponsored a meeting between the parties.

The Report of the World Commission on Dams 27 Chapter 1

The WCD was finally announced in Febru- Fulfilling the WCD ary 1998, and began its work the following Mandate: Process and May, under the Chairmanship of Professor Kader Asmal, then South Africa’s Minister Methodology of Water Affairs and Forestry and later To respond to both parts of the mandate it Minister of Education. Its 12 members were was given in Gland, the Commission began chosen through a global search process to by developing an analytical framework and reflect regional diversity, expertise, and work programme to assemble a consolidated, stakeholder perspectives. The Commission shared knowledge base on the worldwide was independent, with each member serving experience with large dams, that: in an individual capacity and none repre- ■ is grounded in the accepted international senting an institution or a country. norms of sustainable and equitable human development; As defined by the Gland workshop, the ■ aims to explore the key themes at the Commission’s two objectives were to: centre of the dams debate, especially ■ Review the development effectiveness of those that are unresolved; and large dams and assess alternatives for ■ compares the planned performance and water resources and energy development. expectations of dams with the actual ■ Develop internationally acceptable experience after project completion. criteria, guidelines and standards where The Commission used both quantitative appropriate, for the planning, design, and qualitative methods to objectively appraisal, construction, operation, evaluate and answer the key questions monitoring and decommissioning of posed. dams. It did not set out to judge decisions on dam The dual objectives are deliberate, reflecting projects from 50 or 100 years ago, but rather to some extent the priorities of the different to learn lessons about the outcomes of dams participants in the dams debate. The cham- and how these lessons could work to change pions of large dams, while wishing to draw or affect outcomes in the future. useful lessons from the review of past experience, tend to lay the emphasis on In order to ensure a solid foundation of practical tools that will help overcome the material on which to base its analysis and controversy and set a foundation for more conclusions, the WCD commissioned, predictable scenarios. The opponents of organised or accepted: large dams, on the other hand, tend to ■ in-depth Case Studies of eight large dams underscore the importance of the review, on four continents, together with two convinced that it will reveal the depth and country review studies; persistence of the negative impacts that ■ a Cross-Check Survey of large dams dams have caused. They want to see evi- located in 52 countries across the globe; dence that dams can be an acceptable ■ option before giving too much attention to 17 Thematic Reviews grouped along five developing guidelines for building better dimensions of the debate; dams in the future. ■ four regional consultations; and

28 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

■ over 900 submissions from interested Developing the knowledge individuals, groups and institutions. base The Commission set out to develop a Recognising the value of the perspectives knowledge base that would give it access to provided by representatives attending the the full range of issues and perspectives Gland meeting, in the first few months of its concerning large dams. The goal was to go work the Commission decided that the beyond the realm of experts and intergovern- group should be retained as a consultative mental processes to include constituencies body for the Commission process, to be with very different entry points into the dams known as the WCD Forum. It was composed debate. The process was designed to offer the of a mix of former Reference Group mem- opportunity for dialogue among the different bers from the Gland meeting, and new interest groups, while providing a solid founda- stakeholders and interest groups. In select- tion for the Commission’s findings. ing the new members of the Forum the WCD was guided by criteria such as relevance, balance and representation of a diversity of perspectives and interest groups. The WCD Process Establishment of Commission Work Programme developed and Secretariat established The Forum offers a consultation model that works at a somewhat different level than other forms of consultation. It has around 70 Activities members and operates as a ‘sounding board’ for the work of the Commission. It is • Thematic Reviews • Case Studies primarily a mechanism for maintaining a dialogue between the Commission and the • Cross-Check Survey • Regional Consultations constituencies of Forum members.

Since the Commission was facilitating debate on the complex issue of the develop- Knowledge Base ment effectiveness of dams, input derived through consultation with these constituen- Synthesis of Work Programme Results cies was essential for the understanding and • Review of Development Effectiveness of Large Dams • Framework for Options Assessment acceptance of the Commission’s final • Criteria and Guidelines products. The Forum was one means of achieving these objectives.

Chapter 10 looks ahead to the dissemina- Final Report tion and the adoption of the Commission’s recommendations. It provides suggestions on the post-Commission role of stakeholder Global Dissemination Process groups such as the Forum. Global dissemination of findings and publications

The Report of the World Commission on Dams 29 Chapter 1

Main elements of the work studies. They helped to shape the global programme review, the element of the work programme used by the Commission to evaluate the key The Commission has based its report on a issues in the overall debate. The Commis- synthesis of information on past experience sion also used the studies to focus more across all elements of the WCD work specifically on the development effective- programme. A brief description of these ness issue from the viewpoints of the stake- main elements follows. Further details on holder group for each case study. the methodology are shown in Annex III.

Case Studies and Country Studies Cross-Check Survey of dams The Commission undertook two case studies The Commission developed the Cross- in OECD countries and six in developing Check Survey to extend the analysis provid- countries. It prepared country ed in the case studies to target a broader set review studies for India and of dams. Completed survey forms were China, and an issue paper for the received for 125 dams in 52 countries. The Russian Federation and Newly 125 dams included the case study dams, Independent States. The case additional dams from the case study basins, studies provide a thorough dams from existing databases and a random understanding of the challenges selection of dams from the larger population and dilemmas associated with to contribute to the overall diversity of the specific dams set in the context sample. The analysis aimed to detect broad- of the development situation in er patterns and trends in performance and specific countries, and the river decision-making relating to dams. basins. They were implemented in stages with stakeholder participation. A variety of dams of different types (storage, run-of-river); ages (the 1930’s through the The Commission followed a standard 1990’s), functions (water supply, irrigation, framework for each case study, based on the power, flood management, recreation and model developed by the Secretariat. At the other); ownership structures (public, corpo- core were six questions: rate and private); and regional locations were included in the survey. Data was ■ What were the projected versus actual verified by an internal review and by sub- benefits, costs and impacts? mitting contested and randomly selected ■ What were the unexpected benefits, data sets for review by constituencies other costs and impacts? than the dams owners and operators. Annex ■ What was the distribution of costs and III sets out the methodology and summary benefits – who gained and who lost? statistics. Figure 1.10 shows the location of ■ How were decisions made? the case study dams and cross-check survey ■ Did the project comply with the criteria dams. and guidelines of the day? Thematic Reviews ■ How would this project be viewed in A total of 17 Thematic Reviews and some today’s context in terms of lessons learned? 130 papers were commissioned to address These questions served a much wider five major areas of concern identified in the purpose than merely orienting the case WCD strategy and objectives paper:

30 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams Figure 1.10: World map showing the regional location of the Case Studies, Country Studies, Cross-Check Survey dams, Regional Consultations, submissions, Forum members, Commissioners and Secretariat

The Report of the World Commission on Dams 31 Chapter 1

■ social and distributional proaches on the topic and to clarify the issues; areas of potential agreement, and persistent disagreement, on highly controversial issues. ■ environmental issues; ■ economic and financial Regional Consultations issues; Over the course of two years, the Commis- ■ options assessment; and sion held four regional consultations, one ■ governance and institutional each in South Asia, Africa and the Middle processes. East, Latin America, and East and South South Asia Consultation, Colombo, December 1998 East Asia. For each of the regional consulta- The Thematic Reviews provided tions, governments and non-government the baseline information, analy- organisations, project-affected communities, sis and recommendations on financial institutions, industry and private issues that cut across the central sector representatives were invited to submit elements in the large dams proposals on all aspects of their experience debate. They consider past and with dams and alternative options, and with current experience, as well as the water and energy resource development. forward-looking context, by Over 30 presentations were made at each synthesising the state-of-the-art consultation and participants engaged in knowledge, practices and key debates on the issues raised. They gave the Latin America Consultation, Sao viewpoints on each topic. Commission a unique and unprecedented Paulo, August 1999 Within limits set by available exposure to the many voices and perspec- resources and the Commission’s tives in the dams debate in a number of schedule, the level of effort countries and regions. involved in preparing these review papers varies according to The Commission also had the benefit of the complexity of the issue and listening to NGO and affected community the level of controversy sur- presentations at two hearings organised by rounding it. Preparation of the NGOs from Europe and from Southern review papers included setting Africa, as well as obtaining a wider range of up panels and procedures for inputs through attending and participating

East and South-East Asia broader peer review. This helped in meetings, workshops and conferences Consultation, Hanoi, February organised by a wide range of organisations 2000 to bring together a wide spectrum of perspectives and ap- and networks.

Africa-Middle-East Consultation, Cairo, December 1999

32 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

Endnotes 16 Raskin et al, 1998. 17 Brown and Halweil, 1999. 1 World Commission on Water in the 21st 18 WCD Thematic Review II.2 Global Change. Century, 2000. Other examples of strategic initiatives in water include the United 19 Schnitter, 1994; McCully, 1996. Nations Comprehensive Assessment of 20 Zhang, 2000, WCD Regional Consultation Freshwater Resources of the World, the Paper. United Nations Food and Agriculture 21 ICOLD, 1998; see also Annex V; Compila- Organisation Aquastat programme, and the tion of estimates of the number of dams in World Water Vision conferences. the main dam building countries suggest 2 Gleick, 1998, p43. there may be as many as 48 000 large dams. 3 UN, 1999, p1; also Raskin et al, 1998. 22 WCD Thematic Review III.2 Financing Medium projection in 2050 was 8.9 billion. Trends. 4 UN, 1998, p2–29. In 30 years the largest 23 WCD Thematic Review IV.2 Irrigation population growth is projected to be in urban Options. Yields from irrigated areas are on conglomerations; the 10 largest population average double that of rainfed agriculture, centres in descending order of size would be and are generally higher on land irrigated by Mumbai (Bombay), Shanghai, Chennai ground water than on land irrigated with (Calcutta), Beijing, Delhi, Karachi, Tianjin, surface water. Metro Manila, Jakarta, and Dhaka. 24 Cosgrove and Rijsberman, 1999, p40. 5 IPCC, 1999. Purchasing power parity (PPP) 25 Ibid. uses accounting exchange rates to compare 26 The WCD India Country Study noted that or add country economic data and avoid the official figures of the Central Water distortions introduced by market exchange Commission indicate that 30% of irrigated rates. In 1995 world Gross Product reached a land is supplied with water by dams; howev- figure of US$33.4 trillion at purchasing er, the study also suggested an alternative power parity (PPP). figure of 10%. 6 Raskin et al., 1998, p9. OECD Data; 78% at 27 Shiklomanov, op cit. market prices and 55% at purchasing power parity. 28 Roo, 2000, pp1–31; McIntosh and Yñignez, 1997, p189. 7 Shiklomanov, 1998 in Gleick 2000, p51–54; Raskin et al, 1995; based on United Nations, 29 IEA, 2000. The UN estimates that 2 billion World Bank, and World Health Organisation people have no access to electricity. Electric- data. ity consumption per capita varies by more 8 Raskin et al, 1995. than a factor of 10 among different regions of the world: from approximately 10 000 kWh/ 9 Gleick, 1998, p44–45. year in North America to less than 1 000 10 Seckler et al, 1998. kWh/year in Africa. 11 Shiklomanov, op cit. 30 IEA, 1998. 12 WRI et al, 1998, p304–305. 31 IFRCRCS, 1998. 13 Shiklomanov, op cit. Municipal and industri- 32 Berz, 2000. al use increased from 17 to 28% of fresh 33 ICOLD, 1998. water withdrawals globally between 1950 and 1990. 34 Lu, 2000. 14 Van Hofwegen and Svendsen, 2000. 35 Revenga et al, 1998. The WRI’s 1998 study of the conditions of the major watersheds in 15 Raskin et al, 1995, p9; Countries are the world focused on 150 watersheds, considered to be water stressed when the representing 55% of the world’s land area. annual renewable water supply drops below 1700 m3 per capita, and water scarce when it 36 Berger et al, 2000. drops below 1000 m3. 37 ICOLD, 1998; WCD Analysis.

The Report of the World Commission on Dams 33 Chapter 1

38 Wolf, 2000, Contributing Paper for WCD others aspects of the projects such as canals, Thematic Review V.3 River Basins. powerhouses, project infrastructure and 39 Wolf et al, 1999. associated compensatory measures such as 40 Lecornu, 1998. bio-reserves etc. (See end notes 7 to 10 in 41 A considerable portion of the aggregate storage chapter 4 for references to this endnote.) of large dams may be for non-consumptive 45 Fox and Brown, 1998b. hydropower releases. Analysis of all large dams 46 World Bank, 1996a, p77. in the ICOLD Register (1998), but excluding single purpose hydropower dams, shows 4 373 47 Jing, 1999, Contributing Paper for WCD km3 of designed storage capacity. This data set Thematic Review 1.3 Displacement, p2. excludes many of the large dams in China, 48 Wang, no date. estimated to provide 451 km3 of storage (IJHD, 49 WCD Thematic Review I.3 Displacement. 1999) 50 ADB, 1999b, p1–2. 42 Keller et al, 2000, p6–7. 51 Fernandes and Paranjpye , 1997, p15–17. 43 Revenga et al, 1998. 52 World Bank, 1996a, p90–92. 44 The World Bank based on the review of 53 WCD Kariba Case Study. resettlement experience estimated that 54 The protestors included both those to be between 1986-1993, an estimated 4 million resettled and farmers who were supposedly people were displaced annually by the 300 beneficiaries of the Bakolori irrigation large dams that entered on an average into scheme. According to the Nigerian govern- construction every year. In the late 1980s ment, 23 protesters were shot dead; unofficial estimates put the death toll at more than 126. some 10.2 million people were officially 55 Stewart et al, 1996; World Bank, 1996b; recognised as "reservoir resettlers" in China. Chen, 1999, WCD Regional Consultation This figure would be substantially higher if it Paper. At the WCD Regional Consultation is officially updated to include new figures of at Sao Paulo one of the survivors narrated population resettlement. For example, dams how his wife and children were shot dead and reservoirs already built on the tributaries before his eyes when he merely enquired of and the main course of the Yangtze River the authorities ‘where do you want us to alone have caused relocation of at least 10 move to?’ The answer came in the form of four rapid bullets. million people. In India the estimates of 56 McCully, 1996, p281-282. people displaced due to large dams vary, from 57 Goodland, 2000. 21 million to 33 million people. Dams account for 34% of all people displaced by 58 Udall, 1998, p392. development projects in China (displace- 59 Gray, 1998, p269-270. ment due to city construction included in 60 Moore and Sklar, 1998, p286, WCD Submis- the total), 77% in India (urban displacement sion eco048. not included to total displaced) and 65% 61 Fox and Brown, 1998a, p489. among the projects funded by the World 62 Morse and Berge, 1992. Bank involving displacement. All these 63 Manibeli Declaration, 1994. figures are at best only careful estimations 64 OED, 1996a and 1996b. and certainly do not include the millions 65 ICOLD, 1997. who may have been displaced due to several 66 ICID, 2000.

34 Dams and Development: A New Framework for Decision-Making Water, Development and Large Dams

Part One: The WCD Global Review of Large Dams

he World Commission on Dams was charged with Treviewing the development effectiveness of large dams and assessing alternatives for water and energy resources management. Part One of the report contains the results of

our Global Review of large dams. It consists of five chapters.

■ Chapter 2 provides the findings of the Commission’s independent review of the technical, financial and economic performance of large dams. ■ Chapter 3 examines the environmental performance of large dams, including ecosystem and climate impacts. ■ Chapter 4 evaluates the social performance of large dams, looking especially at the displacement of people, and the distribution of gains and losses from dams projects. ■ Chapter 5 assesses the scope of various alternatives to large dams for meeting the needs of irrigation, drinking water, electricity, and flood management in terms of both the opportunities they provide and the obstacles they face. ■ Chapter 6 considers the planning, decision-making and institutional arrangements that guided the development of water and energy resources and the selection, design, construction and operation of dams.

The Report of the World Commission on Dams 35 Technical, Financial and Economic Performance

Chapter 2: Technical, Financial and Economic Performance

ny development project – Aparticularly a large infrastructure one such as a large dam – is conceived, planned, and designed to achieve a set of objectives

that will enhance the welfare of society. In assessing the performance of large dams the Commission first

assessed the dams in the WCD Knowledge Base against the targets set by those planning and designing the

facilities. These projections of costs and benefits provided the rationale and basis for project approval and funding.

In most cases, project proponents set firm physical, financial, economic

The Report of the World Commission on Dams 37 Chapter 2

and, increasingly, social and environmental experiences. The information on dams in performance targets. The Commission has the WCD Knowledge base is reported first. also examined past experience with the Results of the Cross-Check Survey are used benefit of hindsight, that is, within the to indicate the direction, variability and context of presently available information. extent of broad trends and patterns. These Here the Knowledge Base is used to assess broad findings are then supplemented by the impacts of large dams that were not experience captured in Case Studies, explicitly targeted and planned for – such as Thematic Review papers, contributions to cost recovery and adverse impacts on the Regional Consultations, and submis- indigenous peoples – but that are important sions. The Case Studies, in particular, are factors in assessing the contribution of large used to provide indicative illustrations of dams to development. This chapter begins these broad findings and their immediate the analysis with a focus on technical, causes. Existing literature and perspectives financial and economic performance. are then cited to verify the original findings, and also to clarify new directions emerging The evidence and findings presented below from the Commission’s work. indicate that there is considerable scope for improving the selection of projects and the The performance indicators for the imple- operation of existing dams, even prior to mentation phase of dam projects are the consideration of the social and environmen- same for dams with different purposes. Thus tal impacts. The performance of large dams this section begins by reporting on the in terms of achieving technical, financial, degree to which dams have met targets for and economic targets is marked by a high capital costs of projects and proceeded degree of variability, with a considerable according to schedule. It then proceeds portion of dams failing to deliver on their sector-by-sector to provide insight into the overall objectives and many falling short of relative performance of dams built and specific targets. Still, a substan- operated for different purposes. Large dams tial number have met or built for irrigation, hydropower, water supply The emerging message exceeded their targets and and flood control have separate objectives, from the WCD continue generating benefits involve different components, respond to Knowledge Base is that after 50 years or more. None- different markets and are operated in project selection, design, theless, the emerging message different ways. The inquiry is driven by and operations could be from the WCD Knowledge consideration of the delivery of benefits improved substantially Base is that project selection, measured in physical terms (such as water, relative to past design, and operations could be power and crops) and then in terms of performance. improved substantially relative financial and economic profitability. Where to past performance. relevant, sectoral performance on costs and time schedules is brought into the analysis. Structure and The extent of cost recovery is also discussed, Methodology as it provides a window not just on financ- The analysis presented here draws on several ing issues but also on the effect of subsidies independent samples of the experience with on the efficient allocation and use of re- large dams, oriented along different axes and sources. In addition, it informs the issues of supplemented by Commissioners’ personal distribution and equity treated in Chapter 4.

38 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

The secondary economic impacts generated often incur substantial capital cost overruns. by large dams (such as multiplier effects) are The average cost overrun of the 81 large also treated in Chapter 4. Of course, many dam projects included in the WCD Cross- dams fulfil not just one but a number of Check Survey was 56%. Variability was purposes; these multi-purpose dams are high. Of the total sample, one-quarter of the examined in an additional section and their dams achieved less than planned capital cost performance is contrasted with that of targets whilst almost three-quarters of the single-purpose dams. Finally, the sustainabil- dams exhibit capital cost overruns (see ity of dam operations is considered by Figure 2.1). reviewing evidence on dam safety, sedimentation, waterlogging and salinity. Within the Cross-Check sample, multi- purpose dams demonstrated high variability Construction Costs and in performance compared with single purpose dams, ranging Schedules from as low as 22% underruns to CostCost performance performance data data During the implementation period of a dam 180% overruns of planned cost inin the the WCD WCD Knowledge Knowledge project – that is during construction – two key targets. Furthermore, the average basebase suggest suggest that that large large performance indicators are the extent to cost overrun was 63% for the 45 damdam projects projects often often incur incur which projects come in on time and on multi-purpose projects – three substantialsubstantial capital capital cost cost budget. Large dams in the WCD Knowledge times that of the single-purpose overruns.overruns. Base have demonstrated a marked tendency hydropower dams in the sample. towards schedule delays and cost overruns. Comparatively, the cost overrun for single-purpose projects was greatest for Capital costs of large dam water supply dams, with all but one project projects showing a 25 to 100% overrun and the average for this category being twice that of Assessment of actual versus predicted performance on capital costs is important for 1 a number of reasons. First, dam projects are Figure 2.1 Cost overruns on large dam projects typically approved on the basis of a financial budget for the investment. If they end up Global sub-sample: 81 dams costing substantially more than expected, 24 additional funds have to be found. As large 22 25% dams and their associated infrastructure may 20 18 20% cost billions of dollars – three of the Case 16 14 15% Study dam complexes cost over $6 billion 12 each – financial overruns have important 10 10% consequences for public and private budget- 8 centage of dams Number of dams 6 5% 4

ing. Second, projects also often derive tariff Per 2 0% charges based on cost estimates, so under- 0 estimates will undermine financial viability underrun overrun or efforts to recover costs. 0-49 50-79 80-99 >=200 100-119 120-149 150-199 Cost performance data in the WCD Knowl- % actual to planned capital costs (US$ Nominal) edge Base suggest that large dam projects Source: WCD Cross-Check Survey.

The Report of the World Commission on Dams 39 Chapter 2

single-purpose irrigation or hydropower ■ poor development of technical and cost dams. Interestingly, performance was worst estimates and supervision by sponsors; in the sub regions of Latin America, Europe, ■ technical problems that arose during Central Asia, and South Asia, with cost construction; overruns averaging 53%, 69%, 108%, and ■ poor implementation by suppliers and 138% respectively. contractors; and ■ changes in external conditions (econom- The WCD Case Studies – viewed as a sub- ic and regulatory).3 sample separate from the Cross-Check Survey – suggest a more marked tendency Part of the difficulty in developing accurate towards cost overruns. Of these, only Stage projections for construction costs of large 1 of the Kariba dam (built in the late 1950s) dams is that the geotechnical conditions at came in on target (a 3% underrun). Note a site (the quality of the rock for the founda- also that the experience is not limited to tions of the major structure and for tunnels), very large billion-dollar projects, as the four and the quality of the construction materials smaller, million-dollar projects in the cannot be determined precisely until con- Glomma and Laagen basin in the WCD struction is under way. Discovery during Case Studies cost some 60 to 185% more construction of less favourable site condi- than projected. The Pak Mun dam in tions than those assumed in the engineering Thailand, a medium-size hydropower dam designs and construction plans can be a financed by the World Bank and commis- significant contributor to cost overruns and sioned in 1994, had a 68% overrun.2 delays in time schedules. Despite being a common factor in causing overruns, little to The causes of cost variations fall into four no provisions have been made to improve categories: the estimates in this regard.

Figure 2.2 Average cost overruns for large Average for all** 54% dams World Bank HEPa 27% World Bank MPb 39% IDBc 45% ADBd 16% AfDBe 2% IRN Listf 247% WCD Case Studiesg 89% WCD Cross-Check*h 52% WCD Cross-Check: India onlyi 235%

0 50 100 150 200 250 300 % capital cost overruns (US$ Nominal)

Sources: aBacon and Besant-Jones, 1998; b OED, 1996a; cIDB, 1999; dLagman, 2000; eAfDB, 1998; fMcCully, 1999 eco061; gWCD Thematic Review III.1 Economics; h, i WCD Cross-Check Survey Notes: *Excludes Case Study and India dams. **Weighted averages based on number of dams in each sample. HEP refers to hydropower projects and MP to multi-purpose projects.

40 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

Additional estimates of cost performance The evidence gathered by WCD strongly gathered or submitted as part of the WCD confirms the view that there is a systematic process include an Inter-American Develop- bias towards underestimation of the capital ment Bank review of its portfolio of large costs of large dams.11 Aggregating all the dam projects from 1960 to 1999. The results aforementioned samples yields an average suggest an average cost overrun of 45% (see cost overrun for all 248 dam projects of just Figure 2.2).4 In a similar exercise, the WCD over 50% (or 40% without the dams from reviewed the portfolio of large dam con- the IRN list and the Indian Cross-Check). struction projects financed by the Asian The implication is that large dams have Development Bank (ADB) between 1968 performed poorly relative to budgetary targets. and 1999.5 Of the 23 completed ADB large dam projects with available data, there was Poor prediction of inflation is often an an average cost overrun of 16%. This masks important component of these overruns. As considerable variability, as the majority of a consequence, when the figures are adjust- large dams in the ADB sample actually had ed for inflation, the overrun in terms of the cost underruns. The International Rivers real economic costs of the materials and Network (IRN) submitted a list of 14 large resources used is likely to be lower than that dams with cost performance data that reported here.12 For the 81 large showed a 242% overrun, with eight projects dams in the Cross-Check sub- in India dominating the results with an sample the cost overrun as Delays in commissioning average overrun of 262%.6 The results from measured in constant 1998 dollar projects affect the these latter dams, as well as the results for terms came to 21%, a significant delivery of services, the Indian dams in the Cross-Check (235% drop from the 56% cost overrun increase interest overrun) confirms the serious overruns obtained in current dollar payments and delay reported in the India Case Study.7 financial terms, but still large revenue generation. enough to significantly affect the The data on cost performance reported by economics of these projects. WCD confirm the results found in other studies. Perhaps the most cited study is that Project implementation of 70 World Bank financed hydropower schedules projects commissioned between 1965 and A second important indicator of perform- 1986, where costs at completion were on ance during the implementation phase of a average 27% higher than estimated at apprais- large dam project is the extent to which al. This, compared to average cost overruns project time schedules are met. Delays in of just 6% for a sample of 64 thermal power the date at which a project is commissioned projects, and an 11% overrun for a sample of lead to increases in interest accumulated on over 2 000 development projects of all funds borrowed for construction activities types.8 An analysis of the data on cost and to delays in revenues accruing to the indicators for 10 multi-purpose dams includ- owner from the completed project.13 For ed in the World Bank Operations and Evaluation Department’s (OED) 1996 report consumers, delays mean additional periods of on large dams yields an average cost overrun of not being served with electricity or water. 39%.9 Finally, a 1998 review of 10 projects by Delays thus affect the delivery of benefits, as the African Development Bank (AfDB) well as the financial and economic perform- found an average cost overrun of only 2%.10 ance of a project.

The Report of the World Commission on Dams 41 Chapter 2

The WCD Knowledge Base suggests a factors that can lead to schedule delays are marked tendency towards schedule delays late delivery to the site of essential equip- for large dam projects compared with the ment, unrealistic construction schedules, planned time to implementation. Of the 99 contractor and construction management projects included in the analysis of project inefficiencies, labour unrest and protests and schedule performance in the Cross-Check legal challenges by affected groups. Survey, only half the projects came in on schedule (see Figure 2.3). Approximately The existing literature on large dams and 30% of the other half were delayed for one related projects confirms this finding: large or two years, and about 15% were delayed dams tend to be subject to significant between three and six years. Four projects schedule slippages. A recent study of World were delayed more than 10 years. Bank financed hydropower projects reports a 28% delay on average. While this is a The WCD Case Study dams also display a considerable slip, it is no different than that range of results in achieving project sched- recorded in the same study for thermal ules. Stage 1 of Kariba dam came in on power projects (30%).14 schedule, whereas Tarbela took two extra years to finish and Aslantas four. Following Irrigation Dams the initiation of construction in the late Large dams and irrigation projects are a 1970s, financing difficulties led to a nine- nested set of sub-systems involving the dams year delay in the case of Tucurui. This led to as source of supply, the irrigation system much higher than expected payments of (including canals and on-farm irrigation interest during construction. Not counting interest the cost overrun was 51% but this application technology), the agricultural rises to 77% once the comparison includes system (including crop production processes), and the wider rural socio-economic system actual and predicted interest costs. Other and agricultural markets.

Figure 2.3 Project schedule performance Potential performance indicators for large dam irrigation projects include: ■ Global sub-sample: 99 dams physical performance on water delivery, area irrigated and cropping intensity; 48 50% 44 ■ cropping patterns and yields, as well as 40 40% the value of production; and 36 32 ■ net financial and economic benefits.15 28 30% 24 Large irrigation dams in the WCD 20 20% centage of dams Number of dams 16

Per Knowledge Base have typically fallen 12 10% 8 short of physical targets, failed to recover 4 their costs, and been less profitable in 0 0% economic terms than expected. The -1 0 1 2 3 4 5 6 secondary benefits of irrigation projects Years behind or ahead (-) of project schedule were rarely specified as targets (these are discussed later, in Chapter 4). Source: WCD Cross-Check Survey.

42 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

Irrigated area and cropping achieved less than 35% of their target intensity irrigation areas during the first five years. Irrigation components of large dam projects The 52 projects in the survey sub-sample in the WCD Knowledge Base fell well short likewise underachieve in terms of cropping of targets in terms of development of command area (and infrastructure), area actually Figure 2.4 Achievement of command area irrigated, and to a lesser extent the intensity with which areas are actually irrigated. With respect to the achievement of command Irrigation sub-sample: 52 dams area targets, the Cross-Check Survey 22 demonstrates a general pattern of under- 20 35% achievement, with almost half of the 52 30% 16 dams in the sub-sample falling short of the 25% planned target (see Figure 2.4).16 Poor 12 20%

performance is most noticeable during the 15% 8 earlier periods of project life, as the average centage of dams Number of dams 10% achievement of irrigated area targets com- 4 Per 5% pared with what was planned for each 0 0% period increased over time from around 70% below target above target in year five to approximately 100% by year 0-49 50-74 75-89 90-99 >150 30 (see Figure 2.5). Nevertheless, a charac- 100-109110-124125-149 teristic pattern observed in the sample is % actual maximum command area achieved to planned target variability of performance between projects.

In particular, one-quarter of the projects Source: WCD Cross-Check Survey.

Figure 2.5 Actual irrigated area compared to planned targets over time

Irrigation sub-sample 200 1 case 2 cases 175 3 cases get 4 cases 150 5 cases 100% Target 6 cases 125 above tar

100

75 get

% Actual to planned 50 Fit line

25 below tar

0 0 10 20 30 40 (n=no. of dams) (n=34) (n=23) (n=13) (n=6) Years from start of commercial operation

Source: WCD Cross-Check Survey.

The Report of the World Commission on Dams 43 Chapter 2

intensity targets. However, cropping intensi- same can be said of the Gariep dam on the ty performance comes closer to targets than Orange River in South Africa. As Tarbela irrigation area development.17 Less than dam is an integral part of the Indus Basin half of the projects in the sub-sample Irrigation System (IBIS), it is not possible to consistently achieve or exceed planned specify the achievement of area targets. cropping intensities from year one of irriga- However, the amount of water released for tion. Of the remainder, 20% achieve less irrigation from Tarbela has exceeded predic- than three-quarters of targets and the other tions by 20% over its 25-year life. This is 40% fall between 75 and 100%. mainly due to a lower-than-expected rate of reservoir sedimentation. Shortfalls of 9 to With regard to cropping intensity there is 60% in achieving cropping intensity targets little difference in the average values for are observed in the provinces of Punjab and single and multi-purpose dams in the survey. Sindh. Meanwhile, the India Case Study Actual values of maximum irrigated area reports that the level of under-utilisation of achieved by single-purpose irrigation irrigated areas is between 13 and 25%. projects are consistently better on average by a margin of around 10%. Single-purpose In the case of the Aslantas dam in Turkey, irrigation schemes also show more of a 96% of the irrigated area was developed by spread in the values around the average, the end of the implementation period, but with one-quarter of the 21 single-purpose this has now fallen to 87% due to the projects achieving more than 115% of conversion of agricultural land to urban and planned targets. industrial purposes. Additional irrigation area foreseen under the original basin plan When compared with larger projects in the has subsequently been implemented through survey, dams with heights less than 30 meters other projects, although delayed from and reservoir areas of less than 10 km2 original projections. Cropping intensity has tended to be closer to the predicted targets increased from 89% to 134% of the target, and demonstrated less variability for com- but such figures mask a return to growing mand area development, actual wheat, a crop that requires little irrigation irrigated area and actual crop in Turkey. When compared with intensity. All projects below larger projects in the The under-achievement of targets for 90% of area and intensity survey, dams with irrigated area development from large dams targets were larger than 10 km2 heights less than 30 has a number of causes. Institutional failures and higher than 30 meters. meters and reservoir have often been the primary causes, includ- areas of less than ing inadequate distribution channels, over- The WCD Case Studies corrob- 10 km2 tended to be centralised systems of canal administration, orate the results of the Cross- closer to the predicted divided institutional responsibility for main Check Survey and display a targets system and tertiary level systems, and similar breadth of results with inadequate allocation of financing for respect to irrigation area and cropping tertiary canal development. Technical intensity targets. In the case of Grand causes include delays in construction, Coulee, commissioned in 1941, only about inadequate surveys and hydrological assump- half of the predicted area in the Columbia tions, inadequate attention to drainage, and Basin Project was eventually developed; the

44 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

over-optimistic projections of cropping crops. This implies patterns, yields and irrigation efficiencies, higher-than-expected including the late realisation that some gross value of produc- areas were not economically viable. Also, a tion per unit of area, mismatch between the static assumptions of with the caution that the planning agency and the dynamic such increases have nature of the incentives that govern actual varied with the long- farmer behaviour has meant that projections term real price trend quickly became outdated. of the relevant agricultural commodi- Performance data on irrigation projects as ties. But when chang- reported elsewhere support the original es in cropping pat- findings of the Commission, although these terns are combined with shortfalls in area are not solely linked to large dams. An ADB developed and cropping intensity, the end- evaluation of 35 irrigation projects found result is often a shortfall in agricultural actual cropped areas generally at 60–85% of production from the scheme as a whole. appraisal estimates, with only four exceeding Gross value of production is higher where 18 targets. A World Bank study of seven the shift to higher-value crops offsets the irrigation projects found all but one with shortfall in area or intensity targets. crop intensities less than expected (in the 19 range of 65–91%, with one at 107%). A The WCD Case Studies provide examples of 1990 evaluation by the World Bank of 21 both underestimation and overestimation of irrigation projects 5–12 years after comple- the quantity and value of agricultural tion showed that irrigated area had fallen in production. In the case of the Columbia 11 of the projects and that cropping intensi- Basin Project, yields have been 30–50% ty was lower than at completion in 18 of the higher than predicted in planning reports in projects (at 85%).20 1932. Partly because of a shift to higher- value crops, the gross monetary value of Cropping patterns, yields, actual agricultural production per unit of agricultural production, and area almost doubled from that anticipated at gross value of production the time. For example, the area planted to fruit and vegetables in 1992 was 60% Crop yields and the gross value of produc- compared to 20% thirty years earlier. Yields tion from large irrigation dams in the WCD of fodder and cereal crops also increased Knowledge Base have often varied signifi- significantly due to improved varieties and cantly from those predicted at the outset of mechanisation. the projects. Lower yields are often observed for crops specified in planning documents – Similarly, in the case of the Aslantas dam in which emphasise food grain production for Turkey, the cropping pattern changed growing populations – than for the crops dramatically from that planned – partly actually selected by farmers. This occurs as retaining the staple wheat crop and growing farmers respond to the market incentives second high-value crops such as soybean, offered by higher-value crops – either rather than cotton as predicted. Yields of seasonal or longer-term orchard-based crops wheat and cotton reached 75% of predic- – and allocate available resources to these tions, while watermelon and maize exceeded

The Report of the World Commission on Dams 45 Chapter 2

targets by 50–100%. Overall, the gross subsidies and other incentives used by many monetary value of agricultural production countries to support agriculture. This fall in reached 71% of that predicted. In the case prices has contributed to lowering the value of of Tarbela, yields of wheat, rice and cotton production achieved as against predictions. are between 9% and 50% lower than predicted in the feasibility report. A general pattern of shortfalls and variability in agricultural production from irrigation Lower than expected crop yields have been projects in developing countries is also caused by agronomic factors, including revealed by other sources. In the 1990 cultivation practices, poor seed quality, pest World Bank OED study on irrigation cited attack and adverse weather conditions, and earlier, 15 of 21 projects had lower than by lack of labour or financial resources. planned agricultural production at comple- Physical factors such as tion. Evaluations of 192 irrigation projects poor drainage, uneven or approved between 1961 and 1984 by the Box 2.1 Efficiency in the use of irrigation water unsuitable land, ineffi- World Bank indicated that only 67% per- cient and unreliable formed satisfactorily against their targets.23 Surface irrigation systems of the type supported by large dam projects have irrigation application, tended to use water inefficiently. Surface and salinity also hinder water irrigation efficiency is in the range of Financial and economic 25-40% in India, Mexico, Pakistan, the agricultural production profitability Philippines and Thailand; 40-45% in (see below). The effi- Malaysia and Morocco; and 50-60% in Since the 1930s in the industrial countries ciency of water use Israel, Japan and Taiwan. The Aslantas and from the 1970s in developing countries, Case Study suggests that the overall affects not only produc- irrigation efficiency of the Aslantas project financial and economic profitability have is 40%. These examples suggest that the tion but also demand become an important, if not the dominant, ratio of water consumed by crops to the and supply of irrigated water delivered from the source varies decision criteria in water projects.24 Conse- from 25-60%. water (see Box 2.1). quently, approval of many large dam Sources: WCD Thematic Review IV.2 Irrigation Options, Chapter 3; WCD projects was contingent upon estimates of An important cause of Aslantas Case Study their predicted profitability. The measures the difficulties in achiev- typically used to assess profitability are the ing targets for gross value financial internal rate of return (FIRR) and of production is the economic internal rate of return (EIRR) as decline in commodity prices. For example, determined through cost-benefit analyses. in real terms, world prices for grains in the The FIRR tells the project owner if the 1990s were of the order of half those in the project is profitable, while the EIRR is 1950s. Although there were significant intended to tell society if the project im- annual variations, the average price of rice proves the overall economic welfare (or from 1950 to 1981 was $850/ton (in 1997 well-being) of the nation. Under-perform- prices), compared with $350/ton from 1985 ance relative to targets does not necessarily to date.21 Wheat prices have shown a imply that a project is unprofitable in similar decline, but less severe ($330/ton economic terms, as the rate of return of a from 1950 to 1981 in 1998 prices, compared project may fall short of its target but still with $140 from 1985 to 1999).22 The exceed the opportunity cost of capital to the observed fall in prices stems in part from the economy. increase in food production that was spurred by irrigated agriculture and the Green Typically, an EIRR of over 10% is judged Revolution, but also from production acceptable in the context of a developing

46 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

economy.25 On this basis, irrigation dam Figure 2.6 Economic performance of multilateral-financed irrigation projects in the WCD Knowledge Base have dams all too often failed to deliver on promised financial and economic profitability – even 14 when defined narrowly in terms of direct 12 project costs and benefits. 10 Economic ojects Given the lack of evaluation studies on 8 internal rate of return large dams for irrigation purposes, the WCD 6 12%+ compiled financial and economic perform- 4 <12-10% ance data from a series of project appraisal, Number of pr 2 <10-5% completion, and audit reports on large dam <5% 0 projects funded by the World Bank and At appraisal At evaluation 26 ADB (see Figure 2.6). The average EIRR Note: Total number of projects reviewed is fourteen at appraisal for the 14 irrigation dams was slightly above 15%, and at evaluation it was Source: WCD Thematic Review III.1 Economic Analysis, Chapter 10. 10.5%, a significant shortfall in economic performance for the group. Whereas 12 Box 2.2 Economic and financial performance of the Columbia Basin projects had expected returns of over 12% at Project appraisal, this number had fallen to five by The Columbia Basin Project (CBP) was never expected to cover its costs, and evaluation. In four cases, the EIRR at evalua- was criticised by early opponents because it would not be economically profitable. Yet the WCD Case Study reports that a $2 150/ha increase in tion fell below the cut-off rate of 10%. assessed land values has occurred for irrigated land within the project area. When applied to the 268 000 hectares under cultivation, the capitalised increase in the value of the land would be $575 million in 1998 dollars. Even if The results extracted from the ADB and the full value of this increase was due to the CBP alone, it is clear that it would World Bank reports are of course based only not come close to the real cost of the CBP of $3.6 billion, as reported in the Case Study. on evaluation studies undertaken at comple- A better approach to assessing the project is to compare the present value of tion of the implementation phase or just a costs of the project with the benefits. The Case Study calculated the present few years after commissioning. They incor- value of CBP costs at $1.47 billion at a 10% discount rate.27 Two simple assumptions may be used to generate a best-case estimate of benefits. First, porate the effects of cost overruns and initial the full value of the average net production value of $500/ha per year as operating results, but are not long-term or reported by a recent study of the CBP is attributed to irrigation (that is as a return to irrigated water and not other farming or capital inputs). Second, these comprehensive in nature. They typically benefits are considered to occur for all years since 1945 (project start-up) and to only consider the direct project costs and all of the 268 000 hectares. The resulting net present value of benefits from 1945 to 2010 at 10% discount rate is $1.32 billion. Note that the assumptions benefits of the project and do not account made are generous, given that actual acreage and gross value of production for the social and environmental impacts increased only gradually over time, and thus the actual magnitude of early economic benefits would have been considerably less. In addition, the associated with the dam or agricultural opportunity cost of water diverted to the CBP of $39 million per year is left out production. In the case of the Columbia of the calculation. Basin Project, even a cursory analysis of the As the benefits are less than the costs, the results suggest that the CBP did not achieve a 10% rate of return (when measured on an incremental basis, that is long-term performance data available from apart from the Grand Coulee dam (GCD)). While this simplistic analysis the WCD Case Study which shows that the provides just an indication of possible returns from the project, it illustrates the type of information provided by ex-post evaluation of economic profitability. large cost overruns and lower-than-expected Further, it demonstrates the concerns raised about multi-purpose projects, delivery of benefits raise questions about the where the irrigation component fails a cost-benefit test. Would the United States economy have been better off investing only in GCD – and using the economics of the project (see Box 2.2). revenues from GCD to invest in other profitable opportunities in the economy – rather than reinvesting the profits in the CBP?28 In post-evaluation studies of irrigation and Source: calculations based on WCD Grand Coulee Case Study rural development projects by the World

The Report of the World Commission on Dams 47 Chapter 2

Bank and the ADB, half the projects were provided by large dams are valued as con- judged as unprofitable in economic terms, sumption goods or productive inputs, the that is having EIRRs of less than 10%. More absence of cost recovery by the sponsoring than three-quarters of the projects returned agency is often equivalent to a subsidy in less than expected at appraisal. In fact, the the sense that the large dam project pro- average of the EIRRs for the World Bank vides a benefit for which no fees are paid. irrigation projects was 17.7% at appraisal. Which participants in large dam projects At completion, this dropped to 14.8%, and receive these subsidies is examined further at the time of impact evaluation – typically in Chapter 4. A lack of cost recovery is not six to eight years after completion – the just a matter of subsidy, however. Provision EIRRs had fallen to 9.3%. This rather strong of free services and subsidised inputs often downward trend indicates that many irriga- leads to misallocation of resources and tion projects – not just large dams projects – inefficient production. Further, it may lead suffer from a tendency to overstate project to perverse behaviour as people direct their returns at the outset, also referred to as efforts to obtaining such subsidies (rent- 29 ‘appraisal optimism.’ seeking behaviour) rather than productive activities. Cost recovery Public agencies have not always attempted The analysis here assesses the extent to to recover costs associated with public which cost recovery is an explicit objective infrastructure projects. Where the services in large dam irrigation projects, and the extent to which it has met expectations or, if cost recovery was not anticipated, the Box 2.3 Cost recovery for the Aslantas Dam extent to which it has occurred in any event. Not surprisingly, recovery of capital In negotiations leading up to the Aslantas dam project, the World Bank was assured that arrangements would be made for recovering O&M costs for the costs for irrigation has rarely been a target dam and irrigation systems, capital costs for the dam and irrigation system and is even more rarely achieved. Perform- (over 50 years), and on-farm development costs (over 15–20 years). An annual recovery charge of TL 1 250 per hectare to cover the first two items was ance in recovering operational and mainte- proposed in the 1972 appraisal by the World Bank, and assurance was received nance (O&M) costs is typically poor, that this would be reviewed every five years, when the rates would be adjusted for inflation. This comes to roughly $350 per hectare in 1998 dollars and would although increasing recognition of the have represented half of the expected farming profits per hectare as estimated importance of recovery to performance led at appraisal. Even under such a repayment program, the Bank estimated that farmers would pay back only about one-quarter of the total capital and O&M to institutional innovations that increased costs. collection of O&M charges in the 1990s. Actual recovery of capital costs appears to have fared quite poorly. The capital cost recovery charge rate was increased in 1981, 1982 and 1986. Due to rapid local inflation and subsequent devaluation of the Turkish lira, however, the TL Recovery of operational and 7 500 charge set in 1986 came to just $0.03 per hectare in 1998. Collection maintenance costs rates have been a continual problem. In 1984, the collection rates on assessed charges for capital costs were only 63% at which time penalties were intro- There was mixed performance on the duced. More recently, the WCD Case Study suggests that the transfer of O&M recovery of O&M costs in the three WCD responsibilities to the local water user association in 1995 resulted in greatly increased recovery of annual O&M costs. Indirect cost recovery for the sector Case Study dams that involved irrigation. In as a whole in the form of a withholding tax of 5% on agricultural sales was the case of the Aslantas dam, recovery of instituted country-wide in 1981, yielding on average $93 of revenue per hectare of agricultural holding. Still, the project has a very low rate of capital cost costs was expected but only made progress recovery and has not come close to the 25% recovery agreed at appraisal. after 1995 with the adoption of a water user Source: WCD Aslantas Case Study; World Bank, 1973; World Bank, 1985. association (see Box 2.3). In the Indus Basin Irrigation System, where the Tarbela Dam

48 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

plays a pivotal role, revenues from irrigation Recovery of Capital Costs fees basically covered operation and mainte- The tendency to poor finan- nance costs up until 1970. Subsequently, cial and economic perform- revenues have declined, and the gap be- ance and the failure to recover tween O&M expenditure and recoveries O&M costs suggest that even reached 44% by 1992 in Punjab and 30% in where it is an explicit objec- Sindh. At the same time, an increasing tive, recovery of capital costs proportion of the revenues was being will be limited. The Aslantas allocated to agency costs as opposed to dam provides a fairly stark maintenance activities. example of the failure not only to recover these costs but also to stick to agreements A similar situation occurs in India, where made in this regard (see Box 2.3). In the the gross receipts from irrigation charges are World Bank’s 1990 evaluation report on considerably less than the recurrent costs of irrigation projects even the cases of ‘excellent’ O&M. In the 1960s receipts covered cost recovery resulted in only partial recovery expenditures, however, by the end of the of capital costs. 1980s receipts were only of the order of 10% of expenditures.30 Annual operational In sum, the assessment of large dam irrigation losses became a huge fiscal liability, with schemes reveals that while there is considera- annual operational losses exceeding $1 ble variability in performance, such schemes have all too often fallen billion by the mid-1990s. The water charges short of physical targets and failed The bulk of projects have collected represent about 2% of the incre- to recover their costs in cases where delivered power within a mental benefits of irrigation.31 In the that was the intention indicated in close range of pre- Columbia Basin Project, irrigators pay only the project document. Further, in project targets but with a very small portion of the costs of pumping many cases the economic justifica- an overall tendency to water into the CBP system and nothing for tion for the approval of the project fall short of targets. the water itself – which has a large opportu- has not been borne out by actual nity cost in terms of foregone hydropower experience in implementation and production at Grand Coulee. operation due to cost overruns and shortfalls in net benefits of agricultural production. The information gathered by the WCD on O&M cost recovery is confirmed by the literature in this field. A survey done since Hydropower Dams 1992 of 18 irrigation systems worldwide Large hydropower dams in the WCD shows considerable variation of recovery Knowledge Base can be divided into three rates, with public agency systems in the groups. The first two groups represent the range of 30–50% and some locally managed extremes: a number of projects that have systems reaching full cost recovery.32 The greatly exceeded their targets and a few 1990 evaluation by the World Bank reports notable under-performers. The bulk of that in 11 of 21 cases, recovery rates were projects have delivered power within a close too low to cover irrigation O&M costs. In range of pre-project targets but with an Mexico, water user associations have proved overall tendency to fall short of targets. effective in improving cost recovery and Hydropower projects – as with other large management.33 dams – have incurred cost overruns and

The Report of the World Commission on Dams 49 Chapter 2

schedule delays. Limited available evidence In contrast to irrigation, the hydropower suggests that hydropower projects often performance of 63 large dams in the WCD diverge substantially from their economic Cross-Check Survey was on average closer targets, in either a positive or negative to target (see Figure 2.7). But as with direction. Financial performance is more irrigation dams, the variance in performance consistent and with less downside variability. across the projects was extremely large (see Finally, a number of older projects continue Figure 2.8). On average, almost half of the generating benefits even after a half-century sample exceeded the set targets for power or more of operation. generation – with about 15% exceeding targets by a significant amount. It also shows Delivery of services and that around one-fifth of the projects in the benefits sample achieve less than 75% of the planned power targets. Aside from these marked Large dams in the WCD Knowledge Base extremes, Figure 2.7 shows that over half of that were designed to deliver electric power the projects in the sample fall short of their have on average met expectations for the power production targets. Thus average delivery of power but with considerable performance in the sample is sustained by a variability, much of it on the downside. A few over-performers and should not mask number of projects have far exceeded their the variance in performance that is weight- technical, financial and economic targets, ed towards shortfalls in power delivery. whereas others have fallen well short. Delivery of services and benefits are exam- The higher-than-expected output in hydro- ined by assessing performance to targets for power generation from almost half of the installed capacity and delivery of power. large dam projects in the Cross-Check Hydropower also offers ancillary services to Survey is due only in small part to the the power grid. addition of extra installed capacity prior to commissioning, but more especially since Figure 2.7 Project averages for actual versus planned hydropower generation commissioning. One-quarter of the large dams with higher-than-expected output had installed more than 100% of the capacity Hydropower sub-sample: 63 dams planned in the feasibility study. 20 30% 18 The Tucurui dam diverged from feasibility 16 25% 14 design when initial installed capacity was 12 20% raised from 2 700 to 4 000 MW before 10 15% commissioning. Tarbela, Grand Coulee, and 8 the Glomma and Laagen dams have all seen 6 10% centage of dams Number of dams

Per subsequent installation of significant 4 5% 2 amounts of additional capacity that were 0 0% below target above target not foreseen at feasibility. Furthermore, both Kariba and Tucurui are multi-stage projects 0-49 50-74 75-89 90-99 >150 that involve doubling of capacity. The 100-109110-124125-149 Averages of actual to planned percentages per project profiles of hydropower performance of the Case Study dams in Figure 2.9 illustrate how Source: WCD Cross-Check Survey. installation of more than initially expected

50 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

capacity leads to higher-than-expected than expected in the dry season, leading to power output (Grand Coulee and Tarbela). questions regarding its economic viability (as discussed below). However, energy output is often also lower than initially estimated. The Victoria dam Further analysis of the Cross-Check data in Sri Lanka had a predicted energy genera- reveals that the average generation in the tion of 970 GWh/annum, but in reality only first year of commercial operation is 80% of produces an average of 670 GWh, a shortfall the targeted value for large hydroelectric of over 30%.34 Higher-than-expected dams (see Figure 2.8). In years two to five, upstream irrigation abstractions and lower- the average percentage realisation of targets than-predicted natural stream flows were rose to near 100%, however this improve- the causes in this case. Case Study results ment in the average for any time period from Pak Mun, which is a run-of-river masks considerable variation in the sub- project with peaking capacity, showed that sample with half or more of projects still in the first four years of operation after 1994 falling short of predicted power generation. the installed capacity and total annual Delays in the construction phase of projects energy generation were as expected. Howev- (as documented earlier), in reservoir filling er, the ability of the project to deliver (if low rainfall prevails), and in installing energy for its primary purpose of a planned and bringing turbines on-line explain four-hour peak period was considerably less shortfalls in performance in the early years

Figure 2.8 Actual versus planned hydropower generation over time

Hydropower sub-sample 300 1 case 2 cases 275 3 cases 250 4 cases 5 cases 225

200

175 get 100% Target 150

125 above tar

% actual to planned 100

75 get 50 Fit line

25 below tar

0 1 5 10 15 20 25 30 (n=no. of dams) (n=55) (n=51) (n=41) (n=33) (n=24) (n=16) Years from start of commercial operation

Source: WCD Cross-Check Survey.

The Report of the World Commission on Dams 51 Chapter 2

of commercial operation. Delays in install- and exceeded predicted levels in 1992, ing capacity and subsequent delays in which coincided with the commissioning of meeting power targets are noted in two case the full complement of planned capacity studies (Tarbela and Kariba). Tarbela met (see Figure 2.9).

Figure 2.9 WCD case study hydropower performance: capacity and power generation

Grand Coulee Grand Coulee Predicted versus actual installed capacity Predicted versus actual generation GW TWh 8 30 7 25 6 5 20 Actual 4 15 3 Actual 10 2 Predicted 1 Predicted 5 0 0 1940 1960 1980 2000 1940 1960 1980 2000 Year Year

Kariba Kariba Predicted versus actual installed capacity Predicted versus actual generation MW TWh 1500 Actual 10

1250 8 Predicted Predicted 1000 6 750 Actual 4 500

250 2

0 0 1960 1970 1980 1990 2000 1960 1970 1980 1990 2000 Year Year

Tarbela Tarbela Predicted versus actual installed capacity Predicted versus actual generation GW TWh 5 20

4 15 Predicted Actual 3 Predicted 10 2 5 Actual 1

0 0 1975 1980 1985 1990 1995 2000 1975 1980 1985 1990 1995 2000 Year Year

Source: WCD Grand Coulee, Kariba and Tarbela Case Studies.

52 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

The WCD Case Studies show that unex- to be seen. Such variation may Delays in reaching pected events and design changes during the also reflect changes in other targets may have project development stage may lead to demand and supply factors. important consequences delays in achieving power generation for providing power to targets. For example, Tarbela experienced Technical efficiency consumers and securing major structural damage in commissioning and ancillary services early economic benefits trials that led to a two-year loss of power The Glomma and Laagen and from a project. generation, and in South Africa the deci- Grand Coulee Case Studies sion to increase power generation by in- illustrate how unplanned creasing the height of the Van der Kloof improvements in generation were achieved dam delayed its commissioning. Where by a combination of factors such as adding delay is not the result of slower than expect- new powerhouses to the same reservoir, ed growth in demand for power, such delays adding additional turbines, upgrading in reaching targets may have important existing turbine and generation equipment, consequences for providing power to con- or optimising reservoir operations to im- sumers and securing early economic benefits prove performance. from a project. These experiences of improving the per- A further observation on timing of power formance of hydropower generation over a delivery from the Cross-Check Survey is the project’s life are not confined to industrial consistency in the distribution of the sample countries. In Nepal, modifications to the over the period surveyed (see Figure 2.8). intake, provision of an extra desander, The WCD Case Studies provide a series of dredging the forebay and refurbishing the examples (Grand Coulee, Tarbela, and generators/turbines and power house control Tucurui) in which power output from very systems at the Trushuli-Devighat hydropow- large dams increases over long periods (in er station in 1995 improved average annual absolute terms and relative to predictions). power generation by 46% – from 194 to 284 Power production at Grand Coulee, for GWh a year.35 In other cases the optimised example, has trended upward for the last 60 operation of reservoirs has led to increased years – though marked by significant inter- generation during a project’s life.36 The annual variability (see Figure 2.9). trend in the industry in Europe and North Ameri- Box 2.4 Optimising operations with This variation in power production over ca is to optimise reservoir the aid of a computerised decision- support system time within a single project is noted in the operations and power WCD Cross-Check Survey and additional dispatch schedules to Since 1987, two hydroelectric plants in the Case Studies. Normal variations in weather coastal mountains of British Columbia have improve performance by used a computerised decision support and river flows dictate that virtually all using more sophisticated system (DSS) to guide weekly reservoir release decisions. Studies of 1970–74 hydroelectric projects will have year-to-year ‘decision-support’ systems operations (before the support system fluctuations in output. The effect of drought became operational) showed that the rule- (see Box 2.4). curve-based operation had produced 83% years can be easily seen in the large swings of the maximum attainable energy. With in annual power generation from Grand The ancillary services the DSS, the actual energy produced each year between 1989 and 1993 rose to 100, Coulee and Kariba, particularly over the last associated with hydropow- 93, 98, 94, and 96% respectively of the two decades. Whether changes in regional er generation – for exam- maximum possible. and global climate are exacerbating normal Source: WCD Thematic Review ple, reactive power genera- IV.5 Operations weather-related interannual variation remains tion and fast starting

The Report of the World Commission on Dams 53 Chapter 2

reserve generation – reduce or even defer there is a wide spread in terms of actual other investments in the electrical system. financial rates of return.

Financial and economic Evidence compiled by WCD from multilat- profitability eral bank appraisals and evaluation studies shows that although a number of hydropow- Schedule delays, cost overruns and variabili- er projects fall short of their financial and ty in delivery of power suggest a broad economic targets and can be considered variation in economic performance for economically unprofitable, others meet their hydropower projects. Further, evidence from targets or even exceed expected profitability North America suggests that the O&M (see Figure 2.10). Of 20 World Bank, AfDB, costs of hydropower rise over time.37 and ADB projects reviewed, 11 fell below Large hydropower dams in the WCD initial targets and seven rose above these Knowledge Base confirm that there is targets; overall, nine projects had returns considerable variability in meeting econom- below 10% but only six of the projects ic targets and achieving economic profita- actually fell to this level (the other three bility (when defined narrowly in terms of already had low rates of return at apprais- project costs and direct benefits). Unlike al). These evaluation studies do not large dam irrigation projects the variability reflect long-term performance data, but is not only to the down-side, with some rather the effect of cost overruns and projects performing better than expected. initial lags in performance. Much less variation is observed in financial The nine evaluation studies that included performance relative to targets, although data on financial performance (all from the World Bank and ADB) showed much less Figure 2.10 Multilateral bank evaluation results on the economic downward variability, with only one project performance for hydropower dams falling short of target by a significant amount (namely from 11% down to 6%). Three projects improved their performance 16 by 5% over expected returns. The results

14 may reflect the administrative nature of tariff setting, which enables tariffs to be 12 adjusted to suit the financial requirements 10 ojects of a project.

8 Economic internal rate of Long-term data gathered by the Commis- 6 return sion through the Case Studies illustrates the Number of pr 12%+ 4 range of profitability actually achieved by <12-10% 2 <10-5% hydropower projects. Indeed, three of the <5% Case Studies provide a succinct illustration 0 At appraisal At evaluation of projects that are superlative, respectable,

Note: Total number of projects reviewed is twenty or borderline in terms of profitability. A fourth demonstrates how a project may be approved, but in the end fall short in terms Source: WCD Thematic Review III.1 Economic Analysis, Chapter 10. of profitability.

54 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

The financial and economic performance of power plant capacity used in Few formal and Grand Coulee as a stand-alone hydropower the appraisal of the project was comprehensive post- project can only be described as superlative too generous. The Case Study evaluations of the (see Box 2.5). Further, after 60 years of goes on to value the benefits of financial and economic operation it shows no sign of slowing down. Pak Mun based on the avoided profitability of large In the case of the Kariba dam, the Case costs of the alternative. The hydropower dams exist Study reports on a cost-benefit calculation large reduction in the alterna- for comparison with the that took the original economic assumptions tive power plant from 150 MW WCD analysis. and developed both predicted and actual gas turbine to a 21MW gas EIRRs. Based on 40 years of operating data, turbine when combined with and including the cost overruns under the the 68% cost overrun reduces the EIRR second stage of the project, the actual EIRR from 12.1% to 7.9%, which is below the is a respectable 14.5%, down from the opportunity cost of capital in Thailand. It is 16.5% calculated for the project as planned. worth noting that this calculation simply demonstrates that the energy supplied by In the case of Tucurui dam, the project was Pak Mun could have been supplied in a planned under a military dictatorship and more inexpensive fashion, for example by there was little concern during planning for using the 21 MW gas turbine and secondary economic profitability or cost recovery. The power from other plants. WCD Case Study suggests that the project is marginally profitable relative to the Few formal and comprehensive post-evalua- consumer price of electricity, but this omits tions of the financial and economic profita- the costs of transmission and distribution bility of large hydropower dams exist for (see Box 2.6). Perhaps the simplest observa- comparison with the WCD analysis. Despite tion is that had the project met its original cost target it would have produced, in financial terms, low-cost power. However, Box 2.5 Financial and economic performance of hydropower at Grand Coulee dam the large cost overrun – $3.3 billion more than expected – effectively erased the Financial profitability. Profitability of the Grand Coulee dam (GCD) was not an explicit objective. Still, even at a real tariff of $0.02/kWh the GCD will have competitive advantage that the site may generated real revenues of around $15 billion over the last 50 years. These can have had. Further, the relatively high unit be compared with real project costs of $5.7 billion. While the inclusion of operating costs and discounting would reduce this gap, it seems that GCD is cost of electricity from the facility and the quite profitable in financial terms. subsidies to industry established at commis- Economic profitability. No cost-benefit analysis was undertaken in the 1932 sioning imply a failure to recover costs Butler Report. Prior to installation of the third power plant, an analysis by the US Bureau of Reclamation, yielded a benefit-cost ratio of over 3:1. When the cost directly, as the operating utility continues to overruns on the third plant are accounted for in the calculation, the benefit-cost require state subsidies. ratio drops – but at 2:1 it still indicates that the plant would remain a very profitable undertaking. Economic efficiency. The current operating costs of GCD compare favourably In the case of Pak Mun, installed hydropow- (more than 17 times less expensive) with next best current alternatives. This is er capacity and generation have met actual best stated in terms of today’s decision to continue producing power from GCD. Compared with natural gas at $25 MWh, current production levels of 20 000– targets for the few years of operations to this 25 000 GWhs are responsible for saving $475–600 million a year in real costs to point. However, the WCD Case Study the economy. suggests that the failure to produce a firm Of course, these figures reflect only the effects of direct project costs and benefits, not the external social and environmental impacts of the project. supply of peaking power during low flow Source: WCD Grand Coulee Case Study months implies that the alternative thermal

The Report of the World Commission on Dams 55 Chapter 2

Box 2.6 Economic performance and cost recovery of hydropower at Water Supply Dams Tucurui dam Water supply dams in the WCD Knowledge The Tucurui dam is a single-purpose hydropower facility developed and Base have generally fallen short of intended operated by Eletronorte, the public utility for electricity in the north of Brazil. The WCD Case Study calculates generating costs at $40–58/MWh (for 8% and timing and targets for bulk water delivery 12% discount rates). Of the total power produced in 1998 by Tucurui dam, about and have exhibited poor financial cost half (12 000 GWh) went to industry at a price of $24/MWh – from $16 to $34/ MWh below cost. Using these below-cost sales figures as an indication of the recovery and economic performance. These subsidy provided to industry yields a range from $190 million per year (for the results reflect the longer development cost at 8%) to over $400 million (at 12%). Eletronorte calculates its annual subsidy from the public purse in 1998 at $194 million, and Tucurui is the largest of horizon of such dams, as well as over- Eletronorte’s projects. Failing more precise data it appears that Tucurui could estimates of demand, and are similar to the recover its costs but that to date it may not have achieved this, in part due to continued subsidies to industrial producers. Indirect cost recovery through general direction of results in the water taxation of these industries was not documented in the study. supply and sanitation sector. Source: WCD Tucurui Case Study Delivery of bulk water supply The Cross-Check Survey found that one- quarter of the 29 dams with a water supply being probably the single largest financier of function have delivered less than 50% of dam projects in the post-war period, the target. Furthermore, on average 70% of the World Bank did not undertake a dam- sample did not reach their targets over time specific review of its portfolio until the mid- for the delivery of bulk water supply (see 1990s. Even then its 1996 OED study did Figures 2.11 and 2.12). Much of the overall not draw on actual performance of the dams sample variability is due to the multi- in its sample and provides little evidence on purpose, single-purpose distinction in economic performance.38 The African conjunction with reservoir size. The data Development Bank recently reviewed its suggests that all delivery of bulk water in experience with six hydroelectric dams and excess of planned targets can be ascribed to found that only four passed the economic multi-purpose dams. This is contrary to viability test using a 10% discount rate.39 performance trends for other purposes where single purpose dams generally come out Thus the WCD Knowledge Base shows that better. Regarding reservoir size, it is clear a considerable number of hydropower from the survey that the smaller the reser- projects fall short of their initial economic voir area, the closer to target results have targets, although only a smaller number can been – with the exception of the 11 reser- be classified as economically unprofitable voirs larger than 100 square kilometres. (falling short of the rate of return target for These very large reservoirs exhibit extreme the economy as a whole). Meanwhile there variability, ranging from under-performance are almost as many projects in the Knowl- to considerable over-performance and edge Base that actually outperform their delivering maximum achieved bulk water economic targets. Finally, it is worth empha- supplies up to 2.5 times planned targets. sising that cost recovery has not been a substantial problem for hydropower projects; The principal message with regard to water indeed, the focus is more on profitability in supply emerging from the Case Studies is the current context of trends towards that even when it is not planned, demand private-sector participation in electric power for water supply from dams built for other production.40 purposes emerges over time. In the case of

56 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

Tarbela, water from the river system down- Figure 2.11 Project averages for actual versus planned bulk water stream of the dam is diverted through irriga- supply delivery tion canals to Karachi to supplement other sources of municipal water supply. In the case Water supply sub-sample: 29 dams of the Aslantas, the growth of local districts 8 has prompted an application to the authorities 25% 7 for the supply of over 400 million litres per day 6 20% from the reservoir. 5 15% The Cross-Check sample showed a tenden- 4 cy towards under utilisation of capacity that 3 10% centage of dams Number of dams

is reflected more generally in a 1994 post 2 Per 5% evaluation synthesis study of 31 water supply 1 and sanitation projects by ADB. The 0 0% utilisation of capacity varied from 33–80%. below target above target Apart from the fact that evaluations were 0-49 50-74 75-89 90-99 >150 carried out at early stages in the life of the 100-109 110-124 125-149 projects, a number of causes were suggested Averages of actual to planned percentages per project for low rates of utilisation, including lower Source: WCD Cross-Check Survey.

Figure 2.12 Actual versus planned bulk water supply delivery over time

Water supply sub-sample: 29 dams 160 1 case 2 cases 3 cases 140

4 cases get 5 cases 120 6 cases 7 cases above tar

100 get 80 100% Target

60 below tar

% actual to planned Fit line

0 1 5 10 15 20 25 30 (n=no. of dams) (n=21) (n=20) (n=16) (n=16) (n=12) (n=8) Years from start of commercial operation

Source: WCD Cross-Check Survey.

The Report of the World Commission on Dams 57 Chapter 2

than expected growth in both 1995, indicating an increase in financial At current rates, water population and in per capita viability for recurrent costs.44 fees are rarely sufficient consumption.41 to recover both capital Still, a number of studies by the World Bank and recurrent costs for Financial and have demonstrated that people, even those water supply systems in who are less well off in developing coun- many developing economic profitability tries, are often willing to pay for improved countries. Single-purpose water supply water supply services.45 For example, a study dams in the WCD Knowledge in Nigeria demonstrated that on an annual Base had poor performance in financial and basis households pay water vendors over economic terms. Examination of appraisal twice the O&M costs of a piped distribution and evaluation figures for four World Bank system.46 and ADB water supply dams show that three of them dropped from EIRRs above 10% to well below that. As a whole, the EIRRs of Flood Control Dams the group fell by over 6%. For centuries societies have built levees and embankments along riverbanks to contain Poor financial and economic performance and control the effects of floods. The for large dams roughly approximates that for purpose has been to occupy floodplains for the sector as a whole. A sector synthesis of agricultural, urban, and industrial uses and post-evaluation findings from 20 ADB water to reduce any resulting threat to lives and supply and sanitation projects found that 18 property. The WCD Knowledge Base projects had actual FIRRs that failed to highlights two very different perspectives on meet projected FIRRs and 17 of these were the past performance of dams in this regard. 42 below 10%. These types of sectoral results The first is a narrow focus on the role of are confirmed by other sources. A similar dams in flood control and the second is a World Bank analysis found that almost all of broader, more integrated approach to flood 129 water supply and sewerage projects management as an objective. Evidence in 43 reviewed had EIRRs below 10%. the Knowledge Base confirms that while dams have provided important flood control At current rates, water fees are rarely benefits, some dams have increased the sufficient to recover both vulnerability of riverine communities to capital and recurrent costs for floods. Growing concern over the cost and Growing concern over the water supply systems in many effectiveness of large dams and related cost and effectiveness of developing countries. Average structural measures as long-term responses large dams and related tariffs in an ADB survey ranged to floods has led to support for integrated structural measures as from very low in Calcutta flood management as opposed to flood long-term responses to ($0.01 per cubic metre) to control. floods has led to support $0.66 per cubic metre in Cebu, for integrated flood Philippines. Thirty five of the Flood control benefits management as opposed 50 utilities in the survey cov- to flood control. ered O&M costs through their Large dams are used to control floods by tariff collections. The ratio of storing all or a portion of the flood waters in billings to O&M costs for 37 utilities the reservoir and then releasing the water increased from 1.03 to 1.12 during 1991 to slowly over time. Typically, the principal use

58 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

of such dams is to store a portion of the Extensive studies undertaken Large dams in Japan have flood in order to delay or manage when the at that time confirmed the dramatically reduced the peak occurs. This minimises the chance of operation of the reservoirs sudden arrival of floods in coincident peaks from floods in different reduced the flood peak by 2 populated areas where the tributaries arriving at the same time in the meters, leading to a correspond- rivers are exceptionally main-stem of the river, reducing the proba- ing reduction in flood emergen- steep and short, and bility of breaching dykes and overwhelming cy relief and compensation susceptible to flash floods. other flood defences. The main performance payments required by the state. parameter in assessing flood control benefits is therefore the extent of reduction of the For Grand Coulee flood control was not an flood peak. Indicators of the benefits derived initial objective and the inability of the dam from flood control include reductions in the to control the damaging 1948 flood led to area flooded and prevention of any conse- the construction of more reservoirs in quent loss of life, social disruption, health Canada to store snowmelt. Although a impacts and property and economic losses. precise breakdown of the individual contribution of the Grand Coulee is hard to isolate The Aswan High dam is an example of a because a whole cascade of dams contribute to dam that stores the flood. It can store 1.5 flood control in the Columbia basin, estimates times the average annual flow of the Nile indicate its contribution may be in the order River and has provided a high degree of of $20 million a year. protection to the lower Nile simply by retaining the whole flood. At the same Other examples in the Knowledge Base time the beneficial aspects of natural provide similar indications of the perform- flooding – for example restoring the ance of dams in reducing peak flood levels. fertility of the floodplain – have been lost, For example, during the monsoon period in a point returned to below and again in 1995, the Nam Ngum River experienced a Chapter 3.47 50-year flood three times in the same season. The reservoir absorbed the first two Four of the WCD Case Studies provide but was then full. When the third flood other additional examples of flood control, occurred the spillway gates were opened and although none of these dams were built primarily for this purpose. The Tarbela dam regulates about 16% of the annual flow of Box 2.7 Flood protection in Japan the Indus River. Analysis in the Case Study Japan is one of the top five dam-building countries in the world, and flood suggests that the early season flood peak in control is the major purpose for many dams in Japan. An estimated 50% of the Indus was reduced by 20%, however Japan’s population lives in flood-prone areas, and floods have affected 80% of municipalities in the last 10 years. The Chikugo River in the Kyushu District in there was little reduction in the downstream southern Japan has a main river channel of 143 km and a catchment area of flow during the late season flood event of 2 860 km2. A flood in 1953 caused dyke breach in many places, causing great damage in the area (147 people dead, approximately 74 000 houses flooded, 1992. Similarly, the reservoirs in the Glom- and one-fifth of the catchment inundated). Matsubara and Shimouke Dams ma and Laagen basins in Norway regulate were subsequently built for flood control and power generation. The dams have performed well; for example, during a flood in 1982 the Matsubara dam reduced about 16% of the flow in that basin, where the peak river discharge 64% from a high of 2 900 to 1 040 m3s. 20% mitigation in the peak flood level has Source: Berga, 2000, Contributing paper to also been achieved. A major flood event in Thematic IV.4 Flood Management Options; Takeuchi and Harada, 1999: p4. the Glomma and Laagen basin in 1995 prompted a Royal Commission of enquiry.

The Report of the World Commission on Dams 59 Chapter 2

the water level rose above the caused in a similar fashion when gates are The India Case Study full supply level. Due to the opened to release water at peak rainfall shows the potential retention effect of the reservoir, periods. One such serious incident was conflict between the peak of the third flood was reported in Nigeria where a delay in warning keeping space in the reduced by 20%.48 Large dams inhabitants led to a flood that overran approx- reservoir for flood in Japan have dramatically imately 200 communities, submerging 1 500 control and storing as reduced the sudden arrival of houses and killing over 1 000 people.49 much water as possible floods in populated areas where for hydropower and the rivers are exceptionally As shown in the WCD Knowledge Base, irrigation. steep and short, and susceptible flood management may only be needed a to flash floods (see Box 2.8). few days or weeks in any particular year. Thus Delay in flooding provides time large dams used for flood control virtually for public warning and evacuation where always have another function, such as power necessary. generation or irrigation. The India Case Study portrays the potential for conflict between Limitations of flood control flood control objectives for operation of the operations reservoir (where storage space in the reservoir is required) and hydropower and There are also clearly problems that emerge irrigation (where it is desirable to store as from the operation of large dams for flood much water as possible). According to the control. Some dams have increased the Case Study most of the complaints about vulnerability of riverine communities to dams aggravating floods downstream stem floods. For example: from this situation. The Case Study goes on ■ While rare, dam breaks have and do to document the lack of co-ordination or occur and usually during exceptional real-time information exchange between the storms; when they do, communities upstream Tenughat reservoir and the opera- downstream are subject to extreme floods tion agencies of the downstream Damodar amplified by the dam break. Valley Corporation, which put the down- ■ Significant downstream damage to stream river reach and reservoirs at risk. communities has resulted where reservoirs have not been operated properly in Another aspect is that with climate varia- times of emergency or rapidly developing tions, the frequency, duration and intensity situations or when floodgates have of storm events that lead to floods appear to mechanical failures at critical times. be changing. If so, there is a risk that a Often communities have adapted to the changing climate will modify the hydrologi- level of protection normally provided cal basis on which many flood control dams and the contingency plans – or their were designed. This raises concerns about implementation – have been inadequate. the physical adequacy of many dams to perform their flood management functions, There have also been cases where peaking as well as the adequacy of spillways to operation of hydropower stations has caused handle higher flood volumes likely in a an unexpected surge of water in the river; changed climate. United States dams are lives have been lost when measures to warn being reviewed for their flood design and in populations downstream have not been some cases the spillway capacity is being effective or heeded. Local flooding can be increased.50 The technical literature on

60 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

dams is also increasingly focusing on the leading to settlement of flood- role of dams in flood management, looking prone areas. When the excep- at both the safety considerations and the tional flood finally arrives, there means to improve the flood performance.51 are more people and higher- value property at risk than there From flood control to otherwise would have been. integrated flood management Damages may therefore be larger than if floods had continued to The Knowledge Base identifies a series of be normal events within the concerns over the effectiveness of dams and range of regular experience and related structural measures as an element in awareness. a broader flood management strategy including the realisation: In the Nam Ngum case referred to earlier, ■ that dams have encouraged settlement in the dam was blamed for a major inundation areas that are still subject to floods that of downstream agricultural areas despite its exceed the maximum design flood; alleviation of two earlier peaks and reduc- ■ that the costs of ensuring complete tion of the third peak. Since the reservoir protection against all floods are excep- had not spilled for many years people had tionally high; developed a false sense of security and drains were not maintained. In the end the flood- ■ that the effectiveness of structural waters only dispersed slowly and standing measures are reduced over time due to crops did not survive. Another example the accumulation of sediment in river comes from Poland where the 1997 floods beds and reservoirs; and covered an area only half as large as that in ■ that floods have many beneficial uses in 1934 yet three times as many buildings were different river basins and that the elimination or reduction of natural Box 2.8 From flood control to flood management in the United flooding has led to the loss of important States downstream ecosystem functions, as well as loss of livelihood for flood-dependent Between 1960 and 1985, the United States Federal government spent $38 billion on flood control, mostly on structural responses such as large dams. Yet average communities (see next two chapters on annual flood damage, adjusted for inflation, continued to increase – more than environmental and social performance of doubling. Average flood damage in the United States, adjusted for inflation, was as follows: large dams). 1903–33: $1.7 billion 1934–63: $2.8 billion 1964–93: $4.6 billion These concerns indicate the difficulty of 1994–97: $5.1 billion fully controlling floods and of managing the The United States Army Corps of Engineers points out that its dams and 8 500 relationship between floods and people. miles of levees have saved $387 billion in damages since 1928, but it has no figures for damages in areas where the Corps’ projects encouraged develop- This underpins the shift to an emphasis on ment that was later inundated. After the 1993 Mississippi flood, congressional testimony pointed to dams and levees as exacerbating the problem. Subse- integrated flood management – that is the quently, the Corps has called for more consideration of non-structural flood need to set objectives in terms of predicting, management methods, including restoration of wetlands and riparian habitat, limits on development in floodplains, and farm policy that discourages managing and responding to floods – rather conversion of wetlands to cropland. Likewise, communities from Rapid City, than simply in terms of flood control. South Dakota, to Valmeyer, Illinois, to St. Charles, Missouri, to Napa, California, have opted to pursue non-structural approaches to alleviate recurring floods.

Large dams with a flood control component Source: Schildgen, 1999 may provide an increased feeling of security,

The Report of the World Commission on Dams 61 Chapter 2

flooded, 38 times as many bridges and 134 enced by single-purpose projects and, in a times more kilometres of road.52 number of cases, achieved less relative to targets than their single-purpose counterparts. The high cost of flood control and flood damage in the United States, as well as the The WCD Cross-Check Survey shows that emerging shift from a reliance on flood multi-purpose dams have had a high degree control to an approach based on flood of variability in achieving physical targets management is described in Box 2.8. In across most benefit streams. As indicated many countries reliance on structural earlier, single-purpose projects in the Cross- measures, including dykes, leads to a need to Check sub-sample tend to cluster closer to continually invest in additional measures as planned targets for project schedule, hydro- sedimentation decreases their effectiveness power, and water supply performance than over time. For centuries, dykes in Vietnam multi-purpose dams. The exceptions were have been progressively increased in height irrigation projects which showed little as the river bed gradually rises due to accumu- difference between single and multi-purpose lating sediments deposited by floods.53 The projects in variability of performance. same situation prevails in China, where after centuries of raising dykes, the height of the The Cross-Check Survey also suggests that dykes at places on the Yangtze River is over 16 multi-purpose projects have higher cost metres above the floodplain.54 overruns and higher variability in these overruns than single-purpose projects. A Multi-Purpose Dams small sample of 12 multi-purpose projects funded by the World Bank, AfDB, and ADB Many large dams fulfil a number of purposes examined by the Commission indicates that with a single facility. Multi-purpose projects estimates of the EIRR at evaluation were in the WCD Knowledge Base display many about 4% below those projected at appraisal. of the same performance shortfalls experi- This figure masks large variability, with four of the projects moving up and down by almost 10%.55 In the AfDB’s review of four Box 2.9 Cost recovery in a multi-purpose scheme: Grand Coulee and projects, only one was financially and the Columbia Basin Project economically viable.56 In the case of the Grand Coulee dam and the accompanying Columbia Basin Project, the intention was for hydropower revenue to subsidise irrigation capital costs. The Bonneville Power Administration (BPA), which is responsible for GCD, As noted in previous sections, hydropower is charged with paying the United States Treasury for the costs associated with projects tend to perform relatively well in hydropower development (the dam and three powerhouses) plus the share of allocated irrigation costs not paid by irrigators. These amounts are calculated in financial terms while irrigation projects nominal dollars but include interest. As of 1998, the total capital costs for GCD and CBP came to $1.93 billion. Of this, the BPA has already repaid the typically fail to recover O&M and capital hydropower costs of $1.1 billion. costs. In practice this has often led to the The share of the remaining costs allocated to irrigation is $674 million. The last use of hydropower facilities in conjunction portion is for non-reimbursable capital costs such as flood control, as covered with a project designed for irrigation as a directly by the United States Treasury. Of the irrigation share, BPA is responsible for 87% ($585 million) and the irrigators are to cover the remaining 13% ($89 way to cover the costs of the irrigation million). As of 1998, irrigators had paid in $51 million (in dollars uncorrected for facility. The WCD Case Study of Grand inflation). BPA is scheduled to make payments on its share of the irrigation costs during 2009–45. As hydropower revenues from the sale of GCD power come Coulee dam and the Columbia Basin Project close to $500 million, it is clear that the project could easily cover costs under provides an illustrative example of the cross- such terms. Source: WCD Grand Coulee Case Study. subsidies that often result from such arrangements (see Box 2.9).

62 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

In other cases, adding hydropower to an Physical Sustainability irrigation facility is simply a way of increas- Issues ing the overall economic profitability of a scheme. The design for the Aslantas Dam Many factors affect the physical sustainabili- proceeded on this basis. Analysis of the ty of the benefits and services provided by irrigation component as a stand-alone dams. The following section provides brief project indicated a 13% rate of return. A findings from the Knowledge Base on three stand-alone hydropower project was judged of these issues: dam safety, sedimentation not a least-cost source of power, but when and waterlogging and salinity. analysed in terms of just the incremental costs and benefits of being built into an irrigation Dam safety facility, the hydropower component yielded a Dam failure is defined by the International 15.7% rate of return. Combined, the rate of Commission on Large Dams (ICOLD) as return for the multi-purpose project came to the ‘collapse or movement of part of a dam 13.4%. Following a cost overrun in nominal or its foundation, so that the dam cannot terms of 37%, the project completion team retain water.’58 In general, a failure results in reported a recalculated EIRR for the project in the release of large quantities 1985, including flood control, of 13.6%, of water, posing serious risks although the validity of the irrigation benefit for the people or property calculations was questioned by a subsequent Multi-purpose schemes downstream. The findings of a post-project audit report.57 are inherently more recent global compilation of complex, and many information about the failure In summary, these emerging trends and experience operational of dams by ICOLD are as patterns of higher variability and lower conflicts that contribute to follows: average performance of multi-purpose versus under-performance on single-purpose projects are not surprising. ■ The failure rate of large financial and economic While single-purpose dams are designed for dams has been falling over targets. optimal delivery of a particular targeted the last four decades. Of benefit, multi-purpose projects are designed for dams built before 1950, sub-optimal outputs of all intended benefits. 2.2% failed, while the failure rate of They aim to maximise economic efficiency dams built since 1951 is less than 0.5%. achieved through shared costs and infrastruc- ■ The proportion of dams failing varies ture of the proposed scheme. In doing so, little with the height of the dam and so multi-purpose schemes are inherently more most failures involve small dams. complex, and many experience operational ■ Most failures involve newly built dams. conflicts that contribute to under-performance Some 70% of failures occur in the first on financial and economic targets. ten years of life of the dam and proportionately more during the first year after What emerges from the WCD Knowledge commissioning. Base is that these sub-optimal targets set for ■ The highest failure rate is found in dams multi-purpose projects were still not built in the ten years 1910-1920. achieved to the level desired. This suggests that the extent to which conflict arising ■ Foundation problems are the most from multi-benefit operation will affect common cause of failure in concrete performance is probably under-estimated. dams, with internal erosion and insuffi-

The Report of the World Commission on Dams 63 Chapter 2

cient shear strength of the foundation ■ Where other works were the seat of the each accounting for 21% of failures. failure, the most common cause was ■ The most common cause of failure of inadequate spillway capacity (22% as earth and rockfill dams is overtopping primary cause and 30% as secondary (31% as primary cause and 18% as cause). secondary cause). This is followed by ■ The post-failure action most frequently internal erosion in the body of the dam reported was scheme abandoned (36%), (15% as primary cause and 13% as second- construction of a newly designed dam ary cause) and in the foundation (12% as (19%) and overall reconstruction with primary cause and 5% as secondary cause). the same design (16%).59 ■ With masonry dams, the most common cause is overtopping (43%) followed by When most large dam projects are built, the internal erosion in the foundation assumption is that river flows in the future (29%). (total runoff and severe floods) will be much like those in the past. In some cases the historical time series of hydrological data is Box 2.10 Dam safety in the United States too short and may not reflect cyclical The American Society of Civil Engineers (ASCE) gave dams in the United States phenomena. Climate change has introduced a poor grade (a ‘D’) in their ‘1998 Report Card for America’s Infrastructure’ – citing age, downstream development, dam abandonment, and lack of funding another level of uncertainty about changing for dam safety programmes as major unaddressed problems. flows within the life span of most dams. The The Association of State Dam Safety Officials (ASDO) concluded that: safety of large dams is affected by changes in ■ Current levels of dam safety expenditures are insufficient. There are instances when dam safety items are given a lower priority than water or the magnitude or frequency of extreme power delivery. precipitation events. These changes are ■ Some $40 billion is needed to maintain and improve current dams. highly uncertain, but climate change is ■ Little if any information is provided to downstream populations at risk; expected to lead (and perhaps already has less than 10% of jurisdictions surveyed had warning or evacuation plans. led) to larger and more frequent extreme ■ In many cases irrigation districts – which are required by law to share dam safety expenses – withheld funds to pay for necessary safety items. precipitation events. One of the first studies Continued deferment could result in unacceptable public risks. in this area concluded that the discharge of ■ Maintenance costs continue to escalate because of the ageing the 50-year flood on the River Severn, in structures. With constant or declining maintenance funds from federal and irrigation district sources, new sources of revenue are needed. the United Kingdom, may increase by 60 According to the United States Federal Emergency Management Agency: around 20% by 2050. There is concern ■ Dam breaks are rare and there have been few deaths in recent years. whether existing spillways can evacuate There have been 1 449 dam failures over the past 150 years in the such floods in future. United States. The annual dam failure rate was 29 in 1996 and 1997. ■ In massive floods like Hurricane Floyd’s inundation of North Carolina in 1999, 36 dams failed but most were small and claimed no lives. The WCD Cross-Check Survey shows a ■ Failure of the Buffalo Creek dam in West Virginia in 1972 killed 125 trend towards increased attention to the people, while failure of the Teton dam in 1976 killed 11 and of the Kelly Barnes dam in Georgia killed 39. assessment of dam safety, although around ■ Of the 80 000 small and large dams in the United States, 9 326 are rated 20% of dams in the sub-sample that were as ‘high hazard’, meaning that if they should fail, loss of life and serious built in the last three decades do not report property damage would result. Some 1 600 significant hazard dams are within one mile of a downstream city. undertaking a safety assessment (see Figure ■ Less than 40% of high hazard dams have an emergency action plan for 2.13). A report card on the United States people to follow. experience with an ageing stock of dams Sources: ASCE 1998; provides a sobering indication of the impor- Knudsen and Vogel, 1997; Schmid, 2000; ASDSO 2000. tance of dam safety concerns (see Box 2.10). A key element in keeping dams safe is

64 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

providing finance for proper and regular Sedimentation maintenance work. A recent Ontario Hydro Many reservoirs are subject to some degree Study of several hundred North American of sediment inflow and deposition. It is dams indicates that, on average, hydropower estimated that some 0.5–1% of the world operating costs rise significantly after 25 to reservoir volume is lost from sedimentation 35 years of operation due to the increasing annually.62 Sedimentation of the active need for repairs.61 storage affects physical and economic performance, but only where design storage Figure 2.13 Trends in dam safety assessments is more or less fully used.63 Sediment may also cause erosion of turbines if it reaches Global sub-sample: 105 dams power intakes. Eventually, sedimentation 110 will affect project life by silting up the dead 90 storage, leading to intake blockage. 70 50 30 The WCD Knowledge Base indicates that % of dams 10 while sedimentation potentially undermines 0 the performance of a large dam project, the conditions – and therefore the frequency of 1950s 1960s 1970s 1980s 1990s <1950 occurrence of this phenomenon – are Decade project and site-specific. For example,

Source: WCD Cross-Check Survey. Figure 2.15 Loss of active storage due Figure 2.14 Loss of active storage due to sedimentation to sedimentation by reach of river

Global sub-sample: 47 dams Global sub-sample: 47 dams 100 50

80 40

60 30

40 20 % active storage loss % active storage loss Linear fit line 10 20

0 0 0 5 10 15 20 25 30 35 40 45 each each each

Upper r Lower(9 rdams) Age of dams (years) (26 dams)Middle(12 rdams)

Source: WCD Cross-Check Survey. Source: WCD Cross-Check Survey.

The Report of the World Commission on Dams 65 Chapter 2

higher sedimentation rates are indicate storage is depleting at the rate of observed for smaller dams and for 1.1% for reservoirs with storage volumes less dams located in the lower reaches than 500 m3 and 0.6% for larger dams.65 of rivers. Waterlogging and salinity Analysis of the Cross-Check Salinisation of agricultural land is caused by Survey shows that more than the rise of groundwater brought about by 50% of active storage was lost surface irrigation and is related to the due to sediment deposition for problem of waterlogging. When water tables 10% of projects in the sample, all are close to the surface, capillary action of which were in operation for at draws salts naturally occurring in the soil least 25 years (see Figure 2.14). profile to the surface. Even where ground- But there is great variability water use is controlled, secondary salinisa- within this average global figure tion can occur due to poor-quality ground- (see Figure 2.15). In the sample, water. Salinisation reduces yields of crops the average loss of active storage that do not tolerate high salinity levels to was greatest for reservoirs in the the point of eventually rendering the land lower reach of rivers. unproductive. In India, the yields of rice and wheat on salt-affected land were approxi- Of the WCD Case Study dams, only Tarbela mately half those on unaffected land.66 faces a sedimentation problem. In this case, the reservoir has lost 18% of live storage The WCD Knowledge Base indicates that after 25 years. Although this is less than problems of waterlogging and salinity for predicted, the loss of storage capacity irrigation systems have reached serious reduces the dam’s capacity to store water for levels globally and have severe, long-term irrigation; in addition, the build-up of and often permanent impacts on land, sediment close to the dam is threatening agriculture and livelihoods. In the Cross- operation long before the end of its design Check Survey approximately one-fifth of life. Conversely, sedimentation is not an the large dam projects with an irrigation issue in the temperate dams in the Glomma component reported impacts from waterlog- and Laagen basin and Grand Coulee, where ging. Data from 11 major irrigation coun- sediment concentrations are negligible. tries indicate that approximately 20% of irrigated land is affected by salinity.67 But A survey of 547 U.S. dams found that: the variation across countries in the share of ■ one-quarter of U.S. hydropower reservoirs irrigated land affected by salt is also large, have some form of sedimentation problem; ranging from 15% in China and 33% in ■ 15% of owners and operators consider Egypt to 80% in Turkmenistan (see Figure these problems serious; and 2.16). ■ the frequency and perception of sedimentation as a problem is higher for The phenomena of waterlogging and smaller-volume reservoirs.64 salinity are not new. In the case of Grand Coulee dam, the unexpected rise in ground- Other studies confirm a number of these water levels was recognised in the early findings. Data from 42 dams in Morocco 1950s, requiring considerable additional

66 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

expenditure on drainage to control in- surface. Regardless of the Data from 11 major creased water levels. To date, 7 300 ha have timing of actual investment in irrigation countries been taken out of production through a drainage facilities, the exclu- indicate that government-sponsored set-aside programme. sion of drainage facilities from approximately 20% of In the Indus Basin, 38% of the irrigation the original project design – irrigated land is affected system is classified as waterlogged and only to require remedial action by salinity which makes production is estimated to be 25% lower later – may lead to the over- land increasingly than potential as a result of salinity. Major estimation of project net unproductive. engineering works to remove saline effluent benefits. Resolving waterlogging have recently been implemented, but it is and salinity problems entails too early to evaluate the effectiveness of this significant rehabilitation costs (underestima- approach. In justifying a major drainage tion of project costs) and loss of productivity project in Pakistan, the World Bank as- over time (over-estimation of benefits). cribed a production decline of approximately 25% to salinity and waterlogging, with A further difficulty is that predictions may specific cases reaching 40–60%.68 underestimate the time required for such problems to appear. One cause of this Although the need for drainage has been problem is over-irrigation. For example, in evident for some time, proponents have the Chashma project in Pakistan there was a often omitted the necessary infrastructure shift to more water-demanding crops such as from project plans.69 Drainage facilities rice and sugarcane, and excessive irrigation have been difficult to justify at the outset of during the early stages of project develop- a project under prevailing economic analysis ment when water was abundant. As a result, as the main benefit of drainage is realised water tables rose more quickly than expect- only after some time, namely when the ed leading to the need to invest in drainage groundwater levels have risen close to the works at an earlier date than anticipated.70

Figure 2.16 Waterlogging and salinity

Values represent the total irrigated China 6.7 land affected by salt in millions of India 7.0 hectares. United States 4.2 Pakistan 4.2 Iran 1.7 Egypt 0.9 Uzbekistan 2.4 Turkemenistan 1.0 Sub-Total 28.1 World Estimate 47.7

0 10 20 30 40 50 60 70 80 90 % of total irrigated land affected by salt

Source: Postel, 1999.

The Report of the World Commission on Dams 67 Chapter 2

Once drainage is in place, their financial targets but demonstrate recurrent costs of operation and variable economic performance relative maintenance are often not to targets, and include a number of recovered, leading to early notable under- and over-performers; degradation and a reduction in ■ built for municipal and industrial water the effectiveness of drainage supply have generally fallen short of systems. A report by the Inter- intended targets for timing and delivery national Water Management of bulk water supply and have exhibited Institute notes that despite poor financial cost recovery and eco- improved understanding of the nomic performance; process involved, the area ■ with a flood control component have adversely affected by waterlog- provided important benefits in this ging and salinity is increasing at a rate faster regard, but at the same time have led to than reclamation, pointing to conflicting an increased vulnerability to flood 71 sector interests and priorities. hazards due to increased settlement in areas still at risk from floods, and in some Findings and Lessons cases have worsened flood damages for a The degree to which large dams in the number of reasons, including poor WCD Knowledge Base have delivered operation of dams. services and net benefits as planned varies ■ which serve a number of purposes also substantially from one project to the next under-achieve relative to targets, in some with a considerable portion falling short of cases exceeding the shortfalls registered physical and economic targets. In spite of by single purpose projects, demonstrating this, the services produced by dams are that the targets were often over-optimistic. considerable – in the order of 12-16% of world food production, 19% of world The review of performance suggests two electricity supply amongst others, as present- further findings: ed in Chapter 1. In addition, the WCD ■ large dams in the Knowledge Base have a Knowledge Base confirms the longevity of marked tendency towards schedule large dams, as many continue generating delays and significant cost overruns; and benefits – even if less than planned – after ■ growing concern over the cost and 30 to 40 years of operation. effectiveness of large dams and associated structural measures have led to the A sectoral review of technical, financial and adoption of integrated flood manage- economic performance suggests that of the ment that emphasises a mix of policy and dams in the Knowledge Base, those: non-structural measures to reduce the ■ designed to deliver irrigation services vulnerability of communities to flooding. have typically fallen short of physical targets, did not recover their costs and The review also examines threats to the have been less profitable in economic physical sustainability of large dams and terms than expected; their benefit flows. Based on the extent and ■ built to deliver hydropower tend to nature of these threats the findings suggest: perform close to but still below targets ■ Ensuring the safety of dams will require for power generation, generally meet increasing attention and investment as

68 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

the stock of dams ages, maintenance Using the information on the performance costs rise and climate change alters the of large dams collected in the WCD Knowl- hydrological regime used as a basis for edge Base this chapter shows that there is the design of dam spillways. considerable scope for improving the ■ Sedimentation and the consequent long- selection of projects and the operation of term loss of storage is a serious concern existing large dams and their associated globally and the effects will be particular- infrastructure. Considering the vast amounts ly felt by basins with high geological or of capital invested in large dams, substantive human-induced erosion rates, dams in evaluations of project performance are few the lower reaches of rivers and dams with in number, narrow in scope and poorly smaller storage volumes. integrated across impact categories and scales. The resounding message is that we ■ Waterlogging and salinity affect one-fifth need better and continued monitoring of of irrigation land globally – including land technical, financial and economic perform- irrigated by large dams – and have severe, ance. The next chapters turn to the envi- long-term and often permanent impacts on land, agriculture and livelihoods where ronmental and social dimensions of large rehabilitation is not undertaken. dams.

The Report of the World Commission on Dams 69 Chapter 2

Endnotes benefit analysis. The real price of goods and services used in construction - and not the 1 The emphasis here is on the direct capital general price level - must increase at a higher costs of a project. Direct costs are those costs than expected rate for the real costs to that are incurred by public or private project ‘overrun’. owners as part of investment and operations, 13 Delays due to difficulty in reaching financial and not those that remain outside the closure do not add to interest during con- owner’s perspective (the latter being external struction changes. costs). 14 Bacon et al, 1996, p30. 2 WCD Thematic III.1 Economic Analysis, 15 Note that although regional development is Chapter 2. often an objective of irrigation projects, the 3 Bacon et al, 1996, p7-8. Commission has not found cases where it is 4 WCD Thematic III.1 Economic Analysis, made an explicit objective with firm Chapter 2; IDB, 1999. performance targets. Such benefits are also 5 Lagman, 2000, in WCD Thematic III.1 inherently distributional in nature and are Economic Analysis, Chapter 2. treated in the later section on this topic. 6 In nominal US dollar terms. Of the 35 dams 16 Actual command area figures are directly in the IRN list, 11 were already accounted comparable with ultimate command area for in the World Bank and IDB samples and targets in the Survey sub-sample as all 10 had incomplete information to calculate projects had passed the planed time frame for the nominal cost overruns in dollars. WCD achieving full command area development. Thematic III.1 Economic Analysis; McCully, 17 Cropping intensity describes the extent of 1999, eco061, WCD Submission. land utilisation in a year and reflects the 7 The India Case Study reports on a series of degree of multiple cropping. It is the ratio of historical reports of overruns in India, total area cropped per year to the irrigation including a 1983 study of 159 projects command area. showing 232% overruns. These overruns are 18 ADB, 1995, p5. calculated in local currency units not US 19 WCD Thematic IV.2 Irrigation Options, dollars and, thus, follow a different method Table 3.13, Section 3.4.2. then the other studies reported in the text. 20 OED, 1990, p4-2. 8 Bacon et al, 1996, p30; Bacon and Besant- 21 Barker and Dave, in print. Jones, 1998, p321. 22 World Bank Commodities Prices Data. 9 WCD Thematic III.1 Economic Analysis, based on OED, 1996a, p57-67. 23 OED, 1990, p v. 10 AfDB, 1998, in WCD Thematic III.1 24 Eckstein, 1958, p2. Economic Analysis, Chapter 2. 25 World Bank appraisals typically cite 10% as 11 Young, 2000, eco066, WCD Submission, p3, the ‘hurdle rate’ which a project must exceed suggests that US Bureau of Reclamation to be deemed worth undertaking. The ADB projects cost on average three times as much only approves projects that have an EIRR as is planned. over 12%, although in exceptional cases those with an EIRR of less than 12% but still 12 Financial costs of projects are determined by more than 10% are approved; ADB, 1997, projecting quantities of goods and services at p37, in WCD Thematic Review III.1 current year prices and then applying a price Economic Analysis, Technical Annex I, p10- escalation factor to account for expected 14. inflation. When actual inflation exceeds expected inflation, this contributes to 26 The sample included all 13 ADB-funded overruns in terms of cash flows (as against large dams where both completion and original budgets) but it does not necessarily evaluation reports exist and a sampling of 27 alter the underlying real cost of the goods large dams funded by the World Bank, and and services as employed in economic cost- eight projects funded by the African Devel-

70 Dams and Development: A New Framework for Decision-Making Technical, Financial and Economic Performance

opment Bank. For the full analysis see WCD 45 Whittington et al, 1991, p193; World Bank Thematic Review III.1 Economic Analysis, Water Demand Research Team, 1993, p47-52. Chapter 10. 46 Whittington et al., 1991, p193. 27 The present value is less than the full cost 47 Shalaby, 1999. due to the long development period and the 48 Oud, pers. comm. 2000. effect of discounting. 49 Atakpu, 1999, WCD Regional Consultation 28 This assumes that there was no general or Paper, p1. sector-specific excess capacity in the 50 Townshend, 2000, p81. economy. As the project began in 1945 it can only be said that the likelihood of consistent 51 MacDonald and McInally, 1998, p183. excess capacity in the US economy during 52 WCD Thematic Review IV.4 Flood Manage- the project period is low given the high ment Options, section 1.2.1. growth rates observed during this period. 53 World Bank et al, 1996, Annex 1, p51. 29 ADB, 1995, p7; OED, 1990, p4–2. 54 WCD China Country Study, Chapter 2. 30 Thakkar, 1999, p17, Contributing paper to 55 WCD Thematic Review III.1 Economic WCD Thematic IV.2 Irrigation Options. Analysis, Chapter 10. 31 WCD Thematic Review IV.2 Irrigation 56 AfDB, 1998, p6. Options, Section 4.2.2. 57 WCD Aslantas Case Study; World Bank, 1987. 32 Molden et al, 1998, p15. 58 ICOLD, 1995. 33 Johnson, 1997, p28. 59 ICOLD, 1995. 34 ADB, 2000. 60 Tedd, 2000. 35 NEA, 1997. 61 McCully 1997b, p2. 36 WCD Norway Case Study; WCD Grand 62 Mahmood, 1987. Coulee Case Study. 63 Active storage is the reservoir volume above 37 Wong, 1994, in WCD Thematic Review III.1 the outtake, whereas dead storage is the Economic Analysis, Chapter 2. reservoir volume below the outtake. 38 McCully, 1997a, p5, in WCD Thematic 64 Dixon, 2000. Review III.1 Economic Analysis, Chapter 10. 65 Smith, 1999, p13. 39 AfDB, 1998, p6, in WCD Thematic Review 66 Kijne et al, 1998, p26. III.1 Economic Analysis, Chapter 2. 67 Heuperman, 1999, Contributing Paper to 40 WCD Thematic Review III.2 Financing WCD Thematic Review IV.2 Irrigation Trends, Chapter 2. Options, Section 4.2. 41 ADB, 1994, p3. 68 World Bank, 1997, p3. 42 ADB, 1994, p17. 69 Young, 2000 eco066, WCD Submission, p3. 43 OED, 1992, p1 Executive Summary. 70 ADB, 1984. 44 McIntosh and Yñiguez, 1997, p16. 71 Kijne et al, 1998, p26.

The Report of the World Commission on Dams 71 Ecosystems and Large Dams: Environmental Performance

Chapter 3: Ecosystems and Large Dams: Environmental Performance

he nature of the impacts of Tlarge dams on ecosystems is generally well known and scientists, NGOs and professional groups such as the International Commission on

Large Dams (ICOLD), International Hydropower Association (IHA) and the International Energy Agency

(IEA) have written extensively on them.1 A useful indicator of the scale of human intervention in this regard is

a recent estimate that dams, inter- basin transfers, and water withdrawals for irrigation have fragmented 60% of

the world’s rivers.2

The Report of the World Commission on Dams 73 Chapter 3

Ecosystem impacts can be classified accord- impacts with decreasing certainty, from first- ing to whether they are: to third-order impacts. To date efforts to ■ first-order impacts that involve the counter the ecosystem impacts of large dams physical, chemical, and geomorphologi- have had only limited success. This is due to cal consequences of blocking a river and limited efforts to understand the ecosystem altering the natural distribution and and the scope and nature of impacts, the timing of streamflow; inadequate approach to assessing even anticipated impacts and the only partial ■ second-order impacts that involve changes success of minimisation, mitigation and in primary biological productivity of compensation measures. ecosystems including effects on riverine and riparian plant-life and on down- This chapter describes the nature of the stream habitat such as wetlands; or impacts in general, supported by the Case ■ third-order impacts that involve Studies, the Cross-Check Survey and the alterations to fauna (such as fish) Thematic Reviews on Ecosystems (II.1) and caused by a first-order effect (such as on Global Change (II.2). Given that the blocking migration) or a second-order WCD Knowledge Base describes a large effect (such as decrease in the availa- number of ecosystem impacts, the focus is bility of plankton). on summarising these impacts by grouping them as follows: In addition, modifying the ecosystem ■ the impacts of reservoirs on terrestrial changes the biochemical cycle in the ecosystems and biodiversity; natural riverine system. Reser- voirs interrupt the downstream ■ the emission of greenhouse gases associ- flow of organic carbon, leading ated with large dam projects and their to emissions of greenhouse gases reservoirs; such as methane and carbon ■ the impacts of altered downstream flows dioxide that contribute to on aquatic ecosystems and biodiversity; climate change. ■ the impacts of altering the natural flood cycle on downstream floodplains; The current state of knowledge ■ the impacts of dams on fisheries in the indicates that large dams have upstream, reservoir and downstream many mostly negative impacts on areas; ecosystems.3 These impacts are ■ the enhancement of ecosystems through complex, varied and often profound in reservoir creation and other means; and nature. In many cases dams have led to the irreversible loss of species populations and ■ the cumulative impacts of a series of ecosystems. Because the ecosystem impacts dams on a river system. are many and complex it is hard to give a precise and detailed prediction of the The extent to which efforts to reduce or changes likely to result from the construc- eliminate these impacts were undertaken in tion of a dam or series of dams. Based on the the past is described at the end of each geographical location of a dam and the discussion. The final subsection contains a natural river regime it is possible to give a further assessment of past experience with broad indication of the type and direction of efforts to avoid, mitigate, minimise or

74 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

compensate these impacts. Current efforts to conventional wisdom that hydropower restore environmental function through produces only positive atmospheric effects, decommissioning are also reviewed. such as a reduction in emissions of carbon dioxide, nitrous oxides, sulphuric oxides and Terrestrial Ecosystems and particulates when compared with power generation sources that burn fossil fuels.8 It Biodiversity also implies that all reservoirs – not only The construction of a storage dam and hydropower reservoirs – emit GHGs. subsequent inundation of the reservoir area Consequently, reservoir and catchment effectively kills terrestrial plants and forests characteristics must be investigated to find and displaces animals. As many species out the likely level of GHG emissions. prefer valley bottoms, large-scale impoundment may eliminate unique wildlife habitats All large dams and natural lakes in the and affect populations of endangered boreal and tropical regions that have been species.4 Efforts to mitigate the impacts on measured emit greenhouse gases (carbon fauna have met with little success (see Box dioxide, methane, or sometimes both) (see 3.1). Construction of irrigation infrastruc- Figure 3.1 and 3.2).9 Figure 3.1 shows the ture may have similar impacts. range of values recorded by field measurements and models of GHG emissions for 15 Flooding a reservoir may lead to the occupa- reservoirs.10 Some values for gross GHG tion and clearing of upstream catchment emissions are extremely low and may be 10 areas as replacement for land lost to the times less than the thermal option. Yet in reservoir. Land use change provoked in this some circumstances the gross emissions can manner not only has direct effects in terms be considerable, and possibly greater than of habitat loss, elimination of flora and the thermal alternatives.11 These emissions fauna and, in many cases, land degradation, may change significantly over time as the but also feedback effects on the reservoir through alterations in hydrologic function. Box 3.1 Mitigating and compensating for terrestrial impacts The resulting loss of vegetative cover leads Some large dam projects have tried to mitigate terrestrial impacts on to increases in sedimentation, stormflow, biodiversity by physically rescuing animals from the area to be flooded or by and annual water yield; decreases in water anticipating that mobile species will simply move to neighbouring areas. Operation Noah and Operação Curupira are two examples undertaken at Kariba quality; and variable changes in the seasonal and Tucurui dams. The respective WCD Case Studies show that neither timing of water yield.5 programme yielded tangible benefits for the wildlife involved. This may be a result of the implicit and probably incorrect assumption that the recipient habitat was not already at carrying capacity for the species concerned. Greenhouse Gas Emissions An alternative to mitigation is a compensatory project approach, or environmental ‘offsets’. For example, in India there is a legal requirement that forests The emission of greenhouse gases (GHG) flooded by reservoirs must be replanted elsewhere. However, the India Case Study found that only half of the required forest has typically been planted – and from reservoirs due to rotting vegetation and even this is poorly managed, yielding little in the way of comparable benefits or carbon inflows from the catchment is a services. Additional compensatory measures may include either trust funds established through grants from developers (for example Harvey Basin recently identified ecosystem impact (on Restoration Trust, Australia) or trust funds that manage parts of the revenue climate) of storage dams.6 A first estimate stream and use it for environmental purposes. This latter model is proposed for the planned Nam Theun II dam in Laos, with the intention of creating and suggests that the gross emissions from managing a National Park in the catchment. The plan has the potential to reservoirs may account for between 1% and benefit both forest ecosystems and the lifespan of the dam through reduced sedimentation. 28% of the global warming potential of Sources: WCD Kariba, Tucurui and India Case Studies; Bizer, 2000 GHG emissions.7 This challenges the

The Report of the World Commission on Dams 75 Chapter 3

biomass decays within the reservoir during Establishing that a reservoir emits GHGs is the first few years of impoundment. In other not enough to assess the impact of a dam on cases the emissions may depend more on climate change. Natural habitats (undis- carbon inflows from the catchment in the turbed by dams) may also emit GHGs (see longer term and have greater stability over Figure 3.2). Alternatively they may store time, subject to catchment conditions. carbon or act as a net carbon sink. For example, a floodplain tropical forest in Figure 3.1 Gross greenhouse gas emissions from reservoirs Amazonia may emit methane from soils and, at the same time, absorb carbon dioxide in leaves. The balance of all these potentially

4 500 counteracting effects would determine the profile of GHG emissions from the natural /yr)

2 4 000 habitat without the dam. A further compli- /m 2 3 500 cation is that the land use change induced 2 500 by displacement of people, resource extrac- 2 000 tion and other economic activities may also

1 500 form part of the net contribution to GHG emissions associated with the construction 1 000 of the dam. Calculations of the contribution 500 of new reservoirs to climate change must

verage GHG emissions (g CO 0

A therefore include an assessment of the

Brazil Canada Finland natural pre-dam emission or sink in order to determine the net impact of the dam.13

Source:WCD Thematic Review II.2 Global Change Note: Average measured emissions of greenhouse gases from 15 reservoirs in boreal The WCD Case Studies only provide data and tropical regions show large variations within countries and between regions. These on carbon dioxide and methane emissions 12 averages mask strong seasonal and annual variations. from the Tucurui reservoir (see Box 3.2).

Figure 3.2 Greenhouse gas emissions from natural habitats

3 500 /yr)

2 Others 2 500 /m

2 Indonesian 2 000 wetland rice Natural lakes Forest ecosystems 1 500

1 000 Amazon floodplain 500 Northern peatlands 0 Tropical forests -500

verage GHG emissions (g CO Boreal forests A -1 000

Source: WCD Thematic Review II.2 Global Change. Note: Natural habitats may be sources (positive values) or sinks (negative values) for carbon. Emissions from dam reservoirs in Canada and Finland (Fig 3.1) are at similar levels to natural lakes.

76 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

Even in this case there is no data on emis- To date, no experience exists with minimising, sions without the dam, making a conclusion mitigating, or compensating these impacts. on the net effect impossible. This applies Pre-inundation removal of vegetation is one more generally to other dams that have alternative, but the net effects of such an even less information available. Current activity are not well understood. The outcome understanding of emissions suggests that of global negotiations on climate change may shallow, warm tropical dams are more likely bear on future penalties and incentives for net to be major GHG emitters than deep cold GHG emissions from dams. boreal dams. Downstream Aquatic In the case of hydropower dams, tropical Ecosystems and dams that have low installed capacity and large shallow reservoirs are more likely to Biodiversity have gross emissions that approach those of Storage dams are intended to alter the comparable thermal alternatives than those natural distribution and timing of stream- with small, deep reservoirs and high inflow. They compromise the dynamic aspects stalled capacity.14 of rivers that are fundamental to maintain-

Box 3.2 Greenhouse gas emissions at Tucurui, Brazil

Recent monitoring in the 2 600 km2 reservoir of Tucurui show that emissions.16 The figure below compares these gross emissions to greenhouse gas emissions are substantial and highly variable from those of alternative technologies for large-scale power genera- year to year. Values in 1998 exceeded those measured in 1999 by tion.17 Background emissions from natural pre-impoundment more than a factor of 10 for methane and by 65% for carbon dioxide habitats have not yet been measured for Tucurui, so true compari- (see table below).15 sons of net emissions with alternatives remain elusive. Total Gross Emissions (tons/km2/ year) The alternative technology for large-scale electricity generation required for aluminium smelting (the main consumer of electricity) Year Methane Carbon dioxide was thermal power employing diesel fuel when the project was built 1998 76.36 3 808 in the 1970s. Today the alternative would be gas combined cycle 1999 5.33 2 378 plants. Modelling taking into account emissions from water passing through Source: WCD Tucurui Case Study. the turbines or over the spillway leads to higher estimates of total

Range of annual gross GHG emissions at Tucurui Hydropower Project and four thermal alternatives 2 500

2 000 /TWh/yr

2 1 500

1 000

Kt eq. CO 500

0 Tucuruia Tucuruib Tucuruic Gas Dieseld Heavy oild Bituminous combined coald cycled

Sources: aFearnside, 1995; bRosa et al, 1999; cFearnside, 2000; dIEA, 2000.

The Report of the World Commission on Dams 77 Chapter 3

ing the character of ic has positive effects on some species and aquatic ecosystems. negative effects on others (see Box 3.3). Natural rivers and their habitats and Impacts of changes in flow species are a function regimes of the flow, the quanti- Flow regimes are the key driving variable for ty and character of the downstream aquatic ecosystems. Flood sediment in motion timing, duration and frequency are all through the channel, critical for the survival of communities of and the character or plants and animals living downstream. composition of the Small flood events may act as biological materials that make up the bed and banks of triggers for fish and invertebrate migration: the channel. The defining river discharge major events create and maintain habitats includes both high- and low-flow elements. by scouring or transporting sediments. The These dynamics, not the average conditions natural variability of most river systems of controlled dam operations, determine a sustains complex biological communities stream’s physical foundation, which in turn that may be very different from those 18 ensures ecosystem integrity. adapted to the stable flows and conditions of a regulated river. Finally, water temperature The extent of impacts will also depend on and chemistry are altered as a consequence whether water is extracted or diverted for of water storage and the altered timing of consumption, or left instream. Introduction downstream flows. Algal growth may occur of non-native species, modified water in the reservoir and in the channel immedi- quality (temperature, oxygen, nutrients), ately downstream from dams because of the loss of system dynamics, and loss of the nutrient loading of the reservoir releases. ability to maintain continuity of an ecosys- Self-purification processes diminish this tem result in ecologically modified river effect downstream. systems. The establishment of a new dynam- Storage dams, particularly hydropower peaking plants, can significantly disrupt the Box 3.3 How one dam has affected two different species in opposite ways whole flow regime, resulting in both high seasonal and day-to-day fluctuations that Before it was dammed, the Waitaki River in New Zealand was highly unstable, flooded frequently, and had a constantly changing channel. After damming the differ greatly from natural flow levels. As river flood runoff is now stored in the reservoir to produce electricity. This shown in Figure 3.3, the construction of increased the stability of the sandbars downstream, allowing colonisation by vegetation, which further stabilised the channel. The increased flow stability has Glen Canyon dam on the Colorado River in benefited chinook salmon (Oncorhynchus tshawytscha) populations, an exotic the United States reduced daily average species introduced in the early 1900s, because the more stable channels provide more shelter for fry at high flows as well as a larger area of spawning gravel. flows during the annual September peak from Currently the Waitaki has the largest population of salmon in New Zealand. about 2 000 m3/sec to about 700 m3/sec. In However, the beneficial change for salmon has been detrimental to the black stilt (Himantopus novaezealandiae) – a native species. This bird is so endangered addition, as shown in Figure 3.4, streamflow that fewer than 100 individuals remain. They nest exclusively on the large can fluctuate daily more than 425 m3/sec exposed sandbars isolated from the shore, a habitat that was maintained by the unstable nature of the river. The vegetation that has proliferated and stabilised due to dam releases for electricity generation the gravel bars has increased the cover for predators, which in turn have exacted during the peak daytime demand periods. a significant toll on adult stilts, eggs, and nestlings. Source: Ligon et al, 1995, cited in WCD These changes in flow have dramatically Thematic Review II.1 Ecosystems altered the riverine environment, creating consistently colder temperatures due to

78 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

release of water from the bottom of the fish Tandanus tandanus in Australia’s Murray reservoir. A general decline in native fish River disappeared due to short-term fluctua- abundance in the Colorado River is attribut- tions in water level caused by reservoir ed specifically to the cold-water release from releases in response to downstream water- large dams there.19 The population of the user requirements.20

Figure 3.3 Modification of annual flow regimes due to a hydropower dam, Colorado River at Lee’s Ferry, United States

1922-1931 before dam 60 1982-1991 after dam

20 eam flow (1 000 cubic ft/sec) Str 0

1-Jul 1-Jan 1-Feb 1-Mar 1-Apr 1-May 1-Jun 1-Aug 1-Sep 1-Oct 1-Nov 1-Dec

Source: Data from United States Geological Survey, 2000.

Figure 3.4 Fluctuation of daily streamflow regime due to hydropower peaking operations, Colorado River at Lee’s Ferry, September

5 eam flow (1 000 cubic ft/sec) Str 0

1-Oct 15-Sep 16-Sep 17-Sep 18-Sep 19-Sep 20-Sep 21-Sep 22-Sep 23-Sep 24-Sep 25-Sep 26-Sep 27-Sep 28-Sep 29-Sep 30-Sep

Source: Data from United States Bureau of Reclamation, 2000b. Notes: Peak flows are associated with the power generation between 14.00 and 19.00 daily, with minima at 04.00 am, and the fluctuation in demand also varies from day to day.

The Report of the World Commission on Dams 79 Chapter 3

Particularly high hydropower australis) of 41 000 ha of riverbed has The modified habitats dams cause gas to become occurred as a result of stabilised flows on the resulting from large dams supersaturated when water flows Orange River. often create over the spillway. This causes environments that are fish deaths due to a condition Biological linkages also extend laterally more conducive to non- similar to the bends that can away from and parallel to the river, pushing native and exotic plant, affect divers who dive too deep the effect of river changes onto a band of fish, snail, insect, and for too long. The Grand Coulee varying width. As long as the river flow is animal species. Case Study reports that this is a sufficient, other wildlife such as deer, particular problem on the antelope, and elephants will come to the Columbia River in the United States, where water, especially in the dry/hot season, to regulators have fixed a maximum total drink. Hippos in Africa will use water of dissolved gas concentration to reduce sufficient depth as a daytime refuge, emerg- impacts on migratory fish. ing to forage at night. Many species of birds and bats fly in to drink. These lateral The modified habitats resulting from large movements can extend to several kilometres dams often create environments that are from the river. Thus many wildlife species in more conducive to non-native and exotic a fairly wide strip of land on either side of plant, fish, snail, insect, and animal spe- the river depend upon it, and they may all cies.21 These resulting non-native species be affected when the flow of the river is often out-compete the natives and end up disrupted by the construction of a large dam. modifying ecosystems that may become Equally, long reservoirs that extend many unstable, nurture disease vectors, or are no kilometres up river valleys present a barrier longer able to support the historical envi- for terrestrial species inhabiting each ronmental and social components. riverbank that were previously able to cross the river. Compared with a natural river, the root systems of plants in rivers below dams When waters of one basin are diverted into experience reduced effects of another one, changes in volume and season- scour, the plants themselves ality of flow result. New biota from the suffer less stress from high source basin may invade the recipient basin discharges and the rate of chan- and compete with the native species. If all nel migration is reduced, so that the water is diverted from the source basin, an area of the channel-bed this will clearly have serious impacts on any available for the development of unique species or genetically different stocks. aquatic plants can be stabilised. In South Africa, for example, the diversion of In both Africa and Australia, the the Orange River into the Great Fish River elimination of high discharges to resulted in a six-fold increase in flow, making flush systems has allowed the the Great Fish River permanent rather than extensive development of the intermittent. The Orange River Case Study aquatic weeds Water Hyacinth (Eichhornia reports that one beneficiary is the larvae of crassipes) and Water Fern (Salvinia moles- the blackfly Simulium chutteri, which does ta).22 The Orange River Pilot Study docu- not tolerate desiccation. In the absence of ments the colonisation by reeds (Phragmites control measures, this biting fly causes live-

80 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

stock losses, reduces recreational use, and The consequence of reduced sediment also irritates local farmers. extends to long stretches of coastline where the erosive effect of waves is no longer Efforts to minimise the impacts of changes sustained by sediment inputs from rivers. It in flow regime have relied on measures to is estimated that the coastlines of Togo and restore the streamflow regime through the Benin are being eroded at a rate of 10-15 setting of environmental flow releases (EFR) metres a year because the Akosombo dam (see Box 3.4) on the Volta River in Ghana has halted the sediment supply to the sea.25 Another Impacts of trapping sediments example is the Rhone River in France, and nutrients behind a dam where a series of dams has reduced the quantity of sediment transported by the The reduction in sediment and nutrient river to the Mediterranean from 12 million transport in rivers downstream of dams has tons in the 19th century to only 4 to 5 impacts on channel, floodplain and coastal million tons today.26 This has led to erosion delta morphology and causes the loss of rates of up to 5 metres per year for the aquatic habitat for fish and other species. beaches in the regions of the Camargue and Changes in river water turbidity may affect the Languedoc, requiring a coastal defence biota directly. For example, plankton budget running into millions of dollars. production is influenced by many variables, including turbidity. If this is reduced due to Measures for mitigating the impacts of trap- impoundment, plankton development may ping sediments and nutrients are limited. be enhanced and may occur in new sections Where feasible, flushing sediments can be part of a river. of a programme of managed flood releases.

Reduction in sediment moving downstream from the dam leads to degradation of the river channel below the facility.23 This can Box 3.4 Minimising impacts of changes in streamflow regimes: environmental flow requirements lead to the elimination of beaches and backwaters that provided native fish habitat, At least twenty-nine countries seek to minimise ecosystem impacts from large dams by using an EFR to meet predetermined ecosystem maintenance and the reduction or elimination of riparian objectives. The practice of EFRs began as a commitment to ensuring a vegetation that provides nutrients and ‘minimum flow’ in the river (often arbitrarily fixed at 10% of mean annual runoff). It has since grown to include a definition of ecosystem requirements and a habitat for aquatic and waterfowl species, planned flow release programme, which may vary annually or seasonally, to among others. Impounding rivers invariably meet downstream needs for both the environment and people. The level of EFR required is determined by the need to maintain particular ecosystem compo- results in increased degradation of coastal nents downstream, often with reference to national legislation. The countries deltas due to reduction in sediment input. that use this method have recognised that a short-term reduction in financial returns from a project often leads to improved long-term sustainability and For example, the slow accretion of the Nile attainment of broader societal objectives for a healthier environment. Still, this Delta was reversed with the construction of represents a re-distribution of the benefits of a dam project and thus existing beneficiaries such as irrigators and operators of hydropower facilities may resist the Delta barrage in 1868. Today, other EFRs. dams on the Nile, including the Aswan Amongst the WCD Case Study large dams, only Grand Coulee has an environ- High Dam, have further reduced the mental flow requirement, in this case consisting of a specially designed release for flow augmentation for salmon, while avoiding high total dissolved gas amount of sediment reaching the delta. As a concentrations. Implementation of a planned release to maintain downstream result, much of the delta coastline is eroding ecosystems is being considered for the Orange River in South Africa. Sources: Brown et al, 1999, Contributing Paper for WCD Thematic Review II.1 by up to 5–8 metres per year, but in places Ecosystems; Tharme, 2000 this exceeds 240 metres per year.24

The Report of the World Commission on Dams 81 Chapter 3

Blocking migration of aquatic host fish. To help counteract the drift organisms downstream of their larvae, aquatic insect adults such as mayflies and stoneflies move As a physical barrier the dam disrupts the upstream to lay their eggs.27 Dams block movement of species leading to changes in these migrations to varying degrees. upstream and downstream species composition and even The WCD Cross-Check Survey found that species loss. River-dwelling As a physical barrier the impeding the passage of migratory fish species have several migratory dam disrupts the species was the most significant ecosystem patterns. These include anadro- movement of species impact, recorded at over 60% of the projects mous fish such as salmon and leading to changes in for which responses on environmental issues catadromous fish such as eels. upstream and were given. In 36% of these cases, the Adults of the former migrate up downstream species impact of the large dam on migratory fish rivers to spawn and the young composition and even was not anticipated during project planning. species loss. descend, while the reverse occurs with the latter. But many other Migratory fish require different environ- freshwater fish move up rivers or ments for the main phases of their life cycle: their tributaries to spawn, while the glochid- reproduction, production of juveniles, ia larvae of freshwater mussels are carried by growth, and sexual maturation. Many anadromous fish populations such as salmon Box 3.5 Mitigation measures: fish passes and shad have died out as a result of dams blocking their migratory routes. The stur- Fish passes are often used as an engineered mitigation measure for reducing impacts on fish. However, very few of the over 400 large dams in Australia have geon populations in the Caspian Sea now fish passes of any description, only 16 had been constructed on the 450 large rely on stocking from hatcheries (mainly in dams in South Africa by 1994, and only 9.5% of 1 825 hydropower dams in the United States have an upstream fish pass facility. An example is Idaho Power Iran), as dams built by the former Soviet Company which built fish passes into each of its dams in the Hells Canyon Union on rivers entering the sea halted Complex. However, all were unsuccessful and salmon no longer migrate above Hells Canyon Dam. natural spawning migrations.28 The Glomma and Laagen Case Study reports that there are 34 fish passes on the 40 dams in this Norwegian basin. Of these only 26% work with ‘good efficiency,’ Detailed studies in North America indicate 41% work less well, and as many as 32% are not working at all. In general, the efficiency is considered low, and fish migrations are severely affected. At Pak that dam construction is one of the major Mun Dam in Thailand, the case study documents the ineffectiveness of the fish causes for freshwater species extinction. For pass, especially for the large migratory species in the Mekong that may be up to two metres long and cannot fit through the 15x20 cm slots. Grand Coulee, example, a study of the threatened fish of Tucurui, Tarbela and Aslantas have no fish passes despite having migratory fish Oklahoma suggested that the loss of free- species in the river. flowing river habitat due to reservoirs had Even when fish passes have been installed successfully, migrations can be delayed by the absence of navigational cues, such as strong currents. This led to 55% of the human-induced species causes stress on the energy reserves of the fish, as anadromous fish such as loss, while a further 19% was caused by salmon do not feed during migration. dams acting as barriers to fish migration.29 Recent research in Australia, the United States, and Japan has shown that fish passes need to be modified to meet the needs of each species and the The best-documented examples of disrupted particular situation at each dam. They cannot simply be considered an easily transferable technology, as shown by the Pak Mun fish pass, which used a fish migrations are from the Columbia River design appropriate for leaping trout and salmon in mountain streams, but which in the United States, where many stocks of was ineffective for species living in the slower-flowing Mekong. salmon have been lost. The impact of these Sources: Australia in Blackmore, pers. comm. 2000; South Africa in Benade, pers. comm. 1999; USA in Francfort et al, 1994, Executive Summary pviii;Collier disruptions on the productivity of the et al, 1996, p22; Larinier, 2000, Contributing Paper for WCD Thematic Review fishery are described below. II.1 Ecosystems

82 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

Fish passes are typically used where efforts system. This, together with the associated are made to mitigate the effect of dams in loss of floodplain habitats, normally has a blocking migrations of fish (see Box 3.5). marked negative impact on fish diversity and productivity. The connection between Floodplain Ecosystems the river and floodplain or backwater habitats is essential in the life history of Reduction in downstream annual flooding many riverine fish that have evolved to take affects the natural productivity of riparian advantage of the seasonal floods and use the areas, floodplains and deltas. The character- inundated areas for spawning and feeding. istics of riparian plant communities are Loss of this connection can lead to a rapid controlled by the dynamic interaction of decline in productivity of the local fishery flooding and sedimentation. Many riparian and to extinction of some species. Addition- species depend on shallow floodplain ally, dewatering of stream chan- aquifers that are recharged during regular nels immediately downstream flood events. Dams can have significant and from dams can be a serious In Africa, the changed complex impacts on downstream riparian problem. hydrological regime of plant communities. High discharges can rivers has adversely retard the encroachment of true terrestrial The direct loss of annual silt and affected floodplain species, but many riparian plants have nutrient replenishment as a agriculture, fisheries, evolved with and become adapted to the consequence of upstream im- pasture and forests that natural flood regime. poundment is thought to have constituted the contributed to the gradual loss of organising element of Typically, riparian forest tree species are fertility of formerly productive community livelihood dependent on river flows and a shallow floodplain soils as used in agri- and culture. aquifer, and the community and population culture and flood-recession structure of riparian forests is related to the agriculture. Dramatic reductions in bird spatial and temporal patterns of flooding at species are also known, especially in down- a site. For example, the Eucalyptus forests of stream floodplain and delta areas, where the Murray floodplain, Australia, depend on wetlands may not be replenished with water periodic flooding for seed germination, and and nutrients once a dam is installed. headwater impoundment has curtailed Finally, recharge of groundwater in flood- regeneration.30 Conversely, artificial pulses plain areas is severely diminished once generated by dam releases at the wrong time floods are eliminated.32 – in ecological terms – are recognised as a cause of forest destruction. For example, In Africa, the changed hydrological regime Acacia xanthophloea is disappearing from the of rivers has adversely affected floodplain Pongolo system below Pongolapoort dam, agriculture, fisheries, pasture and forests that South Africa, as a result of the modified constituted the organising element of flood regime.31 community livelihood and culture. Econom- ically important wetlands in Africa include The control of floodwaters by large dams, river floodplains, freshwater lakes and which usually reduces flow during natural coastal and estuarine environments. In the flood periods and increases flow during dry Sahel, there are major wetlands in the Delta periods, leads to a discontinuity in the river Intérieur of the River Niger in Mali and

The Report of the World Commission on Dams 83 Chapter 3

Lake Chad (on the border between Niger, Fisheries Chad, Cameroon and Nigeria). These have As indicated earlier, the blockage of sedi- counterparts elsewhere in semi-arid Africa, ment and nutrients, the re-regulation of notably the Sudd in Sudan and the Okavan- streamflow, and elimination of the natural go Delta in Botswana, and in humid areas, flood regime can all have significant, such as the swamps of eastern Zaire. negative effects on downstream fisheries. Marine or estuarine fisheries are also nega- Some of these wetlands cover extensive tively affected when dams alter or divert tracts of land. The fringing floodplain of the freshwater flows. Still, productive reservoir Senegal River covers some 5 000 km2 in fisheries can follow from dam construction, flood, and shrinks to 500 km2 in the dry although they are not always anticipated or season. The fringing floodplain of the Niger part of project design proposals. covers about 6 000 km2 in the flood season, shrinking to about half that at low water, Substantial losses in downstream fishery while the Niger Inland Delta extends to production as a result of dam construction 20 000 to 30 000 km2 in the flood season, are reported from around the world. Along shrinking to 4 000 km2 at low water. In the with subsistence agriculture, fisheries Logone-Chari system, flooding covers some constitute an important livelihood activity 90 000 km2.33 among large rural populations in the developing world. Many of these households Efforts to restore floodplain ecosystem depend on fisheries either as a primary or functions rely on reversing the effects of the supplementary source of livelihood. For dam through a program of managed floods example, the partial closing of the river designed to simulate the floods that oc- channel by Porto Primavera dam in Brazil curred prior to the dam (Box 3.6). blocked fish migration and diminished upstream fish catch by 80%, affecting livelihoods.34 In areas of rich fish species diversity, such as the lower Mekong region Box 3.6 Restoring ecosystem function through managed floods in East Asia, community livelihoods and The WCD Knowledge Base includes a number of cases where artificial floods culture are woven around fisheries. The Pak have been released from large dams to regenerate the natural resource base of downstream floodplains for local livelihoods (for example Manantali dam in Mali Mun Case Study reports a drastic decline in and Senegal and the Pongolapoort dam in South Africa). Managed floods upstream fish catch once the dam had generate economic benefits when downstream communities depend on natural, flood-maintained resources such as grazing, flood-recession agriculture and effectively blocked fish migration from the fishing (see Chapter 4). For example, on the Tana River, Kenya, a released flood from the planned Grand Falls scheme would have a net present value of at least Mekong River upstream into tributaries of $50 million for the downstream floodplain economy. Managed floods also entail the large Mun River watershed. an opportunity cost which may be greater or lesser depending on the value of the released floodwaters to the dam for irrigation, hydropower or other uses. A set of preliminary studies show that in some cases there are clear net economic Data on the losses to downstream fishery benefits to these releases and in other cases the opportunity costs exceed the value of downstream benefits that were identified, quantified and valued in production as a result of dam construction economic terms. The potential for managed floods is often constrained by the are reported from river basins in Africa as design of the sluice-gates, sedimentation in the reservoir and in downstream channels and the development of infrastructure on areas previously prone to well. For example, 11 250 tonnes of fish per flooding. Another constraint may be the political will to support traditional year from the Senegal River system were lost means of livelihood at the expense of benefits from the dam. following dam construction.35 Studies on Source: Acreman et al, 2000, Contributing Paper to the Niger have shown that fish productivity WCD Thematic Review II.1 Ecosystem Impacts; Grand Falls in Emerton, 2000 increases linearly with the volume of river

84 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

flow.36 Other basins in the WCD Knowl- Productive reservoir fisher- edge Base reporting loss in fish production ies often follow from dam include the yaeres in Cameroon, the Pongo- construction, although they lo flood plain in South Africa, and the are not always anticipated Niger in West Africa below Kainji dam.37 or part of project design Adverse impacts have been felt in the delta proposals. While practically and estuary areas in the lower Volta region, all the WCD Case Study in the Nile delta, and on the Zambezi in dams have reservoir fisher- Mozambique.38 ies, predictions of fish production were made in Freshwater flows also help support marine only three cases. In the case of Aslantas, fish production as many marine fish spawn consultants estimated a production of 580 in estuaries or deltas. A decrease in freshwa- tonnes, and actual figures are 86–125 tonnes ter flow and in nutrients due to dam con- (for 1987–95). At Pak Mun, actual produc- struction affects the nursery areas in a tion came to only one-tenth of prediction. number of ways, including increasing In both cases targets were not met as the salinity, allowing predatory marine fish to predictions depended partly on a functional invade, and reducing the available food stocking programme that was not fully supply. These impacts are well illustrated by implemented. At Kariba estimates were the effect of the Aswan High Dam on the made for the artisanal fishery but not for the coastal waters of the Mediterranean, where more productive open-water commercial reduction in nutrients transported to the sea fishery. has lowered production at all trophic levels, resulting in a decline in catches of sardines On the other hand, an unanticipated but and other fish.39 In the Zambezi delta, the productive fishery has developed in Tucurui. impact of modified seasonal flows on local In addition to commercial production at shrimp fisheries has been estimated at $10 Tucurui, a local sport-fishing industry has million per year.40 developed around the sought-after Peacock Bass, Cichla ocellaris (known in Tucurui as Dams can enhance some riverine fisheries, tucunaré), Kariba has seen a similar develop- particularly tailwater fisheries immediately ment of a vibrant sport fishery. At Grand below dams that benefit from discharge of Coulee, the lack of a fish pass deprived nutrients from the upstream reservoir. If salmon of over 1 000 kilometres of upstream discharge is from the lower layer of water in spawning grounds, and affected First Nations the reservoir, lowered temperatures in the tribes in the United States and Canada, receiving tailwater can curtail or eliminate while a fish hatchery largely maintained warmwater river fisheries and require salmon numbers (but not genetic diversity) stocking with exotic coldwater species such in downstream runs in the United States. as salmonids (assuming that the water is sufficiently oxygenated). Productive tailwa- Data from before and after the construction ter fisheries targeting these coldwater fish of Tucurui illustrate the changing nature of can result but generally require supplemen- species composition and fish production in tal hatchery programmes and the introduc- the downstream, reservoir and upstream tion of coldwater invertebrates to serve as areas. The experimental catch data docu- food for these fish.41 ment that the number of species found in

The Report of the World Commission on Dams 85 Chapter 3

each of the three areas has declined signifi- years, with the result that the total fishery cantly (by 30–50 species) following im- (upstream, downstream and in the reservoir) poundment. The data suggest that in total has tripled in size to 4 700 tonnes per year 11 species are no longer found in these areas since the dam was created. (see Figure 3.5a). Within the reservoir and to a lesser degree upstream, species now Mitigation or compensation measures have include more piscivorous species at the been used to reduce the impacts of changes expense of the detritivores that were more in fisheries. Fish passages are the most common prior to construction. In produc- prevalent measure and have been of limited tion terms, the harvest upstream of the applicability and usefulness (see Box 3.5). reservoir remained stable for the first 10 Compensation measures consist of fish years or more, but now appears to be in- hatcheries and stocking programs designed creasing (see Figure 3.5b). Meanwhile, the to reproduce the productivity of the fishery. downstream fishery has shown a continued downward trend. However, the reservoir Ecosystem Enhancement fishery has expanded tenfold in the last 20 The WCD Knowledge Base provides a number of examples of the ecosystem Figure 3.5 Decline in species numbers but increase in fisheries enhancement effects of large dams. The productivity, Tucurui (a&b) Case Studies show, for example, that productive wetlands have been created by 200 Pre-impoundment pumping Grand Coulee water through a Post-impoundment previously dry area in the Columbia River

150 Basin, and along the shores of Lake Kariba, with considerable wildlife and tourism values resulting. 100

Some reservoirs support globally threatened 50 Number of species reptiles (for example Hillsborough dam, Trinidad), and others have been declared 0 Ramsar sites of international importance for (a) Downstream Reservoir Upstream Total birds. Indeed, one measure of the environmental value of water bodies is to consider 3 500 the list of sites designated as internationally Downstream 3 000 important for waterfowl under the Ramsar Reservoir Convention on Wetlands. Of 957 sites 2 500 Upstream designated by December 1998, only 10% 2 000 included artificial wetland types, while 25% 1 500 included natural lake types.42 Many of the 1 000 designated artificial wetlands are dammed

Annual fish harvest (tons) sites: of the almost 100 artificial wetlands 500 designated as internationally important, 78 0 include water storage areas.43 (b) 1981 1988 1989 1998 A study by Wetlands International for Source: WCD Tucurui Case Study. WCD showed that the wintering waterfowl

86 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

assembled on natural and dammed lakes in In general, deeper reservoirs that Switzerland are broadly similar, and the have steep sides or high seasonal same species occur most abundantly on both water-level fluctuations are unlikely types of lake. The study also showed that to support major wetland habitats. the situation in South Africa is very different, as it lacks natural permanent water Dammed lakes support a more bodies and almost all permanent water restricted range of species, and bodies are reservoirs. At least 12 reservoirs several common and uncommon support major and important concentrations species from natural lakes were not of waterbirds. Large dams in South Africa recorded on dammed lakes by have provided generally beneficial condi- Wetlands International. The dam- tions for pelicans, darters and cormorants. ming of rivers has increased the They provide suitable moulting sites for number of open water sites available waterfowl: for example, at least 70% of the to wintering waterfowl in Switzer- global population of the South African land and provided a more suitable Shelduck, Tadorna cana, moults at only 23 habitat for these birds than the localities in South Africa, 21 of which are generally fast-flowing stretches of large reservoirs. Dam reservoirs provide dry- river in-between. However, these season or drought refuges for many water- sites support only relatively small fowl species in the semi-arid parts of the numbers of birds, of mostly common country.44 and widespread species and do not appear to provide as diverse a waterfowl habitat as the But productive wetlands are most likely to natural lakes in the area. Eleven of the be created around reservoirs where these are 2 596 British reservoirs support substantial shallow or have shallow margins and limited and important wintering waterfowl popula- reservoir drawdowns. Where sediment tions, while 60 natural inland waters and 52 inflows from the catchment are heavy, small estuaries support populations of internation- deltaic wetlands may evolve at the inflow. al importance.45

Figure 3.6 Fragmentation in 225 large river basins

All Basins 37% 23% 40%

Southern Basins 31% 28% 41%

Northern Basins 42% 19% 39%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Strongly affected Moderately affected Unaffected

Source: Revenga et al, 2000.

The Report of the World Commission on Dams 87 Chapter 3

Cumulative Impacts The WCD Knowledge Base documents a number of cumulative impacts that include Many of the major catchments in the world water quantity, water quality and species now contain multiple dams. Within a basin, impacts. Flood regimes are clearly affected as the greater the number of dams, the greater increasing the total storage volume by the fragmentation of river ecosystems. An adding additional dams reduces the flood estimated 60% of the world’s large river flows downstream. basins are highly or moderately fragmented by dams (see Figure 3.6). The magnitude of In Pakistan, the Tarbela Case Study reveals river fragmentation can be very that only 21% of the historical dry season high. In Sweden, for example, flow of the Indus reaches the delta; the rest Problems may be only three major rivers longer is diverted for irrigation and water supply by magnified as more large than 150 km and six minor 22 dams and barrages,. Since the Kotri dams are added to a rivers have not been affected barrage was commissioned in the early river system, resulting in by dams.46 1960s, the average number of days with no an increased and river flow downstream in the dry season cumulative loss of Although seldom analysed, increased from zero to 85 (the average from natural resources, cumulative impacts occur 1962 to 1997). Similar impacts have oc- habitat quality, when several dams are built on curred around the Aral Sea (see Box 3.7) environmental a single river. They affect both and in Australia where 80 years of river sustainability and the physical (first-order) ecosystem integrity. variables, such as flow regime regulation, construction of additional and water quality, and the storages, and diversion of water from the productivity and species Murray Darling River have reduced the composition of different rivers. The prob- median flow reaching the sea to 21% of the 47 lems may be magnified as more large dams pre-regulated flow. are added to a river system, resulting in an Water quality parameters recover only increased and cumulative loss of natural slowly when water is released from a dam. resources, habitat quality, environmental Oxygen levels may recover within a kilome- sustainability and ecosystem integrity. The tre or two, while temperature changes may cumulative impacts of interbasin water still exist 100 km downstream. Where the transfers can be of special concern, as this distance between dams does not allow often involves the transfer of species into recovery to natural levels, the biology of new watersheds. many hundreds of kilometres of river may be affected by a handful of dams. Examples Box 3.7 Cumulative impact of dams: the Aral Sea from the WCD Case Studies include the The Aral Sea, fed by the Amu Darya and Syr Darya, was once the fourth largest Orange-Vaal river in South Africa, where inland body of water in the world, ranking just behind Lake Superior. It supported 24 species of fish and a fishing population of 10 000 people. A series the impacts of 24 dams may have led to of dams was built on the rivers to feed an immense irrigation system and grow 2 300 km (63%) of the river having a cotton on 2.5 to 3 million hectares of new farmland. The withdrawal of water has reduced the Aral Sea to about 25% of its 1960 volume, quadrupled the salinity modified temperature regime. On the of the lake and wiped out the fishery. Pollutants that had formerly fed into the Columbia River, Grand Coulee dam re- lake became airborne as dust, causing significant local health problems. The environmental damage caused has been estimated at $1.25 to $2.5 billion ceives water that is already high in total annually. dissolved gasses as a result of upstream Source: Anderson, 1997, Section 1 pii, Section 6 pii. Canadian dams. Before the levels can recover to natural values, spill at Grand

88 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

Coulee increases them again, passing the Anticipating and problem further downstream.48 Construc- predicting ecosystem Global sub-sample: tion of a series of dams may therefore have impacts 87 dams increasing impacts on downstream ecosys- In order for ecosystem impacts to Before 1950 (3 dams) tems and biodiversity. be addressed properly, they have 16.7% to be understood and predicted. Also on the Columbia River, the cumulative The Cross-Check Survey found impact of an additional dam on salmon 83.3% that for the 87 projects that migrations is significant. It is estimated that provided data on ecosystem 5–14% of the adult salmon are killed at each impacts, almost 60% of the 1950s (8 dams) of the eight large dams they pass while 49 impacts identified were unantici- swimming up the river. 20.9% pated prior to project construction, largely due to inadequate What is not well researched is the change in 79.1% the magnitude of the incremental response studies. While the sample size is small for some time periods, the of ecosystem function and biodiversity as a Cross-Check also suggests that river is increasingly fragmented. Thus it is 1960s (17 dams) over time the trend is increasing- not known if there is some threshold level at 33.3% which the marginal impacts of the addition ly to anticipate impacts (see Figure 3.7). This confirms the of one or more dams to a particular cascade 66.7% of dams will begin to decline. It is therefore expectation that the trend towards the use of environmental a case by case call whether the ecosystem impacts of further modifying a particular impact assessments (EIA) would result in improved identification 1970s (30 dams) river may at some point be of less conse- of potential impacts (see Chapter quence than, for example, putting the first 38.6% 6 for discussion of EIAs). dam on a free-flowing river. 61.4% Anticipating an impact is, Anticipating and however, not synonymous with

Responding to Ecosystem accurately predicting the direc- 1980s (17 dams) Impacts tion and magnitude of its effect on ecosystems and biodiversity. 59.4% Examination of efforts to counter the Nor does it guarantee understand- ecosystem impacts of large dams in the ing of the further impact of such 40.6% WCD Knowledge Base indicates that they changes on the livelihoods and have met with limited success owing to: economic welfare of affected ■ the lack of attention paid to anticipating people. While the generalised 1990s (12 dams) and avoiding impacts; impacts of reservoir creation on 64.3% ■ the poor quality and uncertainty of terrestrial ecosystems and biodi- predictions; versity are well-known, specifics, 35.7% ■ the difficulty of coping with all impacts; and Figure 3.7 Anticipated and Anticipated ■ partial implementation and success of unanticipated ecosystem impacts Unanticipated mitigation measures. Source: WCD Cross-Check Survey.

The Report of the World Commission on Dams 89 Chapter 3

such as the net emissions of time, most ecosystem impacts remain site- greenhouse gases from a specific. Their exact nature cannot be particular dam site, cannot be predicted in the absence of appropriate field predicted with any certainty at studies of individual river systems. present. Further research into the factors determining net Avoidance, minimisation, emissions may reduce this mitigation, compensation and uncertainty over time. ecosystem restoration The WCD Knowledge Base reveals that Downstream impacts on efforts to avoid or minimise impacts through aquatic ecosystems and biodi- choice of alternative projects or alternative versity and on floodplain designs were more successful than efforts to ecosystems represent the sum manage the impacts once they were built of many complex interactions into the design of the dam. Avoidance and and thus are inherently diffi- minimisation of impacts, by their very cult to predict where baseline data are nature, reduce ecosystem impacts on the site absent or unreliable. However, the overall concerned. But where alternative sites or direction of the impacts is generally nega- designs have been chosen, the net conse- tive. As shown for the case of floodplain quences for ecosystems have rarely been effects, the impact of large dams on these recorded. ecosystems will vary. With regard to fisheries, while it appears that the effect on Project planners and proponents have species composition is generally negative at employed five principal measures to respond all levels (upstream, reservoir and down- to ecosystem impacts: stream), downstream losses in productivity may be accompa- ■ measures that avoid the anticipated nied by increases in reservoir adverse effects of a large dam through Downstream impacts on fishery production. Finally, the the selection of alternative projects; aquatic ecosystems and nature of cumulative impacts as ■ measures to minimise impacts by altering biodiversity are additional dams are added to a project design features once a dam is inherently difficult to river system may be significant, decided upon; predict where baseline but a lack of research on the data are absent or ■ mitigation measures that are incorporat- topic makes predictive assess- unreliable. ed into a new or existing dam design or ment difficult. operating regime in order to reduce ecosystem impacts to acceptable levels; In sum, past anticipation and ■ prediction of ecosystem impacts was limited, measures that compensate for unavoidable residual effects by enhancing ecosys- in part due to a lack of reliable baseline tem attributes in watersheds above dams data, scientific uncertainty regarding the or at other sites; and nature of the interactions, inadequate attention paid to these issues and a corre- ■ measures to restore aspects of riverine spondingly limited ability to model these ecosystems. complex systems. While improvements in measurement, scientific understanding and The primary option for avoiding ecosystem modelling capability have occurred over impacts from large dams has been not to

90 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

build the dams in the first place. This is While there are cases of good Good site selection, such given a legal basis in Austria, Finland, mitigation, the success is never- as not building large France, Norway, Sweden, Switzerland, the theless contingent upon stringent dams on the main-stem United States and Zimbabwe, where legal conditions including: of a river system, and ‘set-aside’ provisions to protect particular ■ a good information base and better dam design also river segments or basins from regulation or competent professional staff played significant roles in development have been established.50 available to formulate com- avoiding or minimising plex choices for decision- impacts. Good site selection, such as not building makers; large dams on the main-stem of a river ■ an adequate legal framework system, and better dam design also played and compliance mechanisms; significant roles in avoiding or minimising ■ a co-operative process with the design impacts in a number of cases in the WCD team and stakeholders; Knowledge Base.51 The International Energy Agency also supported such policies ■ monitoring of feedback and evaluation of in its recent policy paper Hydropower and mitigation effectiveness; and Environment.52 As reported earlier, an ■ adequate financial and institutional increasing number of countries are using resources.53 environmental flow requirements to minimise downstream impacts (see Box 3.4) If any one of these conditions is absent, sometimes in the form of managed floods mitigation is unlikely to succeed. Mitiga- (see Box 3.6). tion, though often possible in principle, presents many uncertainties in field situa- Mitigation was the most widely practised tions and is therefore at present not a response to ecosystem impacts for the large credible option in all cases and all circum- dams in the WCD Knowledge Base. As stances. In addition, the weaknesses of the noted earlier, mitigation has failed or EIA process for many projects reduces the worked only sporadically in the case of possibility of positive outcomes.54 This wildlife rescue operations and fish passes supports the use of alternative strategies (Box 3.1 and Box 3.5). In the Cross-Check rather than simply one of mitigation. sub-sample of 87 projects for which ecosystem impacts were recorded, mitigation was Compensation for lost resources may be ‘in- undertaken for less than one-quarter of the kind’ (for example construction of a fish anticipated ecosystem impacts (10% of all hatchery for lost fish spawning areas) or ecosystem impacts that occurred). Of these ‘out-of kind’ (for example watershed protec- projects, 47 also recorded the effectiveness tion in the upper catchment for loss of of mitigation measures implemented. riverine or wetland habitat). Compensation Respondents stated that about 20% worked may also be paid ‘in-basin’ (for example effectively, 40% did not mitigate the impact, restoration of forest area within the river and 40% were moderately effective. The basin for forest lost to inundation) or ‘out- conclusion can be drawn that only a small of-basin’ (for example assistance in expand- percentage of ecosystem impacts that ing management capability at similar occurred were actually mitigated effectively, locations in another river basin). These are while the relative significance of these applied to offset ecosystem and biodiversity impacts remains unknown. loss, as well as to replace lost productive use

The Report of the World Commission on Dams 91 Chapter 3

of natural resources (as with fish Ecosystem impacts of decommissioning are hatcheries). Concerns with the also complex and site-specific. One major effectiveness of compensation issue in dam decommissioning is what to do include questions about the with possibly polluted sediment accumulat- possibility of ‘replacing’ ecosys- ed behind the dam. The fate of this sedi- tem functions and species (for ment when the dam is removed is frequently example, are fish raised in a a major obstacle to restoration. hatchery equivalent to native fish stocks?) and the conse- Current large dam designs are often not quences of such efforts, for sufficiently flexible to allow for changed example whether fish hatcheries operating regimes to meet environmental actually damage native fish (or other) goals. Global experience shows stocks through disease and that these long-lived structures may be competition. called on to operate differently in the future than in the past as society’s needs and values Restoring ecosystem evolve and as other dams are added in the through catchment area. decommissioning In some cases, the dam design is completed before the environmental flow needs are Ecosystem restoration has been determined, and cannot accommodate water undertaken in a range of countries where evolving national releases of the required quantity and quality. legislation has required higher Five dams on the Colorado River have now standards of environmental performance been retroactively fitted with variable level (see Box 3.8). In the United States and offtakes to draw off surface water, increase France, dams have been decommissioned to the temperature of the downstream river, 56 restore key environmental values, often and satisfy the needs of native fish. related to migratory fish (salmon), and often as a condition of project relicensing.55 Findings and Lessons Large dams generally have extensive im-

Box 3.8 Ecosystem restoration through decommissioning in the pacts on rivers, watersheds and aquatic United States ecosystems. From the WCD Knowledge

A total of 467 dams have been removed to date in the United States, 28 of these Base it is clear that large dams have led to: are large dams higher than 15 metres. Reasons for removal have included safety ■ the loss of forests and wildlife habitat, concerns, the restoration of riverine fisheries, financial considerations, or removal of unauthorised structures. the loss of species populations and the One example of a removal is the Grangeville dam on Clearwater Creek, Idaho. degradation of upstream catchment areas Built in 1903, it housed a 10-MW power plant. The removal was motivated by due to inundation of the reservoir area; excessive sedimentation in the reservoir and blockage of migratory fish following the collapse of the fish pass in 1949. The dam was removed in 1963, ■ emissions of greenhouse gases from and the river washed out the accumulated sediment within six months, with no recorded downstream effects. Removal restored access for salmon and reservoirs due to the rotting of vegeta- steelhead runs to 67 km of main stem river and over 160 km of tributary habitat tion and carbon inflows from the basin; in the upper reaches of the Clearwater River. It also allowed members of the Nez Perce tribe to regain a traditional fishery long denied them despite the ■ the loss of aquatic biodiversity, upstream provisions of the 1855 treaty with the United States. and downstream fisheries and the servic- Source: Bowman et al, 1999, pxix, 27–31. es of downstream floodplains, wetlands

92 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

and riverine estuarine and adjacent al ecosystems and biodiversity, and marine ecosystems; efforts to ‘rescue’ wildlife have met with ■ the creation of productive fringing little sustainable success; wetland ecosystems with fish and water- ■ the use of fish passes to mitigate the fowl habitat opportunities in some blockage of migratory fish has had little reservoirs; and success, as the technology has often not ■ cumulative impacts on water quality, been tailored to specific sites and species; natural flooding, and species composi- ■ good mitigation results from a good tion where a number of dams are sited on information base, early co-operation the same river. between ecologists, the dam design team and affected people, and regular monitor- The ecosystem impacts are more negative ing and feedback on the effectiveness of than positive and they have led, in many mitigation measures; cases, to irreversible loss of species and ■ environmental flow requirements (which ecosystems. In the Cross-Check Survey 67% include managed flood releases) are of the ecosystem impacts recorded were increasingly used to reduce the impacts negative. The social consequences of of changed streamflow regimes on environmental impacts are examined in aquatic, floodplain and coastal ecosys- Chapter 4. tems downstream; and ■ avoidance or minimisation of ecosystem Efforts to date to mitigate the ecosystem impacts can be achieved through legisla- impacts of large dams in the WCD Knowl- tive or policy measures that set-aside edge base have met with limited success particular river segments or basins, or owing to the lack of attention given to through good site selection (such as anticipating and avoiding impacts, the poor avoiding main stem dams). quality and uncertainty of predictions, the difficulty of coping with all impacts and the Finally, a number of countries, particularly only partial implementation and success of the United States, are making efforts to mitigation measures. More specifically: restore ecosystem function and native fish ■ it is not possible to mitigate many of the populations by decommissioning large and impacts of reservoir creation on terrestri- small dams.

The Report of the World Commission on Dams 93 Chapter 3

Endnotes 15 Field measurements from Rosa et al, 1999. 16 Fearnside, 2000, develops a mathematical 1 ICOLD, 1981 and 1988 cited in WCD model. Thematic Review II.1 Ecosystems; IEA, 17 Emissions per km2 are converted to emis- 2000, cited in WCD Thematic Review II.1 sions/TWh using the mean annual genera- Ecosystems; see also other papers in WCD tion 1995-1999. cited Thematic Review II.1 Ecosystems. 18 Dietrich, 1999 env082, WCD Submission. 2 Revenga et al, 2000. 19 Holden and Stalnaker, 1975, p217, 229. 3 Respondents to the Cross-Check Survey found that 67% of the recorded ecosystem 20 Walker, 1979, p156-57. impacts were negative. 21 Thomas, 1998, p2. 4 WCD Thematic Review II.1 Ecosystems, 22 WCD Thematic Review II.1 Ecosystems, section 3. section 3.6.2.2. 5 Where there is a loss of vegetation cover 23 Collier et al, 1996, p56-58. there will be an increase in annual yield, but 24 Abdel Megeed and Aly Makky, 1993, p298; the direction of change in dry season flows Stanley and Warne, 1993, p628, 630. will depend on the balance between eva- 25 Bourke, 1988, p117. potranspiration and infiltration effects 26 Balland, 1991. (Bruijnzeel, 1990). In most cases it is expected that the evapotranspiration effect 27 Hynes, 1970, p422-423. will dominate leading to lower dry season 28 Jackson and Marmulla, 2000, Contributing baseflow (Lamb and Gilmour, 2000). Paper for WCD Thematic Review II.1 6 The WCD Thematic Review II.2 Global Ecosystems, p12–13; Larinier, 2000, Contrib- Change and a WCD Workshop on the topic uting Paper for Thematic Review II.1 provide reviews of the literature and the Ecosystems, pii. perspectives of scientists working in this 29 Hubbs and Pigg, 1976, p115. field. 30 Walker, op cit, p152. 7 The authors stress that the large range 31 Furness, 1978. underscores the need for further quantifica- 32 For example: Hadejia Nguru in Hollis et al, tion in order to better understand the 1993. contribution of reservoirs to global GHGs. St Louis et al, in press. 33 WCD Thematic Review I.1 Social Impacts, Annex I. 8 Bosi, 2000, p12. 34 Kudlavicz, 1999 env129, p3 and 2000 9 WCD Thematic Review II.2 Global Change, env063, p1, WCD Submissions. Executive Summary, pv. 35 Jackson and Marmulla, op cit, p8. 10 WCD Thematic Review II.2 Global Change; 36 Welcomme, 1976, p361. IEA, op cit. 37 Benech, 1992, p161; Jubb, 1972; Lowe- 11 To make the comparison with the thermal McConnell, 1985, p120. alternatives requires the measured reservoir emissions to be converted to emissions per 38 Lower Volta in Adams, 1992, p145-146; kWh generated (see Box 3.2). Kassas, 1973; Gammelsrød, 1996, p120. 12 Values for methane are converted to a carbon 39 Aleem, 1972, p205; Drinkwater and Frank, dioxide equivalent using a Global Warming 1994, p141. Potential of 21 and expressed as grams of 40 Gammelsrød, op cit, p123.

CO2 equivalent (IPCC, 1996, cited in WCD 41 Jackson and Marmulla, op cit, Executive Thematic Review II.2 Global Change). Summary piv. 13 WCD Thematic Review II.2 Global Change. 42 Frazier, 1999, p17-18. 14 Ibid. 43 Ramsar Convention Database, 1999.

94 Dams and Development: A New Framework for Decision-Making Ecosystems and Large Dams: Environmental Performance

44 Davidson and Delany, 2000, Contributing 50 WCD Thematic Review II.1 Ecosystems, paper for WCD Thematic Review II.1 section 4.2.1. Ecosystems, p4, 13. 51 WCD Thematic Review II.1 Ecosystems, 45 Ramsar Convention Database, op cit; Bridle section 6.2. and Sims, 1999, p3. 52 IEA, op cit, p27-29. 46 Lovgren, 1999 env136, WCD Submission, 53 WCD Thematic Review II.1 Ecosystems, p2, 8. Executive Summary, p xii. 47 Crabb, 1997, p42. 54 WCD Thematic Review V.2 Environmental 48 This problem is lessened when the water all and Social Assessment, section 2.5. passes through the turbines, and there is no 55 Bowman et al, 1999, Executive Summary p xii; spill. Epple, 2000 opt136, WCD Submission, p3. 49 Eley and Watkins, 1991, p21. 56 US Bureau of Reclamation, 2000a.

The Report of the World Commission on Dams 95 People and Large Dams – Social Performance

Chapter 4 People and Large Dams – Social Performance

he social impacts of large Tdams – their impact on people’s livelihoods, health, social systems, and cultures – are an integral part of their performance record.

Dams are one of a series of public infrastructure projects aimed at the economic development of a region,

nation, or river basin. The direct benefits they provide to people are typically reduced to monetary figures

for economic analysis and are not often recorded in human terms. In addition, simply accounting for these

direct benefits often fails to capture the full social benefits associated with

The Report of the World Commission on Dams 97 Chapter 4

providing water, electricity, and share of the economic benefits. Where the flood control, and including any broader costs and benefits – economic, indirect economic benefits or environmental and social – fall unequally multiplier effects. within society, decision-making on project appraisal and selection based simply on At the same time, however, summing up the positives and negatives is dams have negatively affected inadequate on equity grounds. At the same many people and societies. This time it is clear from the emerging experience is clear throughout the WCD with good practice on benefit sharing mecha- Knowledge Base, most poignant- nisms and reparations detailed in the Knowl- ly through many of the presenta- edge Base that a continuation of the legacy of tions made by dam-affected inequity associated with many large dams is peoples in the WCD Regional not only unacceptable, but unnecessary. Consultations and the non- governmental organisation This chapter uses the WCD Knowledge (NGO) hearings in Europe and Base to present a number of issues of strate- South Africa. Globally, the gic importance surrounding the social overall magnitude, extent and impacts of large dam projects, particularly complexity of these adverse social impacts those that underpin the growing worldwide for the displaced and for those dependent on opposition to dams by affected communities. the riverine ecosystem – both upstream and To acknowledge the costs in this regard is downstream from a dam – are of such not to deny the benefits. However, in order significance as to merit detailed considera- to provide a foundation for the way forward, tion in any assessment of the rationale for it is essential to understand the extent, dam construction. Further, it is apparent diversity, and range of social impacts – that these impacts are – even today – often particularly the adverse impacts. This not acknowledged or considered in the understanding must extend to all manifesta- planning process and may remain unrecog- tions and implications for various popula- nised during project operations. Where tion groups; the extent to which such measures are put in place to mitigate im- impacts were addressed; and the outcome of pacts on affected people they typically fail to mitigation, resettlement and development address adequately the problems caused by measures. the decision to build a large dam. The chapter begins with a brief overview of Just as with the economic impacts of large social impacts during the planning and dams, the social and environmental impacts project cycle, with an emphasis on the social of dams can be classified as gains or losses costs and benefits, as well as the indirect accruing to different social groups – now economic benefits, that accompany dam and in the future. Analysis of the WCD projects and the services they provide. The Knowledge Base, and in particular the impacts on people displaced from their WCD Case Studies, indicates that the poor, homes and livelihoods, indigenous peoples, vulnerable groups and future generations are downstream communities, gender, human likely to bear a disproportionate share of the health and cultural heritage are then social and environmental costs of large dam explored at length. Analysis of the costs and projects without gaining a commensurate benefits of large dam projects and their

98 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

distribution across different groups in society sustained by the business generated by the leads to conclusions on equity and the planning process, including contractors, adequacy of the ‘balance sheet’ approach of consultants and workers employed on the evaluating large dams. Initiatives addressing project. In the case of developing countries, equity concerns are then described, followed particularly smaller countries with a limited by conclusions for the chapter. ‘dam industry’, foreign consulting firms have often undertaken the preparatory studies.1 Socio-economic Impacts through the Planning and Construction Project Cycle During the construction period, dam projects require At the planning and design a large number of unskilled stage, an important social Given the different types, sizes and locations workers and smaller but impact is the delay of large dams in the WCD Knowledge Base, significant amounts of skilled between the decision to generalisations on the socio-economic labour. New jobs are therefore build a dam and the onset impacts of dams are limited and, in many created both for skilled and of construction. This can cases the nature or significance of the unskilled workers during the result in communities living impacts are contested. This section briefly construction phase. The for decades starved of illustrates and characterises the social costs WCD Case Studies report that development and welfare and benefits, and the indirect economic Kariba and Grand Coulee investments. costs that are associated with the planning employed between 10 000 and and project cycle, with particular attention 15 000 workers each. During to those that accompany the provision of the peak construction period, a labour force dam-related services. of about 15 000 people was employed at Tarbela, helping to build local capacity for Planning and design subsequent national development projects. At the planning and design stage an impor- While skilled labour is typically drawn from tant social impact is the delay between the the national labour market, international decision to build a dam and the onset of contractors are often involved at this stage construction. Dams are often discussed years as well. The construction of dams and their before project development is seriously associated infrastructure bring significant considered and once a site is identified a benefits to the employees and shareholders form of ‘planning blight’ can take place, of companies engaged in construction and making governments, businesses, farmers the supply of equipment and materials. and others reluctant to undertake further productive investments in areas that subse- The beneficial effect on local communities quently might be flooded. Communities can is often transient due to the short-lived, live for decades starved of development and pulse impact of the construction economy welfare investments. A related problem is on dam construction sites. Careful planning the fear felt by many people living in a may, however, enhance the ‘boom’ phase possible reservoir area. Such psychological and lead to long-lasting benefits. Roads, stress cannot be effectively quantified in power lines, social services and other economic terms, but it is a real issue. At this infrastructure installed through the building stage, project beneficiaries include those of a dam provide access to previously

The Report of the World Commission on Dams 99 Chapter 4

inaccessible areas, allowing output and ignore the use of water for settlement and connecting local horticulture, livestock and fish production, economies to national markets. as well as domestic water supply.2 This has both positive and negative social impacts. Exposed How much of a stimulus irrigation projects to the national economy, indige- have provided to wider regional develop- nous and vulnerable groups find ment is a complex issue that needs more their lands and livelihoods explicit recognition during project formula- threatened by forces beyond tion to ensure that the enabling framework their experience or control. is in place to encourage growth. In this Similarly, existing settlements at regard, regional multiplier effects are useful construction sites have found indicators of the distribution of project themselves subject to increased benefits to a project region. However, they health problems (including represent a gain in the economic welfare of malaria, sexually transmitted the nation only when excess capacity diseases, and HIV-AIDS) and a (unemployed resources) exists. Where it loss of social cohesion with the does not exist they simply represent a large influx of outsiders. transfer (or re-distribution) of resources from one region of the economy to another. Irrigation The role of dams in improving nutrition and As with livelihood enhancement, the food security is contested. Increases in food broader impacts of irrigation projects on production from irrigated agriculture may rural and regional development were often lead to both income and price effects as not quantified. Dams, along with other farming households in irrigated areas in- economic investments, generate indirect crease their purchasing power and the price economic benefits as expenditure on the of staple foodstuffs falls for urban (and other project and income derived from it lead to non-farming) households. For these house- added expenditure and income in the local holds irrigated agriculture and, implicitly, or regional economy. The WCD Case large irrigation dams are likely to have Studies give examples of these ‘multiplier’ contributed to greater food security and benefits resulting from irrigation projects. In improved nutrition at the household level. the case of Grand Coulee, agribusiness and local communities prospered due to irrigated At the national level, nutritional levels production yielding multiplier effects of increased over the 25 years from 1970 by 1.5-1.7. Similarly, the Aslantas dam spun off 14% in India and 30% in China.3 These projects that led to a tripling in agricultural two countries are two of the largest builders processing and machinery manufacturing for of irrigation dams. However, the actual the area. Beyond these economic impacts, extent of the contribution of large dams to irrigation schemes also produce a series of these improvements is difficult to deter- indirect social benefits derived from the – mine. The India Case Study calculates that typically unanticipated – multiple use of the share of the total increase in food irrigation water. Estimates of project bene- production from 1950-1993 attributable to fits usually rely simply on the expected crop additional land brought under irrigation is

100 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

10%. The study does not assess the extent of experience in the informal settlements productivity gains derived from access to (favelas) in São Paulo, Brazil illustrates the irrigation water. India’s Central Water social and environmental benefits electricity Commission puts the share at 30%. In the services can bring to people (see Box 4.1). past 50 years India achieved a marginal As in Sao Paulo, policy measures are in- surplus in terms of per capita food availabili- creasingly used to bring electricity to poorer ty. In percentage terms India also saw a sections of the population. The slum areas in decrease in the share of the rural population Mumbai in India, Manila in the Philippines, below the poverty line. However during this and other cities and towns in the developing period the absolute number of people below world are further examples of such efforts. the poverty line – that is without capacity to buy food – increased from 180 million to Employment 4 300 million. Again, the precise impact of In terms of generating employment, the dams on these numbers is not known. principal impact of large dam projects – aside from construction jobs – arises from National statistics mask significant local the new productive enterprises allowed by variation. Of concern is that people in food- the provision of water or electricity. The deficit areas are disadvantaged in terms of Case Studies provide a number of examples access to basic food grains and pay higher of anticipated and unanticipated employ- prices than those in food-surplus areas. ment generation (also see Table 4.1, p121). Contributing to these concerns is the tendency of large irrigation schemes to lead to the production of more cash crops than envisaged at planning and less food crops (as Box 4.1 Bringing electricity to the favelas in São Paulo, Brazil indicated in Chapter 2). Producing higher Between 1973 and 1993, São Paulo’s favelas swelled from about 700 000 value cash crops is a rational decision for inhabitants to over 2 million. Initially, the squatters’ shacks had only sporadic and illegal electrical connections, partly because the electric utility had no commercial farmers who may choose to procedure for electrification of such structures and partly because municipal purchase foodstuffs. Once poorer farmers authorities thought that improvement of the favelas would condone illegal occupation of the land. achieve household food self-sufficiency, they By 1979, the city and the electric utility, Electropaulo, came to an agreement and will probably choose to market surplus connected some favelas to the grid using simplified installation kits and no produce and, therefore, shift to higher value meters. Consumers were billed a flat rate, which was subsidised, for a minimal monthly consumption of 50 kWh – enough to run a couple of lamps and a radio crops. The concern is that those people who or other domestic appliance. The cost of metering was considered too high for do not participate in the irrigation project such low usage levels. or are otherwise marginalised due to dam By 1983, some 100 000 shacks were connected, and the quality of life improved. Better lighting simplified the tasks of cleaning and maintaining the shack and construction may face higher food prices caring for children and sick people. Without smoke from candles and kerosene and decreased food security as a result. lamps health improved. People started to use TV sets, irons and refrigerators. Where water services were provided, electric showers also became more common. For the squatters, an important benefit was receiving bills with their name and address, which gave them a certain social recognition as well as Hydropower access to credit. New energy services provided by dams have A decade later, electricity consumption per shack had increased to 175 kWh. Many of the dwellings had been greatly improved, and services were more benefited urban populations and others reliable. Some demographers attribute the strong decline in Brazil’s population connected to power distribution systems. growth rate – from 3.8% per year in 1970 to 1.4% today – to the adoption of new cultural values that spread partly through television, which electricity made Typically, in countries with low levels of available. energy services, even small energy inputs Source: Boa Nova and Goldemberg, 1999. bring significant welfare improvements. The

The Report of the World Commission on Dams 101 Chapter 4

In the case of the Aslantas dam, an increase Displacement of People in employment in farming was projected. It and Livelihoods did not materialise due to a shift to less labour-intensive crops, mechanisation and a Many development interventions to trans- general migration to urban areas. In the form natural resources, particularly large- Orange River Development Project, al- scale infrastructure projects - involve some though farm jobs decreased throughout form of displacement of people from their South Africa from 1960 through the 1980s livelihoods and homes. Large dams are (the latest period for which data are availa- perhaps unique amongst such projects in ble), they dropped less in the command area that they can have widespread and far- of the project. When compared to national ranging ecosystem impacts due simply to the trends the loss of at least 4 000 regular jobs blocking of a river. The result is a series of was avoided and effectively some terrestrial, aquatic and riparian impacts that 16 000 seasonal jobs were not only affect ecosystems and biodiversity As with any indirect created in downstream areas by but also have serious consequences for economic impact, it is the dams and associated irriga- people who live both near and far from the important to consider tion development.5 The employ- dam site. A large, multi-functional resource not just the gross ment impacts accrued largely to base like a river and its surroundings is number of jobs created Black Africans and the ‘Col- characterised by a complex web of diverse, by a project but also oured’ racial groups who interconnected, implicit and explicit whether alternative represented 97% of paid farm functional roles, dependencies and interac- uses of project workers. tions. Consequently the social and cultural resources would implications of putting a dam into such a generate similar gains. Employment gains are also landscape are spatially significant, locally engendered by hydropower disruptive, lasting and often irreversible. production and other services provided by reservoirs. In the case of Tucu- Large dams have significantly altered many rui and Grand Coulee, a sizeable percentage of the world’s river basins, with disruptive, of the power produced goes to industries in lasting and usually involuntary impacts on the respective regions. Similarly, the crea- the livelihoods and socio-cultural foundation of commercial and sport-fishing indus- tions of tens of millions of people living in tries, as well as reservoir-based recreation these regions. The impacts of dam-building and tourism, has led to job creation at on people and livelihoods – both above and Grand Coulee, Tucurui and Kariba. Inland below dams – have been particularly devas- navigation can also provide substantial tating in Asia, Africa and Latin America, employment. The Panama Canal, based on where existing river systems supported local two large dams, directly employs 8 000 economies and the cultural way of life of a people on Canal operations and creates jobs large population containing diverse commu- in the local shipping services industry and nities. duty-free zone.6 As with any indirect economic impact, it is important to consider Displacement is defined here as referring to not just the gross number of jobs created by both ‘physical displacement’ and ‘livelihood’ a project but also whether alternative uses of displacement (or deprivation). In the project resources would generate similar narrow sense displacement results in the gains. physical displacement of people living in the

102 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

reservoir or other project area. This occurs Scale of physical displacement not only from the inundation of reservoirs The WCD Knowledge Base confirms that but from the installation of project facilities there are many dams that have caused and associated infrastructure. The WCD physical displacement – and indeed that Knowledge Base records that all too often large dam construction has physically this physical displacement is involuntary displaced tens of millions of people world- and involves coercion and force – in a few wide in the last half century. The scale and cases even killing. extent of impacts will vary depending on location, size and other dam characteristics However, the inundation of land and such as inundated area, and population alteration of riverine ecosystems – whether density in the river basin. In the eight WCD upstream or downstream – also affects the Case Studies, the only one without any resources available for land- and riverine- physical displacement was the chain of dams based productive activities. In the case of in the Glomma and Laagen Basin. In the communities dependent on land and the Cross-Check Survey, physical displacement natural resources base, this often results in is reported in 68 of the 123 dams (56%). Of the loss of access to traditional means of the dams in this sample, 52 out of 68 dams livelihood, including agricultural produc- are in Latin America, Asia, and sub-Saharan tion, fishing, livestock grazing, fuelwood Africa. Large dams on the main-stem of a gathering and collection of forest products, to name a few. Not only does this disrupt local economies, it effectively displaces Box 4.2 Economic, socio-cultural, and health impacts of livelihood people – in a wider sense – from access to a displacement series of natural resource and environmental Resettlement programmes have predominantly focused on the process of inputs into their livelihoods. This form of physical relocation rather than the economic and social development of the livelihood displacement deprives people of displaced and other negatively affected people. The result has been the impoverishment of a majority of resettlers from most dam projects throughout their means of production and dislocates the world. them from their existing socio-cultural According to Cernea’s Impoverishment Risks and Reconstruction model, milieu. (See Box 4.2) The term ‘affected’ displacement epitomises social exclusion of certain groups of people. It culminates in physical exclusion from a geographic territory and economic and thus applies to people facing either type of social exclusion from a set of functioning social networks. Thus, affected people displacement. face a broad range of impoverishment risks that include landlessness, joblessness, homelessness, marginalisation, food insecurity, increased morbidity, loss of common resources, and community disarticulation that result in a loss of socio- The timing of these social impacts varies, cultural resilience. depending on the proximate cause. In the The key economic risks to affected people are from the loss of livelihood and income sources such as arable land, common property resources (forests, case of loss of home and livelihood due to grazing land, ground and surface water, fisheries, and so on), and changed the filling of a reservoir, the social impacts access and control of productive resources. The loss of economic power with the breakdown of complex livelihood systems results in a temporary or are quite immediate. The implications for permanent, often irreversible decline in living standards, leading to downstream livelihoods, however, come to marginalisation. Higher risks and uncertainties are introduced when diversified livelihood sources are lost. Loss of livelihood and disruption of agricultural the fore only after completion of the dam. activity can adversely affect household food security, leading to under- At this point they may set in quickly, as nourishment. Higher incidence of diseases associated with deteriorating water quality can result in increased morbidity and mortality. High mortality rates with changes in flow and their impact on immediately following involuntary resettlement from the reservoir areas of the recession agriculture or slowly, as with Kariba and Aswan High dams are cases in point. Forced displacement tears apart the existing social fabric, leading to socio-cultural disarticulation. physical and chemical changes that are Source: Cernea, 1999; Cernea, 2000; Cernea and translated into degradation of ecosystem Guggenheim, 1993; McDowell, undated; Scudder, 1997a,b. function and loss of biodiversity.

The Report of the World Commission on Dams 103 Chapter 4

river and in densely populated dam-displaced people in China is much regions of the world will higher than the official figure, with 10 inevitably displace people. In million displaced in the Yangtze Valley the Cross-Check sample, 26% alone.8 Large dams in India displaced an of dams with a surface area estimated 16-38 million people.9 Thus, in less than 1 km2 report physi- India and China together, large dams could cal displacement compared have displaced between 26-58 million with 82% of dams over 100 people between 1950 and 1990. The level of km2 in area. Yet the Cross- displacement has increased substantially Check figure may understate after 1990 with the construction of projects the occurrence of physical such as Three Gorges in China. Among the displacement given the larger tendency projects involving displacement funded by towards systematic under-enumeration the World Bank, large dams account for discussed below. 63% of displacement.10

The overall global level of physical displace- These figures are at best only estimates and ment could range from 40 to 80 million. certainly do not include the millions dis- According to official statistics, dams have placed due to other aspects of the projects displaced 10.2 million people in China such as canals, powerhouses, project infra- between 1950 and 1990 (34% of all devel- structure, and associated compensatory opment-related displacement including that measures, such as biological reserves and so due to urban construction).7 Independent on. They also refer to physical displacement sources estimate that the actual number of only and thus do not include communities upstream and downstream of dams that have

Box 4.3 Missing numbers of affected people: Sardar Sarovar project, suffered livelihood displacement. India, and Pak Mun dam, Thailand

For the Sardar Sarovar project, the 1979 Narmada Water Disputes Tribunal gave Under-counting of the the number of displaced as 6 147 families, or about 39 700 people. The World Bank’s 1987 mission placed the total at 12 000 families (60 000 people). In 1991, displaced the project authorities provided an estimate of 27 000 families. According to three state governments, the current estimate of displaced families stands at At the planning stage, the numbers of both 41 000 (205 000 people). This number will probably increase, since 13 years after directly and indirectly affected people have full-scale dam construction began, resettlement surveys have still not been completed. The current estimate does not include at least 157 000 people frequently been under-estimated (see Box displaced by canals. Nor does it include those moved to make space for the 4.3), and there has been inadequate under- creation of a wildlife sanctuary and for the resettlement of people displaced by the dam, or the 900 families displaced in the early 1960s to make room for standing of the nature and extent of the construction site infrastructure. The nature and extent of the dam’s impact on negative impacts. In all the WCD Case downstream livelihoods were not assessed. Serious efforts to survey the affected villages and people began several years after the start of dam construction work Studies, the initial assessments of the in response to intense struggles by the affected since 1985. projects failed to account for all the affected In 1991, when construction started on the Pak Mun dam, 241 families were people. The level of under-enumeration counted as displaced. By the time construction was completed it was clear that another 1 459 households had to be relocated. The true extent of the social ranges from 2 000 to 40 000 people. Exam- impact only became evident when the impact of the dam on fisheries livelihoods ples from large dam projects in Africa was admitted in response to prolonged agitation by the affected people. By March 2000, the Thai government had paid interim compensation – pending a include the tri-national Ruzizi hydroelectric final solution to the permanent loss of fisheries livelihoods – to 6 204 house- project involving Zaire, Rwanda, and holds for livelihood loss during construction. Burundi, the Funtua dam in Nigeria, and the Source: Sardar Sarovar in Brody, 1999, Contributing Paper for WCD Thematic Review I.1 Social Impacts, Section 5.2; Supreme Court of India, 1999; Morse Kiambere reservoir on the Tana River in and Berger, 1992, p51, 89; WCD Pak Mun Case Study. Kenya, with discrepancies ranging from 1 000

104 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

to 15 000 people.11 Similar observations the loss of their lands and liveli- Often, people physically emerge from other regions.12 Among projects hoods. With such criteria for displaced by canals, funded by the World Bank, the actual number eligibility, indigenous peoples and powerhouses, and of people to be resettled was 47% higher than ethnic minorities suffer dispropor- associated compensation the estimate made at the time of appraisal.13 tionately as they may lack citizen- measures such as nature ship, tenancy, or land tenure reserves are not The WCD Cross-Check Survey reveals a papers. One-fifth of those physical- enumerated and similar trend towards under-estimation ly displaced by the Kao Laem dam considered for insofar as 35% more people were resettled in Thailand were from the Karen resettlement. than initially planned. This figure must ethnic group. Because they lacked represent a lower bound on the error in the legal residence permits, they were Cross-Check dams, given the poor reliability considered ineligible for resettlement.14 of the estimates. Data provided by NGOs as part of the external review of the Cross- Often, people physically displaced by canals, Check Survey confirm this view, as many of powerhouses, and associated compensation the actual figures for the physically dis- measures such as nature reserves are not placed were disputed. enumerated and considered for resettlement. Examples of this type emerge from all parts Affected groups that are not of the world, including Sulawesi, Indonesia; counted or compensated the Mahaweli Development Programme, Sri Lanka; and the Sardar Sarovar project, Surveys of the categories of people to be India.15 Further, compensation is often not affected by dams have generally been paid to those affected by such additional inadequate. The scope of definition of the components of a project.16 affected has been limited, and the totality of affected groups has not always been deter- While not all large dams have involved mined. The principal categories excluded physical displacement it would be much from assessments include the landless, rarer to find a river whose natural function downstream communities and indigenous is not used or appreciated by people in some people. The WCD Case Studies show that fashion. And in many cases in densely communities situated downstream from the populated tropics large dams will lead to dam, those without land or legal title, both physical and livelihood displacement. indigenous people and those affected by For example, the Urrá 1 dam on the Upper project infrastructure (and not just the Sinú River in Colombia displaced 12 000 reservoir) were not considered as affected people but also affected severely more than people at the time of design. 60 000 fishermen in the lower Sinú, where the fish population diminished drastically as Among those assessed, compensation has a result of the dam.17 usually gone only to those in possession of legal titles, leaving out a large number of Physically displaced people – often the poorest – who depend on populations enumerated but common resources such as forests and grazing grounds for subsistence. In the WCD not resettled Case Studies on Grand Coulee, Tarbela, Among physically displaced people officially Aslantas and Tucurui, only those affected recognised as ‘project affected,’ not all are people with legal title were compensated for given assistance to resettle in new locations.

The Report of the World Commission on Dams 105 Chapter 4

In India, those actually resettled range from But none of the owners had been paid com- less than 10% of the physically displaced in pensation. The Colville agent reported that the case of the Bargi dam to around 90% for the Indians were growing resentful, they the Dhom dam.18 The Yacyreta project in needed money to build new homes and Argentina and Paraguay is a classic illustra- improve their remaining property and they tion of delayed and incomplete resettle- knew that white owners across the river had ment. It took the project developers 20 years already been paid.20 The Colville and to resettle just over 30% of the displaced Spokane tribes only received cash compensa- people, leaving the remainder to be resettled tion for reservation land in 1941. Two tribal in the less than two years before the reservoir towns, Keller and Inchelium, were rebuilt and would be filled. If the experience of other still exist today, but other smaller settlements projects involving large displace- were lost with the inundation.21 ment in a region is taken into Little or no meaningful consideration, a large proportion participation of affected Experience of affected people of the Yacyreta displaced are people in the planning with resettlement, mitigation, unlikely to be resettled.19 and implementation of and compensation dam projects – including The WCD Case Study on Little or no meaningful participation of resettlement and Tarbela reports that of the affected people in the planning and imple- rehabilitation – has taken 96 000 physically displaced mentation of dam projects – including place. people enumerated for the resettlement and rehabilitation – has taken Tarbela dam in Pakistan, two- place. Involuntary, traumatic and delayed thirds qualified for replacement agricultural relocation, as well as the denial of develop- land in Punjab and Sindh provinces. Of ment opportunities for years and often these, some 2 000 families or approximately decades, has characterised the resettlement 20 000 people did not receive land when the process.22 For millions of people on all amount of land provided by Sindh fell short continents, displacement has essentially of that promised. In the case of Aslantas, occurred through official coercion.23 The only 75 of an estimated 1 000 displaced starkest example from the WCD Case families asked for resettlement, with the Studies comes from one of the earlier dam remainder choosing cash compensation. Of projects, Kariba, where the resistance of the these, 49 were considered eligible for Tonga people ended with the fatal shooting resettlement and subsequently received new of eight people.24 The displacement of housing. In the case of Tucurui, of the people at the Sri Sailam project in India in indigenous groups physically displaced only 1981 was also achieved through force.25 the Parakaná people were resettled; the other Eviction of people at the Chixoy dam site in indigenous group that lost land to the dam was Guatemala led to the killing of about 376 not considered for resettlement benefits. Maya Achi people from the submergence area.26 In implementing the Miguel Aleman In the Grand Coulee project, the Colville dam in Mexico, employees from the Papal- and Spokane reservation lands, in addition oapan River Commission set fire to homes to three towns, were inundated. By the of 21 000 Mazatec Indians who were refus- summer of 1940, the rising water covered ing to move.27 In other cases – such as the the first tracts of land and the government submergence of 162 villages when the Bargi was clearing allotments and burning houses. dam in India was filled without warning –

106 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

authorities have resorted to eviction displaced persons themselves. They have through the filling of reservoirs prior to the often been forced to resettle in resource- departure of the displaced.28 depleted and environmentally degraded areas around the reservoir. Such lands Cash compensation is a principal vehicle for rapidly lost their capacity to support the delivering resettlement benefits, but it has resettled population. Among the earliest often been delayed and, even when paid on instances is the Liu-Yan-Ba project on the time, has usually failed to replace lost Yellow River in China, which displaced livelihoods. Compensation is understood to 40 000 people from fertile valleys and refer to specific measures intended to make relocated them to windswept uplands. good the losses suffered by people affected Erosion and loss of fertility ultimately led to by the dam. It usually takes the form of a the abandonment of painstakingly reclaimed one-off payment either in cash or kind for farmland, and the drastic reduction of land, housing and other assets.29 The WCD farmland led to extreme poverty.33 Similar Case Studies show that downstream com- experiences have been recorded from Hoa munities affected by loss of floodplain Binh in Vietnam, Sirindhorn in Thailand, vegetation and fisheries in Tucurui dam in Batang Ai in Sarawak Malaysia, and other Brazil and Tarbela dam in Pakistan were not rice-growing East Asian countries with large compensated. The Gavaio da Montanha rural populations.34 indigenous people, whose lands were affected by the transmission lines in the Tucurui The loss of cultivable land and inability to project, were initially not considered gain good-quality replacement land has eligible for compensation but were later significantly affected indigenous peoples and given cash compensation. In the cases of peasant farmers. Examples are the Chinan- Aslantas (Turkey), Tarbela (Pakistan) and tec and Mazatec Indians displaced by the Kiambere (Kenya) dams, affected people did Miguel Aleman and Cerro de Orro dams in not receive adequate compensation to buy Mexico; the Kuna and Embera people in alternate land.30 Panama; the Parakana, Asurini, and Gavio da Montanha people in Brazil; and the Further, there have been many cases illus- Tonga in Zambia and Zimbabwe.35 trating inadequate compensation, unsuitable mitigation, and lack of recourse, including The replacement of agricultural land, basic the Sri Sailam project in India and the Kao services and infrastructure at resettlement Laem in Thailand.31 Delays in compensa- sites has often failed to materialise, was tion provisions, titles to landholdings and inadequate, or was delayed for many years. houses, and provision of basic services have Absence of livelihood opportunities forces occurred. Cases illustrating inordinate affected people to abandon resettlement delays – from 5 to 15 years – include the sites and migrate. Examples include Tarbela, Aswan High dam in Egypt, the Nangbeto in where allotted agricultural land was of poor Togo, the Akosombo in Ghana, the Itá in quality and basic services such as electricity, Brazil, and the Bhumibol in Thailand.32 health facilities and schools were not provided. Electricity was only provided after Resettlement sites are often selected with- 25 years. Similar experiences are recorded out reference to the availability of liveli- from resettlement sites at Tucurui, Sirind- hood opportunities or the preferences of horn dam in Thailand, and Akosombo in

The Report of the World Commission on Dams 107 Chapter 4

Ghana.36 In northeast the case of India, 75% of the people dis- Thailand, 15 000 farming placed by dams have not been rehabilitated families were left without and are impoverished.41 A monitoring study lands as a result of failed in 1993 found that 72% of the 32 000 resettlement schemes displaced people from the Kedung Ombo between 1960 and 1970.37 dam in Indonesia were worse off after Government reports in resettlement.42 Conditions among the 800 China characterised reser- ethnic minority Nya Heun families dis- voir resettlement problems placed by the recently constructed Houay as ‘seven difficulties’ (qui Ho dam in Laos are reported to be appalling, nan) and ‘four inadequacies’ (si cha). The with people suffering from severe lack of seven difficulties include shortages of food, shortage of arable land and insufficient electricity, drinking water, schools, food, clean water.43 medical services and means of communication and transportation. The four inadequacies Often the sheer scale and level of displace- refer to the insufficient amount and poor ment makes adequate rehabilitation and quality of irrigation, housing, flood control livelihood restoration difficult. An inverse and reservoir maintenance facilities.38 relationship between the scale of displacement and the possibility of properly carrying Resettlement programmes have predomi- out resettlement is evident. For example, nantly focused on the process of physical the inundation zone of the Danjiangkou relocation rather than on the economic and project in the Hubei province of China, social development of the displaced and implemented in 1958, covered four rural other negatively affected people.39 Lack of counties and 345 villages. Although the accountability on the part of the state for Chinese Government tried hard in the promised entitlements has led to poor (and 1980s and 1990s to improve the living incomplete) implementation of resettlement standards of the physically displaced in the measures. Finally, long delays in the onset of Danjiangkou area, many unsolved problems resettlement programmes are common and persist. In 1996, an estimated 35 000 of lead to great uncertainty and those resettled around the city of Shiyan psychological and social anxiety had incomes below the official poverty An inverse relationship for those awaiting resettlement. line.44 And in India, the sheer extent of between the scale of These and other problems have displacement is making resettlement a displacement and the severely eroded the effectiveness daunting task for the Sardar Sarovar project possibility of properly of resettlement and rehabilita- (see Box 4.3). Since the start of the resettle- carrying out tion programmes in creating ment process in 1984, less than 20% of the resettlement is evident. development opportunities for recognised displaced people have been the resettled and have height- resettled.45 ened the risk of impoverishment for those being resettled. Elements for positive mitigation, development and That the livelihoods of those resettled have resettlement outcomes not been restored therefore comes as little surprise. At least 46% of the 10 million Impoverishment of affected people is Chinese resettled as a consequence of increasingly seen as unacceptable but it is reservoirs are still in ‘extreme poverty’.40 In also unnecessary since there are a wide

108 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

range of opportunities available for making measures were taken to maintain communi- not only resettlers, but all affected people ty cohesion and identity.47 Social science project beneficiaries. This is in the interests input had a strong influence in determining of all stakeholders since, as beneficiaries, the outcome there. Baseline demographic affected people add to the stream of project and socio-cultural studies played an impor- benefits, while reducing costs. The problem tant role in informing planners about of making them beneficiaries lies not with distinct social and cultural features of people affected people, who time and again have living in the proposed impact areas. shown the capacity to respond to opportunities that are available, but with the inade- In cases where compensation packages were quate laws, policies, plans, financing capaci- negotiated with project affected people and ty and political will of governments and other stakeholders, the process has resulted project authorities. in fewer instances of injustice and better outcomes for the resettle- For resettlement to lead to the development ment process. Even where every- For resettlement to lead of those resettled, the process has to address one may not see negotiated to the development of the complexities of resettlement itself and to compensation as the most appro- those resettled, the effectively engage the full range of political priate or effective option, affected process has to address and institutional actors. A positive outcome people tend to feel more satisfied the complexities of requires several enabling conditions such as for having engaged in the negoti- resettlement itself and to low level of displacement, resettlement as ation process, as attested by the effectively engage the development policy with supporting legisla- Zimapan resettlement program in full range of political and 48 institutional actors. tion, a combination of land and non-land Mexico. In case of the Mubuku based sustainable livelihood provisions, III hydropower project in Uganda, strong community participation and ac- public consultation meetings with countability and commitment from govern- Ugandan local council system and commu- ment and project developers. nity leaders were held for identification and valuation of land.49 This minimised dis- Providing a legal framework that governs placement by enabling adjustments to the the process of displacement is important for routing of canals. protecting rights of affected people. For instance, China’s Reservoir Resettlement The plan developed by the Chinese govern- Act specifies the rights of affected people ment for people affected by the Xiaolangdi and defines the obligations of the State and dam provides an example of an integrated the procedures for settling conflicts and the strategy that combines land and non-land redress of complaints.46 Recent changes in based activities to ensure livelihoods.50 Chinese policy serve as an interesting model Resettlement plans focus on building the for other countries. Minimising displace- skills of the adversely affected through ment is another enabling condition to substantial investments in imparting new, effectively address resettlement needs. relevant skills that are in demand in the regional and local economy, enhancement In some cases, project proponents have of existing skills, and special measures to made an effort to resettle people as commu- facilitate capacity building amongst women. nities in order to minimise socio-cultural This approach calls for resettlement plans to disruptions. In the Kainji project, Nigeria develop linkages between negatively affect-

The Report of the World Commission on Dams 109 Chapter 4

ed people and other sectors of the economy, Past and current experiences of affected underlining the importance of incorporating people and the rapidly changing context the overall development of the affected reinforces the argument that displacement economy into the resettlement programme. needs to be located in the broader perspective of the tensions between the local versus An inclusive process involving all groups – the national and international interests. Just including host communities – enables as displacement is not an inevitable conse- initiatives to promote resettlement as quence of infrastructure development, development to be managed jointly by the resettlement need not necessarily result in people and project and government institu- impoverishment. Empowering people, tions as a long-term process that can con- particularly the economically and socially tribute to the stream of project benefits. In marginalised, by respecting their rights and the case of Itá dam in Brazil, a ensuring that resettlement with develop- sustained struggle by the local ment becomes a process governed by negoti- Empowering people, community for proper resettle- ated agreements is critical to positive particularly the ment resulted in joint negotia- resettlement and rehabilitation. economically and socially tion for benefit-sharing, resettle- marginalised, by ment as a community, and Indigenous Peoples respecting their rights consultative implementation of and ensuring that the programme.51 The accord Large dams have had serious impacts on the resettlement with between affected people and the lives, livelihoods, cultures and spiritual development becomes a utility Electrosul resulted in a existence of indigenous and tribal peoples. process governed by community managed resettle- Due to neglect and lack of capacity to secure negotiated agreements ment programme. justice because of structural inequities, is critical to positive cultural dissonance, discrimination and resettlement and Evolving policy frameworks in economic and political marginalisation, rehabilitation. countries such as Ghana and indigenous and tribal peoples have suffered China reflect two of the more disproportionately from the negative im- positive attempts to learn from pacts of large dams, while often being past resettlement experience. In the Ghana- excluded from sharing in the benefits.54 In ian case, with the benefit of the administra- the Philippines, almost all the larger dam tive continuity of the Volta Resettlement schemes that have been built or proposed Authority, planners at Kpong dam were able were on the land of the country’s 6–7 to avoid some of the mistakes made earlier million indigenous people.55 Similarly in at Akosombo.52 While the legal framework India, 40–50% of those displaced by devel- related to land and resettlement is compre- opment projects were tribal people, who hensive and improvement was noted at account for just 8% of the nation’s 1 billion Kpong, not all good intentions were success- people.56 These costs are not balanced by fully pursued.53 China’s resettlement experi- any receipt of services from dams or by ence before 1980 was in many ways inade- access to the benefits of ancillary services or quate, and the new policy sought to improve indirect economic multipliers in the formal matters. How effectively such policy im- economy.57 provements are translated into successful resettlement and development outcomes In general, development planning and remains to be seen. implementation have inadequately ad-

110 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

dressed the special needs and vulnerabilities The rights of indigenous peoples of indigenous and tribal peoples. In many and ethnic minorities are often cases, large dams have only perpetuated this poorly defined or enshrined in the disregard and exacerbated the problem – national legal frameworks, and even causing multiple displacements of consequently their entitlements these peoples. The Waimiri-Atroari of have lacked effective protection. northern Brazil numbered 6 000 in 1905. The Bayano dam in Panama that Eighty years later, massacres and disease left forced the indigenous Kuna and only 374 Waimiri-Atroari alive. In 1987, Embera peoples from their tradi- the Balbina dam flooded two of their tional territories resettled them villages, displacing 107 people.58 Similarly, on land that was less fertile and subject to in the unique Biobio region in Chile, the encroachment by loggers. The Panamanian Pehuenches were pushed higher and higher government systematically failed to fulfil up in the valley throughout the last century. agreements made with the affected indige- The Pangue and Ralco projects would nous people at the time of construction, as inundate much of the remaining ancestral well as commitments negotiated later. land of the Pehuenches.59 For the Ibaloy Among the violations was the government’s indigenous people currently living in the failure to compensate adequately for the loss fertile Agno river basin in the Philippines, of traditional territories and provide legal 63 the San Roque dams would be the third to titles to the new lands. What happened in impact their lands.60 Similar experiences are Panama in the 1970s is similar to what has 64 recorded in Indonesia, Malaysia, Thailand, happened in Malaysia in the 1990s. In the Brazil, Argentina, Mexico, Panama, Colom- case of the Bakun project, rights to indige- bia, Guatemala, United States, Canada, and nous common land in the Ulu Belaga site 65 Siberia.61 were not recognised or properly assessed.

Industrial countries’ experience with indige- For indigenous peoples and ethnic minori- nous peoples in the era of building large ties, dam-induced displacement can trigger a dams was not very different from that of spiral of events that spreads beyond the developing countries. Dams built during the submergence area. A case in point is the 1950s and 1960s cost the indigenous nations situation of the 100 000 Chakma people of the Missouri River basin in the United displaced by the Kaptai hydropower dam in States an estimated 142 000 hectares of their the Chittagong Hill Tracts, Bangladesh. The best land, including a number of burial and project submerged two-fifths of their culti- other sacred sites, leading to further impover- vable land; as a consequence, 40 000 Chak- ishment and severe cultural and emotional ma left for India and another 20 000 were trauma. A guarantee used to rationalise the supposed to have moved into Arakan in plan – that some 87 000 hectares of Indian Burma. The Chakma have never gained land would be irrigated – was scrapped as the citizenship for themselves or their children project neared completion.66 in India. The conflict triggered by land shortage between the Buddhist Chakma Despite changes over the years, new projects people and Muslim Bengali settlers has cost in industrial countries raise similar issues. A 10 000 lives since the project was completed case in point is the second stage of the in 1962.62 Churchill Rivers project in Labrador,

The Report of the World Commission on Dams 111 Chapter 4

Canada, consisting of two dams seek mitigation, let alone development and two river diversions that will benefits. flood a large area of hunting territory of the Innu people who Downstream communities throughout the live on both sides of the provin- tropics and subtropics face some of the most cial boundary. The Innu have drastic impacts of large dams, particularly yet to be clearly recognised as where the changed hydrological regime of the owners of their lands, and rivers has adversely affected floodplains that the whole area is the subject of supported local livelihoods through flood- an unresolved Innu land claim recession agriculture, fishing, herding and currently being negotiated with the Canadi- gathering floodplain forest products. The an government.67 disruption of downstream economies that results from the insertion of a dam and the In the last two decades, international and subsequent reduction in natural floods can national laws progressively aimed at empow- create uncertainty in livelihoods and render existing skills unproductive – leading to ering indigenous peoples to play a decisive migration, dependence on informal wage role in development planning and imple- labour in urban areas and impoverishment. mentation. Constitutions in some countries recognise the vulnerability of indigenous In northwestern Nigeria, the Bakolori dam people in mainstream development process- on the Sokoto River reduced average flood es arising from their distinct culture and levels by 50%, leading to a fall in cropped history and have designed safeguards to area of 53% and to a quarter of the house- 68 protect their rights. The scope of interna- holds dropping dry-season cultivation as a tional law has widened and currently includes component of their livelihood strategy.69 a body of conventional and customary norms Similarly, significant impacts on floodplain concerning indigenous peoples, grounded on agriculture are seen in Niger, Chad, Nigeria, self-determination. In a context of increasing Sudan, Senegal and Mali.70 In the case of recognition of the self-determination of Manantali dam on the Senegal River, indigenous peoples, the principle of free, between 500 000 and 800 000 people prior, and informed consent to development suffered from loss of access to productive projects and plans affecting these groups has floodplains that provided most or part of emerged as the standard to be applied in their means of survival.71 The creation of protecting and promoting their rights in the the Sobradinho reservoir in Brazil affected development process. the livelihood of 11 000 farm families downstream from the reservoir who depend- Downstream Livelihoods ed on traditional floodplain agriculture.72 The WCD Case Studies reveal that the Downstream impacts can extend for many cumulative impact of Tarbela dam and Kotri hundreds of kilometres and well beyond the barrage has affected the grazing activities of confines of the river channel. The serious pastoral communities in Pakistan. implications come to the fore only after completion of the dam and a number of the Substantial losses to downstream fishery impacts only develop over time. In general, production as a result of dam construction the downstream riverine communities have are reported from around the world. Along lacked social, economic and political power to with subsistence agriculture, fisheries

112 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

constitute an important livelihood activity Downstream impacts are not In the case of Manantali among large rural populations as well as an only among the most signifi- dam on the Senegal River, important low cost source of protein. Many cant unassessed and unad- between 500 000 and of these households depend on fisheries dressed aspects of large dams, 800 000 people suffered either as their primary or a supplementary they are also indicative of the from loss of access to source of livelihood. The impact of dams on magnitude and spread of productive floodplains that fisheries is no less in temperate regions, impacts associated with an provided most or part of where salmon runs over large stretches of altered river regime. The their means of survival. river in North America and Canada have extent to which mitigation been destroyed, affecting the livelihood, and development can be food security and cultural organisation of designed and implemented to Native American communities. The Grand address these complex and diverse concerns Coulee Case Study reports that the loss of effectively is open to question. As demon- salmon also had severe cultural and spiritual strated in a case from northern Nigeria, an consequences integral to the First Nations’ examination of the economic value of way of life. One problem faced by indige- downstream uses of water may also provide a nous peoples who live downstream from convincing argument for setting aside dam dams in boreal zones is the discharge of projects (Box 4.4). power generating water on top of frozen rivers during winter, as in case of the Koly- Box 4.4 Economic value of downstream floodplains, Hadejia-Nguru, ma River in northern Yakutia.73 Nigeria In northern Nigeria, extensive floodplains exists where the Hadejia and Jama’are Rivers converge. The floodplains provide essential income and nutrition benefits Adverse impacts on downstream fisheries in the form of agriculture, grazing resources, non-timber forest products, have been severe, even in projects imple- fuelwood and fishing for local populations, and help to recharge the regional aquifer which is an essential groundwater source. However, in recent decades mented in the 1990s. These impacts were the floodplains have come under increasing pressure from the construction of not adequately assessed in, among others, the Tiga and Challawa Gorge dams upstream. The maximum extent of flooding has declined from 300 000 hectares in the 1960s to around 70 000–100 000 the Urrá I project in Colombia; the Singkar- hectares more recently with plans for a new dam at Kafin Zaki. Economic analysis ak project in West Sumatra, Indonesia; the of the Kano River Project, a major irrigation scheme benefiting from the upstream dams, shows returns to water of $1.73 per 1 000 m3 and when the Lingjintan project in China; Theun Hin- operational costs are included, the net benefits of the project are reduced to boun in Laos; and Pak Mun in Thailand.74 $0.04 per 1 000 m3. Downstream communities and livelihoods A combined economic and hydrological analysis was conducted to simulate the impacts of these upstream projects on the flood extent that determines the are typically left out of any explicit reckon- downstream floodplain area. The economic gains of the upstream water projects ing of project impacts and subsequent efforts were then compared to the resulting economic losses to downstream agricultural, fuelwood and fishing benefits (valued at $32 per 1 000 m3 of water in 1989 to manage these impacts – apart, perhaps, prices). Given the high productivity of the floodplains, the losses in economic from installation of fish passages. One benefits due to changes in flood extent for all scenarios are large, ranging from $3 million to $24 million. As expected, there is a direct trade-off between explanation for this is that the downstream increasing irrigation upstream and impacts on the floodplains downstream. Full communities are not only dispersed but also implementation of all the upstream dams and large-scale irrigation schemes would produce the greatest overall net losses, around $20 million. have typically lacked social, economic, and These results suggest that the expansion of the existing irrigation schemes political power to press their case for mitiga- within the river basin is effectively ‘uneconomic’. The introduction of a regulated tion and development. While the people flooding regime would reduce the scale of this negative balance substantially, to around $16 million. The overall combined value of production from irrigation affected by the flooding of the reservoir and the floodplains would however still fall well below the levels experienced if could assert their right to mitigation by the additional upstream schemes were not constructed. refusing to move, those affected downstream Source: Acreman et al, 2000, Contributing Paper for WCD Thematic Review II.1 Ecosystems have no such leverage.

The Report of the World Commission on Dams 113 Chapter 4

Gender oblivious of the gender aspect of resettlement.’76 After the Asian Development Bank Gender relationships and power structures approved a gender policy in 1998, a review are all too often detrimental to women. of its dam projects observed that the impacts Extensive research has documented gender on gender at the project preparation and inequalities in access to, and control of, implementation stages were often not economic and natural resources. In Asia and considered.77 Where planning is insensitive Africa for example, women may have use to gender, project impacts can at best be rights over land and forests, but are rarely neutral, and at worst aggravate existing allowed to own and/or inherit the land they gender disparities to the extent of radically use.75 Communities near the Tarbela dam in affecting the pre-project gender balance. Pakistan practised the purdah system (seclusion of women) for centuries. This Dam projects often impose the gender bias section employs the WCD Knowledge Base of the developer (typically the state), with to examine whether dams aggravate, or negative effects for local arrangements that ameliorate gender disparities. provide livelihood opportunities for women. When the Mahaweli dam in Sri Lanka was Given the gender-blindness of the planning built the prevalent inheritance rule, which process large dam projects typically build on allowed women the independent right to co- the imbalance in existing gender relations. own and control land, was undermined by a For affected communities dams have wid- new arrangement that allowed the house- ened gender disparities either by imposing a hold to nominate one heir, usually a son.78 disproportionate share of social costs on In tribal communities in India women do women or through an inequitable allocation not have land rights and therefore they have of the benefits generated. However, the WCD not been compensated for the land they Knowledge Base also provides evidence of have lost as users. 79 Instead women’s cases where dams have served as opportunities interests are seen as linked to the household for reducing gender disparities, primarily among women in and only men and major sons are given land households or communities that according to the local government’s resettle- receive access to project services. ment policy for the Sardar Sarovar project. Women traditionally held land rights Widened gender amongst the egalitarian Gwembe Tonga disparities community in Northern Rhodesia (now Zambia). However, the British colonial In spite of the fact that many authorities that built the Kariba dam only countries and funding agencies recognised men as land owners and women have adopted specific gender lost their land without compensation during policies in recent years aimed at displacement and resettlement.80 mainstreaming gender issues in their development interventions, actual Forests, fisheries and other common proper- project planning and implementation ty resources, which support subsistence continue to overlook gender aspects. An livelihoods, are often not replaced during assessment by the World Bank’s OED of a resettlement with women often bearing a number of World Bank projects noted that disproportionate share of the resulting costs. the experiences studied ‘…were largely The Tarbela Case Study notes that women

114 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

have suffered more than men have from the increased domestic violence.82 As men face disruption of their social life resulting from powerlessness, women (and children) involuntary displacement from their ances- become scapegoats.83 Reduced fishing tral land, which severed their relationship opportunities in coastal and mangrove areas with water, forests and other natural re- downstream of Tarbela dam destroyed the sources. At the Pak Mun dam the loss of structures of families traditionally organised local edible plants due to submergence around this activity and accelerated male resulted in loss of income and sources of out-migration. Women faced increased subsistence. Again this affected women responsibilities as de facto heads of house- disproportionately, as they are responsible holds, while household income was severely for collecting and processing these plants. affected.

The general impoverishment of communi- While women in affected communities bear ties and the social disruption, trauma and a disproportionate share of the costs, they health impacts resulting from displacement have often had less access to the benefits have typically had more severe impacts on generated by dams. The employment women. In Ghana, the general impoverish- created during the construc- ment resulting from involuntary displace- tion of large dams generally ment associated with the Akosombo dam benefits men, as illustrated in While women in affected led to increased male migration to urban the Grand Coulee Case Study communities bear a areas and an increase in households headed where it was only at the later disproportionate share of by women.81 In South Africa, farm-workers’ stages of the construction the costs, they have households headed by women suffered phase that the government often had less access to unduly during the displacement process agreed to hire women, and the benefits generated caused by the Gariep and Vanderkloof dams: only for the administrative by dams. 30% of men-headed households followed clerical work. The allocation farmers to new farms compared to 15% of of the irrigated land made available by dams women-headed households. As a result, 75% is also often done in a manner that exacer- of women-headed households ended up bates gender inequalities. In the Mahaweli living for more than one year in the no- irrigation scheme in Sri Lanka, fully 86% of man’s land along the highways called the the land allocations were made to men, and corridor, where some even gave birth. As only two local women-headed households the Kariba case illustrates, the influx of were granted land. In addition, the preva- immigrants during construction and the lent inheritance rule, which allowed women resulting urbanisation can increase levels of the independent right to co-own and sexually transmitted diseases and, more control land, was undermined by a new recently, HIV-AIDS prevalence rates which arrangement that allowed the household to have negatively affected local women. nominate one heir, usually a son.84

Displacement can make women’s position Dams as opportunities for inside and outside the family more precari- addressing existing inequalities ous. For example, at resettlement sites for the Sardar Sarovar dam in India, Kariba in There are also examples in the Knowledge Zambia-Zimbabwe and Nangbeto dam in Base where dams have provided benefits to Togo-Benin, increased alcoholism markedly women. As gender is a relational concept,

The Report of the World Commission on Dams 115 Chapter 4

access by women to the benefits generated in Egypt, Tunisia, and Sri Lanka demon- by a dam is a necessary but not sufficient strate how land tenure reforms have benefit- condition for positive gender impacts.85 As ed poor people (including women) recruited these impacts are seldom as new landholders in irrigation schemes.87 documented, little is known A survey of 32 villages conducted in 1991- Large infrastructure regarding the ways in which 92 showed that women owned 6% percent projects like dams or benefits generated by dams of the almost 2 500 irrigated plots in the their associated irrigation affected existing gender Middle Valley of the Senegal River. While schemes can present disparities and relationships. this reflects a continued gender imbalance, unique opportunities for Where dams have improved in the context of the Senegal valley it reforms. the general supply of services, represents a positive gender impact. In (see Table 4.1) the increased traditional systems of rainfed and recession availability of water for agriculture women and marginal communi- household uses, electricity and food ties had only use and not property rights. In (through irrigation) are likely to have this case government agencies used the benefited women by reducing time spent on opportunity of centralised control over the chores and improving nutrition. Where allocation of irrigated land to provide social services are provided as part of reset- women-headed households with ownership tlement programmes, these may represent an rights.88 improvement compared to the pre-displacement situation. For example, 80 000 people Cultural Heritage resettled from the Akosombo dam benefited from services including the following: 82 Although improvements have been noted in school blocks, 46 markets, 146 public recent years, potential cultural heritage latrines, 52 boreholes, 6 wells and 162 water impacts are still largely ignored in the stand pipes.86 planning process, especially in industrialised countries.89 Large dams have had significant Where dams achieve an improvement in adverse effects on this heritage through the living standards in impact areas, this can loss of local cultural resources (temples, have a positive spill over effect on gender shrines, and sacred elements of the land- equity. For example, improved family scape, artefacts and buildings) and the income resulting from irrigated agriculture submergence and degradation of archaeolog- at Aslantas dam enabled farmers to give ical resources (plant and animal remains, both boys and girls higher education. This, burial sites and architectural elements). The along with the eradication of illiteracy, has latter may be part of the cultural life of local contributed to ending polygamy in the communities, or they may predate the basin. arrival of people currently inhabiting the dam site. Shoreline erosion processes can Because gender disparities and even severe expose subsurface archaeological remains, marginalisation of women exist in many encouraging looting and illicit digging for countries, large infrastructure projects like artefacts and valuable remains. Dams can dams or their associated irrigation schemes also cause loss or damage of cultural heritage can present unique opportunities for re- through land reclamation and irrigation forms, in areas like land tenure, that can projects and the construction of power lines, contribute to reversing the situation. Cases roads, railways and workers towns.

116 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

In most cases no measures have been taken In the WCD Case Studies, two to minimise or mitigate the loss of cultural dams – Pak Mun and Aslantas and archaeological resources. Affected – were redesigned to avoid communities repeatedly raised the treatment impacts on cultural and of burial sites at the WCD Regional Consul- archaeological resources. The tations and other public hearings. During Aswan High dam, (see Box the construction of the Inanda dam in 4.5) admittedly an exceptional South Africa, remains of human bodies case, illustrates not only how buried under the reservoir site were ex- important potential losses of humed and all buried in one hole, profound- cultural heritage can be, but ly disturbing local communities.90 The also how efforts to conserve cultural resourc- Grand Coulee Case Study records the es can improve understanding of cultural submergence of Native American burial heritage. A study in the United States sites by the dam waters. The tribes used demonstrated that although submergence funds provided by the authorities, and their may be a way of preserving archaeological own means, to relocate burial sites exposed resources, it is more cost-effective to exca- by receding reservoir waters. The risk of vate and manage these resources prior to submerging ancestral graves is one of the reservoir inundation than to leave them for main reasons the Himba people in Namibia possible future underwater archaeological oppose the planned Epupa dam.91 expeditions.94

The assessment of lost or buried cultural Despite the established potential for signifi- heritage resources not directly linked to cant and often irreversible losses of cultural local people has been at least equally resources due to dam construction, cultural significant, but often more difficult to heritage management is still not adequately estimate. The difficulty lies in the fact that considered in the planning process. In no investigation of cultural and archaeologi- Turkey, for example, only 25 of 298 existing cal resources has taken place as part of the dam projects have been surveyed for cultural planning process of most dams. Given that heritage, and of these only five have had river valleys often hosted the most ancient systematic rescue work conducted.95 In civilisations, the importance of losses from Argentina, despite the fact that many existing dams can be assessed by default, on the basis of the quality and quantity of finds in areas affected by dams where some Box 4.5 The Aswan High dam: a milestone in the history of archaeology cultural heritage assessment did take place. When the Madden dam in Panama dropped The potential adverse effects of the Aswan High dam on the monuments of ancient Nubia were recognised in 1954, one year after selection of the dam site. to its lowest historical limit in 1998, it Thanks to an international effort driven by UNESCO, the ancient monuments of exposed thousands of artefacts, cultural Egypt and Sudan were saved from inundation. Equally important, the international rescue operation led to decades of intensified archaeological research in 92 features and human burial sites. In 1988 in the vicinity of the dam, greatly enhancing understanding of the civilisation of India, reconnaissance surveys in 93 of the Nubia. This radically altered the knowledge of Egyptian archaeology, resulting in the rewriting of the prehistory of the Nile Valley. What the Director General of 254 villages to be submerged in the Narma- UNESCO called ‘a task without parallel in history’ subsequently led to the da Sagar dam impoundment area yielded launch of numerous other operations supported by UNESCO to save world cultural heritage. hundreds of archaeological sites ranging Source: Hassan, 2000, in Brandt and Hassan, 2000, from Lower Palaeolithic to historic temples WCD Working Paper on Cultural Heritage Management and iron smelting sites.93

The Report of the World Commission on Dams 117 Chapter 4

provinces have updated their legislation on Akosombo.99 Rift Valley Fever has also cultural heritage, cultural resources manage- spread due to the Aswan and Kariba dams ment activities related to large dams were and irrigation systems along the Blue Nile in either poorly done or not at all.96 Sudan.100 Most reservoir and irrigation projects undertaken in malaria-endemic The India Case Study reveals that although areas increase malaria transmission and projects like Narmada Sagar, Tungabhadra, disease.101 The increase was more pro- Bhadra, and Nagarjunsagar have paid some nounced for dams below 1 900 meters of attention to major temples and places of altitude and less pronounced above that worship, almost all the dams built so far altitude.102 Similarly in India, the Sardar suffer from lack of cultural heritage studies Sarovar and Upper Krishna projects demon- (let alone mitigation measures). In China, strated a high potential for malaria transmis- the Three Gorges Project illustrates the sion in the short term and thereafter leading potential to do damage through neglect of to transmission of Japanese encephalitis.103 cultural heritage. The combined problems of time constraints, under-budgeting, and a In new dams in tropical, sub-tropical, and shortage of qualified personnel are seriously arid regions there is rapid eutrophication hampering the salvation and preservation of resulting in problems of excessive aquatic the impressive archaeological and cultural weed growth or ‘blooms’ of toxic cyanobac- sites in the areas to be affected.97 teria. This is reinforced by enhanced nutrient pollution through growth of towns, Human Health agriculture and mining operations in the catchment. In China, a high incidence of Environmental change and social disruption primary liver cancer has been linked to the resulting from large dams and associated presence of cyanobacterial toxins in drink- infrastructure developments such as irriga- ing water.104 tion schemes can have significant adverse health outcomes for local populations and Another problem is the accumulation of downstream communities. The issue of high levels of mercury in reservoir fish. equity – in terms of pre-existing nutritional Mercury, naturally present in a harmless and health conditions of the population and form in many soils, is transformed by bacte- the capacity to resist new health problems – ria feeding on the rotting biomass in reser- is at the root of the adverse health impacts voirs into methylmercury, a central nervous of dams.98 Among the resettled, access to system toxin. Alternatively, effluent from drinking water, health services and ability to human activities such as mining may lead to cope with new social and physical environ- the accumulation of mercury in reservoirs. ment determines health conditions. As methylmercury passes up the food chain it becomes increasingly concentrated in the Numerous vector-borne diseases are associ- tissue of the animals eating contaminated ated with reservoir development in tropical prey, potentially threatening human health areas. Schistosomiasis (or Bilharzia) spread (see Box 4.6). through snails breeding in still or slow- moving waters was a significant public Socio-cultural disruptions can be traumatic health problem that emerged from many for communities. The Kariba Case Study early projects, such as Kariba, Aswan and reports on the strong emotional response of

118 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

the Gwembe Valley Tonga to their involun- National Research Institute for Amazonia tary resettlement. When 50 people died a study for Eletronorte underlining the mysterious and sudden death in 1959 in the association between macrophytes and the Lusitu area, those relocated attributed these proliferation of insect vectors. After Tucurui deaths, together with deaths from dysentery was filled in 1984, an unusual proliferation and measles that occurred earlier in the year, of Mansonia mosquitoes in rural areas close to ‘bad spirits’. The ‘good spirits’ of the to the reservoir forced farm families to leave Tonga had been drowned by the lake and their homes. Test subjects received over could no longer protect them from such 500 bites per hour at the height of the ailments. infestation.

Destruction of community productive bases Despite several decades of precedence, in agriculture and fisheries can give rise to documented experiences from different food shortages, leading to hunger and regions, and the availability of sophisticated malnutrition. The Kariba Case Study recalls assessment techniques and instruments such that the serious food shortages of 1958-60 as Health Impact Assessment, health can be largely traced to resettlement- concerns were not integrated in the design associated factors. Food shortages due to of the dams and infrastructure to the extent resettlement are also reported in Vietnam, possible. Mitigation suffered from lack of China, Malaysia, Thailand and India.106 preparedness and commitment and health concerns were not addressed effectively. For In recent years, the high incidence of HIV/ AIDS in construction and settlement areas is a growing concern. In the Lesotho High- Box 4.6 Mercury and human health at Tucurui lands Project Area, infection rates are far higher than in surrounding areas.107 Com- Mercury can have lasting negative impacts on human health. Levels at 50 to 125 mg/kg in human hair indicate a low risk of neurological damage. Foetal damage munities are concerned about transmission can occur at half the lower limit. Clearly defined neurological effects appear at from migrant workers arriving to work in concentrations of over 125 mg/kg. the Maguga project in Swaziland.108 In the early 1990s scientists from the University of Helsinki in Finland carried out a series of studies at Tucurui to assess the origins and effects of mercury in tropical reservoirs, with co-operation of Brazilian institutions (including Initial assessment and other available Electronorte – the utility operating Tucurui). The studies provide the following findings: information is often not considered until the ■ Tucunaré fish caught at five locations had an average of 1.1 mg/kg net impacts manifest themselves in alarming weight of mercury, more than double the maximum safety level of 0.5 proportions and mitigation measures are mg/kg. unprepared and inadequate. When the ■ the average concentrations of mercury found in hair taken from adults in the fishing community was 47 mg/kg (with a standard deviation of Diama and Manantali dams were filled in 10.2mg/kg); the mid-1970s, an epidemic of Rift Valley ■ one individual was found with a concentration of 240 mg/kg ; Fever occurred, schistosomiasis prevalence ■ this concentration was seven times that of non-fish eating people, adults rates reached record levels and riverside in this group had fish at meals 14 times a week; and inhabitants experienced diarrhoeal disease, ■ the main source of mercury is gold mining operations upstream. malnutrition and malaria.109 All this oc- The WCD Case Study revealed considerable lack of agreement with the results curred despite the experiences with the of the study on the part of Electronorte. Given the irreversible, accumulating, and serious nature of the health impacts of mercury poisoning further research transmission of these diseases from earlier to resolve this issue is of great importance.105 African dams. Among the impact forecasts Sources: WCD Tucurui Case Study; Jobin, 1999, p175. for the Tucurui region were the results of a

The Report of the World Commission on Dams 119 Chapter 4

those affected, this translates peoples), or unrepresented (such as future After Tucurui was filled into an increase in pain and generations). Considering present societal in 1984, an unusual suffering and a decline in commitments to human rights and sustaina- proliferation of educational achievement ble development it is clear that dams in the Mansonia mosquitoes in and productivity. It can also Knowledge Base have led to inequitable rural areas close to the put severe pressure on the outcomes. In the light of the range of reservoir forced farm capacities of public health opportunities that exist for making affected families to leave their systems in developing people beneficiaries and enabling them to homes. countries. contribute to the stream of project benefits, such outcomes are unacceptable on equity Equity and the Distribution grounds.

of Costs and Benefits In the past, if the expected benefits of a dam To this point, the Global Review has exceeded the predicted costs the project identified – and in some cases quantified in went ahead. The narrow nature of the physical, economic, environmental, or technical and economic analyses undertak- social terms – a series of costs and benefits en does not necessarily mean that public that arise from the decision to build a large authorities that chose dams as a develop- dam. As indicated, shortfalls in technical, ment option were unaware of the social and financial, and economic performance led a environmental costs. Rather, within the significant number of the large dams in the context of knowledge available and the WCD Knowledge Base to fail a limited post value system of those making decisions at construction cost-benefit evaluation using any given time, the sacrifices were judged to the metric on which they were approved. be worth the benefits of pushing ahead with The Global Review has also established that the project. This approach to decision- building and operating dams has serious – making continues largely intact today. and largely negative impacts – on ecosystems, biodiversity and human livelihoods. The emergence of equity as a critical ingre- This chapter has documented a wide range dient of development underlines that this of adverse social impacts due to large dams, ‘balance sheet’ approach is unacceptable as including displacement, health and cultural it ignores the typical mismatch between the heritage impacts. Large dams have also distribution of the gains and losses of a provided substantial socio-economic bene- project across different societal groups. Large fits through the delivery of water, electricity dams can be seen as an extreme example of and flood control, as well as various ancil- this dilemma, as public resources – both lary services. In many cases these benefits monies and rivers – are devoted to projects extend beyond the time frame proposed in that all too often result in inequitable the original project documents distribution of costs and benefits.

This section shows that large dams in the This section uses the benefits and costs of WCD Knowledge Base tend to produce the large dams in the WCD Case Studies to benefits that accrue to groups other than illustrate and examine these issues. First, it those who bear the social and environmen- presents the typical approach to balancing tal costs. Those who bear the costs are quite costs and benefits along with an overview of often poor, vulnerable (such as indigenous the benefits generated by the dams in the

120 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

case studies. Then it discusses the distribu- (see Table 4.1). The Kariba dam, for in- tion of project gains and losses to different stance, provided inexpensive electricity to groups with the assistance of a review of the the mining sector and urban areas of Zambia groups adversely affected by the case study and Zimbabwe and also contributed to the dams. Using this information it draws the economy through tourism, irrigation and implications for equity and the balance-sheet commercial fishing, creating thousands of approach. The chapter closes with a presenta- jobs, directly or indirectly, in these sectors. tion of recent initiatives that have addressed Tucurui dam supplies electricity to industry the past legacy of inequity and provided and to two Brazilian states (Para and Mara- guidance on mechanisms for benefit sharing. nhao) that depend almost entirely on it for their supply (97% and 100%, respectively). Benefits, costs and the balance sheet approach to net benefits The 84 000 ha of land irrigated by the Each WCD Case Study dam generated a Aslantas dam has led to substantial im- number of services and benefits for people provements in the living standards of

Table 4.1 Illustration of the services and benefits generated by large dams in the WCD Case Studies Services and benefits Socio-economic significance to beneficiaries Glomma & Laagen Basin ■ Electricity benefited 1.3 million basin residents, landowners, and industries; Public revenues from Electricity: 10 145 GWh/yr (1998) power installations in G&L area of around $60 million in 1998 of which 80% went to G&L region.; 2,350 people are permanently employed in the G&L area power sector Flood protection ■ Flood peaks reduced by 20% Grand Coulee ■ Supply to industrial sector, agriculture and urban areas. Electricity: 24 050 GWh/yr (1995-1998) ■ About 1 400 farms; value of production : US$637 million (1998) Irrigation: 276 700 ha irrigated Tourism: 3 million visitors/yr ■ Thousands of jobs created to service 44 major tourist facilities and parks and recreation areas in the project area. Kariba Electricity: 3 860 GWh/yr (1996) ■ Supplying mainly the mining sector, urban populations and commercial farmers Tourism: 485 000 visitors/yr ■ 20 hotels (about 1,000 beds) established on the reservoir. Fishing: 23 250 tons of kapenta/year from ■ The fishery benefits mainly commercial kapenta fishing companies Kariba Lake Irrigation: 2 700 ha irrigated ■ 450 permanent and 3,000 casual jobs in irrigation schemes Tarbela Irrigation: 9% of total annual water supplies ■ Benefiting millions of holders of irrigated land and associated workers (22 % in dry season) for the 18 million hectares of the Indus system Electricity: 15 100 GWh/yr (1998) ■ Provides 28% of Pakistan’ s annual power generation from the inter-connected national grid system Tarbela employment ■ 4 000 permanent jobs in the project; secondary employment in agro-industry and irrigation Gariep and VanderKloof Irrigated area: 138 000-164 000 hectares110 ■ 40 000 job units created or saved (regular and seasonal labour combined) Electricity: 660 GWh/yr (avg 1971 to 1998) ■ Supply to national grid Water Supply: 151 Mm3/year ■ Supply to industrial sector and cities through inter-basin transfers Tourism: 200 000 visitors/yr ■ 200 people employed in 18 recreational facilities identified (with at least 1 000 beds) Aslantas Irrigated area: 84 000 hectares111 ■ Over 11 000 farm families or 80 000 people (but decrease in total farm labour force); Gross value of production $160 million, Supports about 100 agro-industrial units Electricity: 650 GWh/yr (avg 1995-99) ■ Municipalities and industries in the basin Tucurui Electricity: 20 000 to 23 000 GWh/yr (1995- ■ Aluminum industry consumes over half of the electricity. 8 million people supplied with 1998) electricity in northern Brazil Reservoir Fisheries: 3 200 tons/yr ■ Commercial and subsistence fisheries Pak Mun Electricity: 290 GWh/yr (avg 1995-99) ■ Supply to Northeast region of Thailand Tourism (existed pre-project) ■ 140,000 visitors in Keng Saphue in 1999 (less than pre-project) Source: WCD Case Studies.

The Report of the World Commission on Dams 121 Chapter 4

thousands of farmers, while ■ loss of 285 000 hectares of tropical forest the electricity supplied by the and associated populations and species of dam has benefited municipali- wildlife; ties and industries in the basin. ■ the suffering of communities subject to Grand Coulee dam in the forced displacement and the plague of United States irrigates more Mansonia mosquitoes; than 276 700 ha, generates ■ the risk of mercury poisoning; 24 050 GWh per year, and ■ the loss of downstream fisheries and together with Lake Roosevelt agricultural productivity; and and the Columbia Basin ■ irrigation project is the basis the physical and livelihood displacement for a flourishing tourism sector with 3 million of the Parakana, Gavio da Montanha, visitors annually. Tarbela dam generates and Asurini peoples. 28% of Pakistan’s electricity. In addition, the dam provides over one-fifth of the water Furthermore, there is no agreement on in the dry season for the massive Indus Basin whether the net greenhouse gas emissions irrigation system. Gariep and Vanderkloof from the reservoir, spillway, and turbines are dams in South Africa irrigate more than offset by the saving in emissions from fossil 160 000 ha in the most arid provinces of the fuel sources made possible by the large country and, despite the fact that they were amount of power produced by Tucurui. built during the Apartheid era, they created (or saved) jobs that mostly went to then- Clearly, in the case of large dams that marginalised groups. Through inter-basin produce a multitude of benefits and costs, transfers, the Orange River Development the application of a balance-sheet approach Project dams have also supplied fresh water is not easy to do in a comprehensive fash- to distant cities and industrial areas. ion. In many cases the benefits and costs (especially social and environmental costs) In many cases the exact number of benefici- are not of the same currency and cannot be aries is not known, particularly where the explicitly weighed against each other. water or power is connected to a common transmission or distribution system. Still, it is The Commission asked participants in the possible to roughly approximate these num- final WCD Case Study stakeholder meet- bers, for example, the hydropower dams in the ings for their views on the overall develop- Glomma and Laagen system supply electricity ment impacts of the WCD Case Study to 1.3 million people living in the basin. dams. Some stakeholders were able – in their own minds – to sum up the positive These benefits need to be viewed in the and negative aspects of a project. Others context of the costs implied, especially the rejected the questions as too simplistic an economic, social and environmental costs approach to a very complex, multi-dimen- documented earlier in this chapter. For sional problem. The Global Review itself example, in the case of Tucurui the economic demonstrates that poor accounting in costs and benefits of power production must be economic terms for the social and environ- balanced against a series of social and environmental costs and benefits of large dams mental impacts, including those already implies that the true economic efficiency referred to in this chapter and Chapter 3: and profitability of these schemes remains

122 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

largely unknown. In addition, as covered in transform them into another set of more detail in Chapter 6, even efforts to resources – a reservoir, hydro power and extend traditional cost-benefit analysis to irrigation, providing benefits to people cover this task have failed to accommodate living elsewhere. There is a sense, the myriad impacts and concerns that arise. therefore, in which large dams export rivers and lands, removing them from Who gained and who lost? the productive domain of one community to make them available to another.112 One of the drawbacks of the approach taken above is that it reduces the benefits and Grand Coulee provides a vivid example. costs of large dams to abstract flows (or Native Americans were physically dis- numbers), and hides the impacts that have placed by a project that provided power to had very real consequences for people and industry and households in a city ecosystems. An analysis of the distribution some 250 km away. Furthermore, of costs and benefits – that is who gained the water and land that had Poor accounting in and who lost – provides another method of previously supported their economic terms for the assessing costs and benefits of large dam livelihoods (particularly the social and environmental projects. salmon fishery) was dammed costs and benefits of Re-distribution of access to and diverted to provide white large dams implies that resources settlers with irrigated farmland. the true economic efficiency and The construction of a dam requires an profitability of these investment of human-made capital and, as a In other cases, the resources as schemes remains largely result, generates a series of new entitlements transformed simply continued to unknown. that are then distributed – either through benefit those who previously had political-administrative means or through access to them. The Tarbela dam markets – to members of society. At the illustrates this: in addition to same time, the construction of a large dam securing existing water supply, the expan- will have profound effects on the natural sion of the cropped area by 39% and the and social landscape of the setting in which increase in the cropping intensity have the dam is located. These changes will affect primarily benefited those owning land prior the legal, customary, or de facto entitle- to the construction of the dam. ments to natural resources, environmental quality and socio-cultural integrity experi- In the cases listed the distribution of project enced by local communities and others benefits was explicit. In other cases, however, holding rights to the resources in the area. particularly where ancillary benefits are concerned, the redistribution of resources Dams are unique among large infrastructure and benefits happens in a haphazard fash- projects in the scope and manner in which ion. The illustration provided earlier of the they affect the pattern of access to resources, increase in fisheries in the Tucurui reservoir and their distribution across space, time, and is an example. In this case, those living near societal groups: the reservoir have benefited while those living downstream from the dam have seen dams take a set of resources – a river their fisheries decrease substantially. Simi- and the lands along its banks, generating larly, the India Case Study noted that in food and livelihood for local people; and many irrigation dams, people lose access to

The Report of the World Commission on Dams 123 Chapter 4

their forest and highly preserved effects on the environment will likewise The WCD Case Studies natural resources along the river to have many lasting and even irreversible show that the direct give way for a dam that irrigates environmental impacts. adverse impacts of land for downstream people who dams have fallen have often overexploited their Profile of adversely affected groups disproportionately on local natural resources. These The WCD Case Studies show that the rural dwellers, examples show that even if no direct adverse impacts of dams have fallen subsistence farmers, intentional redistribution is made, indigenous peoples, disproportionately on rural dwellers, subsist- a dam can effectively take a ethnic minorities, and ence farmers, indigenous peoples, ethnic resource from one group and women. minorities, and women (see Table 4.2). allocate it to another. These groups, who are also often the poorest segments of society, tend to be over-repre- Finally, the impacts of large dams sented in the numbers of people who are on ecosystems and biodiversity have conse- displaced from reservoirs sites or lose access quences for future generations. Dams are to their traditional livelihoods. Within the long-lived assets and thus are designed to displaced, compensation programmes tend provide benefits beyond the medium term. to ignore livelihood impacts on landless The decision to dam a river and the ensuing groups and women. In downstream areas,

Table 4.2 Profile of groups adversely affected by large dams: illustrations from WCD Case Studies Dam project Displaced people Profile of displaced people Others adversely affected Glomma and Laagen None displaced. None documented in case study. 1945-70 Grand Coulee 5 000 to 6 500 1 300 to 2 000 from Colville and Spokane tribes Several thousand members of First Nation groups 1934-75 people. not compensated until 1990s; rest are settlers. (Colville, Spokane, Nez Perce and Canadian tribes) located throughout the upstream basin, due to flooding of fishing falls and blockage of salmon. Kariba 57 000 people. Tonga subsistence farmers; ethnic minority in Thousands of downstream people lost floodplain 1955-59 Zimbabwe; most were resettled in resource- livelihoods and lakeside inhabitants experienced depleted areas. increased prevalence of schistosomiasis. Tarbela 96 000 people. Composed of 93% farmers; 5% artisans or semi- Pastoralists, landless people, low cast groups of 1968-76 skilled workers; and 2% boatmen. fisherfolk, boatmen, basket makers, and weavers suffered from loss of wetlands, forests, and grazing. Gariep and 1 380 people: Most white farmers felt they were fairly Farmers living along the river, where 1 million Vanderkloof families of 40 white compensated. Black farm-workers were not sheep suffered from proliferation of biting 1963 to late 1970s farmers and 180 eligible for compensation and eventually moved blackfly leading to livestock losses. black farm workers to other farms and urban areas. Aslantas 1 000 families. Mostly former immigrants from Eastern European None documented in case study, although it is 1975-85 countries; composed of small and medium-size claimed that customary users of forests that were landowners and landless families. not recognised were the most adversely affected. Tucurui 25-35 000 people Subsistence farmers, fisherfolk, pastoralists, and 100 000 people (subsistence farmers, pastoralists, 1975-85 plus riverbank cultivators (126 of 2 247 disputes are fisherfolk, and other natural-resources-dependent indigenous peoples. not yet settled.) communities) affected by reduced water quality, loss of riverbed cultivation, and decreased downstream fish populations. Pak Mun 1 700 families. Rural families dependent on rice farming and More than 6 000 families of subsistence farmers 1991-94 fisheries income; cash compensation failed to and fisherfolk suffered loss of livelihood from provide livelihood regeneration. fisheries reduction. Source: WCD Case Studies.

124 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

communities suffering from altered river Aslantas, Grand Coulee) such flows are mainly subsistence farmers whose schemes also benefit large livelihoods are largely based on the exploi- landowners and those from tation of resources offered by the natural privileged groups (Orange flow of the river (fisheries, floodplain River). Electricity generation has farmlands and pastures). mostly benefited the industrial and mining sector (Kariba, Finally, to the extent that burgeoning popula- Tucurui) and urban areas (Grand tions are at once evermore reliant on a Coulee, Pak Mun). Flood control degraded natural resource base and demand is provided to urban areas more environmental quality (as incomes rise), (Grand Coulee as part of the future generations are also likely to be Columbia River system) as well among the adversely affected. Where the as rural areas (Glomma and Laagen). Fresh performance of large dams in economic terms water supplied by dams is mainly directed to falls short, future generations may also bear a the industrial sector and urban areas. disproportionate share of the costs without a Equity commensurate share of the benefits. The Case Study dams demonstrate a lack of Profile of beneficiaries connection between groups benefiting from As noted earlier, dams have benefited the dam projects and those adversely affected, general public through their contribution to confirming the general finding from the food production and increased access to Knowledge Base. Assessment of these electricity, along with providing other direct distributional impacts of a dam does not in benefits and multiplier effects. The preced- and of itself provide a value system to ing sections of this chapter have also includ- prescribe what should be done in judging ed some specific examples of how the direct projects and confronting the distributional and indirect benefits of large dams have issues raised. Judging the social and moral been shared with the poor, such as through acceptability or fairness of the patterns of the provision of water supply, electricity, distribution of positive and negative impacts employment on construction sites and jobs is an equity issue. The debate on the social in dam-based irrigation, industrial and impacts of dams is thus about the equity tourism industries. dimension – whether the distribution of costs and benefits is judged to be fair or just. Examination of the Case Study dams A strong equity constraint dictates that confirms that those who receive the bene- there be no losers, that those who bear the fits, usually urban dwellers, commercial costs from large dams should receive a farmers and industries, are typically not the proportionate amount of benefits. However, same groups that bear the social costs. equity often speaks more to the costs of Immigrants generally provide most of the dams than to their benefits, directing labour force in construction works (qualified attention to those who are most at risk. This and non-qualified jobs). They develop and speaks to the vulnerabilities of isolated, less operate tourism facilities and manage powerful populations for whom ‘develop- commercial fishing companies (Kariba). ment’ can all too easily mean loss. Put While many small to medium farmers simply, a large dam that renders the poor participate in irrigation schemes (Tarbela, and the vulnerable worse off is inequitable.

The Report of the World Commission on Dams 125 Chapter 4

Implications for the balance-sheet selves, invalidate the balance-sheet ap- approach proach. Rather, the crucial distinction is the That large dams in the Knowledge Base failure to balance the loss of entitlements have led to inequitable outcomes challenges that some groups experience with a corre- the assumptions that underpin the ‘balance sponding gain in new entitlements. sheet’ approach and hence the idea that simply ‘adding’ up the costs, benefits, and The lack of validity of the balance-sheet impacts of large dams will lead – all things approach in such a situation is confirmed by considered – to the best choice for society. economic theory. Where costs and benefits Indeed, there are significant moral and accrue to different groups, the standard ethical concerns that such a balance-sheet procedures for adding up and discounting the exercise does not address. It is implicitly expected costs and benefits do not provide an based on the assumption appropriate measure of changes in societal that if the overall balance of welfare.113 In order to apply the balance sheet Where costs and benefits impacts is positive, then approach equitably, the costs to affected accrue to different those who gain would share groups need to be minimised and an equita- groups, the standard the benefits with those who ble share of benefits ensured. With regard to procedures for adding up lose out. Thus, all would be the environment and intergenerational equity and discounting the better off in the end. How- this implies the need to ensure that ecosystem expected costs and ever, examination of the needs are met in the present so that future benefits do not provide distribution of gains and generations can have access to a non-declin- an appropriate measure losses in the case studies ing income stream, to which natural capital of changes in societal demonstrates that such makes an important contribution.114 welfare. benefit sharing has seldom occurred. Those who bear Initiatives for the equitable the social and environmental costs and risks distribution of costs and of large dams are frequently not the same benefits people who receive the social and economic benefits of the water, electricity, and ancil- The major equity issues arising from the lary services that dams produce. discussion of the distributional impacts of large dams is the impoverishment of those The importance of the distribution analysis who previously inhabited the reservoir site of impacts and the concept of equity to and those who derived their livelihood from decision-making can be understood as the resource base that is transformed by dam follows. If the loss of access to previous construction and operation. There is injus- sources of livelihood is offset by access to tice when the rights of physically displaced new benefits made available by the dam, people are violated, including when they former resource owners and users can have lose their land and access to the river and different but better living conditions than when downstream people experience before. If loss of ecosystem function results reduced access to floodplains and fisheries, from large dam projects the resulting costs but are excluded from access to project may be included on the balance sheet. In benefits. Societies are increasingly rejecting other words, adverse social and environmen- these outcomes and searching for more tal impacts of large dams do not, by them- equitable solutions.

126 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

Chapter 3 has already covered the measures including compensation for property and other that are available to avoid, minimise or losses, improvement of the living conditions reduce the ecosystem impacts of large dams. and industrial base of the affected area, and These are likely to be an important compo- measures for resettling people through the nent of efforts to resolve intergenerational Fund for Reservoir Area Development. The issues or social inequities that are linked to beneficiary municipalities, affected munici- ecosystem impacts. This final section of the palities and central government contribute chapter describes some recent initiatives to to this Fund, which finances development explicitly confront past social inequities in the reservoir area. The Fund also pro- through reparations or to ensure that new motes solidarity between the downstream projects deal with equity issues in a proac- beneficiaries and the displaced people.119 tive manner through benefit sharing. In the Senegal River valley, state-led Progressive national legislation and distribution of irrigated land gave lower policies caste groups access to land ownership, Adopting a benefit-sharing approach previously denied to them under traditional requires that the project design and plan- tenure systems.120 The India Case Study ning process consider such mechanisms from provides examples of agriculture and home- the outset.115 Progressive national legisla- stead land being provided even to those who tion and policies provide the legal frame- were landless as part of resettlement process- work and standardise benefit sharing, thus es. In some cases, previously marginalised having a far broader impact than project farmers were given more land than they had level approaches. This approach has been originally. widely implemented in the energy sector where project proponents allocate a per- A comparison of access to electricity in centage of the electricity sales revenue to Zimbabwe and South Africa documents the resettlers and local administrative units.116 Examples include, the Lubuge, Yantan, Shuikou and Ertan Hydroelectric projects in Box 4.7 Royalties to communities: a Brazilian law for hydropower benefit-sharing China, the Rio Grande Hydroelectric Project in Colombia, as well as several In Brazil, Law No. 7990, dated 28 December 1989, requires that royalties be paid to the federal government for using water for power generation purposes. The projects in Brazil (see Box 4.7). Other royalties paid by each power plant generating more than 10 MW represent 6% mechanisms for benefit sharing include the of the value of the power produced. The royalties are distributed as follows: 10% to the federal government, 45% to the state(s) where the venture is located and supply of energy at preferential rates (as 45% to the municipal districts affected by the venture. The total amount paid required in Norway) and payment of proper- out by the Tucurui dam in 1996 reached $19 million, with the total royalties for 1991 through 1996 topping $103 million. The Itaipu dam, in the south of Brazil, ty or local government taxes (as required in pays annually about $13 million in royalties. The royalties are among the leading France and Norway) which are assigned to sources of income for some of the municipal districts. affected areas.117 However, royalties by themselves will not address social injustice, as the way they are used to benefit local government units depends on broader political and social factors. In some cases the allocation of these resources is done in a Japan’s Act on Special Measures for Reser- non-transparent way. In others, results are visible. A municipality like Itaipulandia has paved all the roads in the city and provides agricultural supplies to the voir Area Development provides various population. In addition, local young residents are funded to study at Brazilian measures for people who are affected by a universities on the condition that they return to the community for five years. dam project and for the development of Source: WCD Tucurui Case Study; Itaipu dam in Ferradas, 1999, WCD Contributing Paper for Thematic Review I.1 Social Impacts areas around the dam/reservoir.118 The Act provides for a combination of measures,

The Report of the World Commission on Dams 127 Chapter 4

importance of proactive govern- ties.124 In North America, the process of The inequitable ment policy and investment in paying reparation to Native Americans for distribution noted in the confronting existing inequities dam-induced impacts has started. The past in sharing risks and in the electricity sector as a Grand Coulee Case Study relates that in allocating benefits is not whole.121 Almost 40 years after 1994 the United States Congress upheld the inevitable. commissioning the Kariba dam, claims for damages and reparation made in only about one-fifth of Zimba- 1951 by Colville Confederated Tribes who bwean households (mainly located in urban had lost homes, lands and salmon runs to areas) have access to electricity. Prohibitive the Grand Coulee Dam in the Colombia power pricing policies adopted by the basin in the 1940s. A total of $54 million government continue to exclude poor was paid in reparations in addition to an people from this service. In contrast, as a annual payment of $15 million as long as result of deliberate policy on the part of the the dam continues to produce electricity. A post-Apartheid government in South $200-million Missouri River Trust fund Africa, the percentage of households there proposed through legislation in early 2 000 having access to electricity increased from aims to move American Indian Tribes closer 20 to 50% in the 1990s. to compensation for land lost to federal dams.125 Project level benefit-sharing mechanisms A few precedents are emerging on repara- Some project proponents also initiate tions in developing countries as well. benefit-sharing agreements with concerned Approval of a World Bank loan for the communities where national or local regula- Ghazi Barotha hydro project in Pakistan in tory frameworks do not exist. A wide variety December 1995 was made conditional on a of mechanisms are developed, ranging from process for the resolution of disputes over making affected communities primary compensation for people displaced 20 years beneficiaries of the project services – for before by the Tarbela dam. An initiative has example irrigation, electricity, water supply, begun in Zambia to raise the living stand- fishing rights – to formal business agree- ards of communities suffering the conse- ments concerning equity and revenue quences of displacement by the Kariba dam sharing in the project itself. Several dam more than four decades ago through the projects in the WCD Knowledge Base Gwembe Tonga Rehabilitation and Devel- shared direct project benefits with resettlers opment Programme. In a negotiated settle- by moving displaced people into the newly ment resulting from prolonged agitation, irrigated areas, for example Andhra Pradesh more than 10 000 families affected by Bargi II and III, Brazil Ceara Water Resources, project on the Narmada River in India China’s Daguangba Multipurpose project.122 gained rights to cultivate drawdown lands, In the case of Hydro-Québec in Canada, the fish in the reservoir and manage forests in Provincial power utility proposed partner- the catchment area jointly.126 ship agreements to local communities for all new hydropower projects.123 China attempted to systematically address the problems faced by the reservoir-resettled Reparations people beginning with new policy and There are an increasing number of examples institutional initiatives in the early 1980s. of reparations being made for past inequi- In 1986, the Ministry of Water Resources

128 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

and Electric Power launched a large rehabil- The WCD Knowledge Base pro- itation programme aimed at improving vides the following findings on the living conditions of some 5 million reser- adverse impacts of the displacement voir-resettled people across 46 resettlement of people from their homes and areas in the country. 127 Although such livelihood by large dams: measures have had a significant impact, a ■ 40-80 million people were 1994 World Bank report cited the Chinese physically displaced by dams government as saying that some 46% of the worldwide; country’s reservoir resettlers ‘were at great ■ millions of people living down- risk.’128 stream from dams – particularly those reliant on natural flood- An appropriate legal and policy environ- plain function and fisheries – ment, accompanied by clear political will to have also suffered serious harm act, can therefore ensure that poor and to their livelihoods and had the vulnerable groups marginalised in the past future productivity of their by large dam projects can begin to share in resources put at risk; the benefits generated by such projects. This means that the inequitable distribution ■ many of the displaced were not noted in the past in sharing risks and recognised (or enumerated) as such, allocating benefits is not inevitable. Indeed, and therefore were not resettled or as discussed in Chapter 6, there are a compensated; number of explanations for past failures in ■ where compensation was provided it performance – explanations that lead to a often proved inadequate and where the series of recommendations for making the physically displaced were enumerated transition to more equitable outcomes. many were not included in resettlement programmes; Findings and Lessons ■ those who were resettled have rarely had their livelihoods restored, as resettlement Past decision-making and planning efforts programmes have focused on physical have often neither adequately assessed nor relocation rather than on the economic accounted for the adverse social impacts of and social development of the displaced; large dams. As a result, the construction and operation of large dams has had serious and ■ even in the 1990s, impacts on down- lasting effects on the lives, livelihoods and stream livelihoods were not adequately health of affected communities, and led to assessed or accounted for in the planning the loss of cultural resources and heritage. and design of large dams; and At the same time a simple accounting for ■ there is a clear relationship between the the direct benefits provided by large dams – magnitude of displacement and the the provision of irrigation water, electricity, ability to rehabilitate and restore liveli- municipal and industrial water supply, and hoods adequately - the larger the number flood control – often fails to capture the full of displaced people, the less likely it is set of social benefits associated with these that livelihoods can be restored. services. It also misses a set of ancillary benefits and indirect economic (or multipli- In sum, the WCD Knowledge Base demon- er) benefits of dam projects. strates a generalised lack of commitment or

The Report of the World Commission on Dams 129 Chapter 4

lack of capacity to cope with displacement. social costs and were often discriminated Large dams in the WCD Knowledge Base against in the sharing of benefits. have also had significant adverse effects on cultural heritage through the loss of cultural These inequitable outcomes documented in resources of local communities and the the WCD Knowledge Base invalidate the submergence and degradation of plant and prevailing ‘balance-sheet’ approach to animal remains, burial sites and archaeological decision-making. The balancing of gains monuments. and losses as a way of judging the merits of a large dam project – or selecting the best The WCD Knowledge Base indicates that option – is not acceptable where the mismatch the poor, other vulnerable groups and future between who gain from the benefits and those generations are likely to bear a dispropor- who pay the costs is of such a serious, perva- tionate share of the social and environmen- sive, and sometimes irreversible nature. tal costs of large dam projects without gaining a commensurate share of the eco- The review also shows that the true economic benefits. Specific cases include: nomic profitability of large dam projects ■ Indigenous and tribal peoples and remains elusive as the environmental and vulnerable ethnic minorities have social costs of large dams were poorly suffered disproportionate levels of displace- accounted for in economic terms. More to ment and negative impacts on livelihood, the point, failures to account adequately for culture and spiritual existence; these impacts and to fulfil commitments ■ affected populations living near reser- that were made have led to the impoverish- voirs, displaced people and downstream ment and suffering of millions, giving rise to communities have often faced adverse growing opposition to dams by affected health and livelihood outcomes from communities worldwide. Innovative exam- environmental change and social disrup- ples of processes for making reparations and tion; and sharing project benefits are emerging that ■ among affected communities gender gaps provide the basis for optimism that past have widened and women have frequent- injustices can be remedied and future ones ly borne a disproportionate share of the avoided.

130 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

Endnotes 25 Fact Finding Committee on the Sri Sailam Project, 1986, cited in Mander et al, op cit, 1 Lang et al, 2000 eco041, WCD Submission. p10. 2 Meinzen-Dick, 1997, p50. 26 Stewart et al, 1996; World Bank, 1996b; 3 UNDP, 1999, p213, 214. Chen, 1999, WCD Regional Consultation 4 Saxena, pers comm, 2000; UNDP, 1998, Paper; Colajacomo, 1999, Contributing p175, 177. Paper for WCD Thematic Review I.2 Indigenous People. 5 These figures relate only to areas along the Orange River downstream of the Gariep and 27 WCD Thematic Review I.2 Indigenous Vanderkloof dams, and do not include People, section 2.1.9. irrigated areas along the Fish and Sundays 28 Mander et al, op cit, p6. rivers, which received water, diverted from 29 WCD Thematic Review I.3 Displacement, the Orange River by ORDP as well. section 3.2; Bartolome and Danklmaier, op cit. 6 Panama Canal Office of Public Affairs, 30 Kiambere dam in Mburugu, 1994, p53. Note undated. that in Turkey farmers are asked to register 7 ADB, 1999b, p1. the value of their land for tax purposes. 8 Jing, 1999, WCD Contributing Paper to Farmers that record a lower value than the Thematic Review I.3 Displacement. real asset value in order to avoid tax will 9 Fernandes and Paranjpye, 1997, p17. have difficulty in purchasing equivalent land elsewhere as the government uses these 10 World Bank, 1996a, p90-92. declarations as the basis for calculating land 11 Cook, 1994, p25. compensation amounts. 12 OED, 1993, p11. 31 Fact Finding Commission on Sri Sailam 13 World Bank, 1996a, p88. Project, 1986, cited in Mander et al, op cit; 14 WCD Thematic Review I.2 Indigenous Kao Laem dam in WCD Thematic Review People, section 2.1.7. I.2 Indigenous People. 15 WCD Thematic Review I.2 Indigenous 32 Aswan High dam in Fahim, 1981, and People, section 2.1.11. Fernea, 1998, cited in De Wet, op cit, p11; 16 Morse and Berger, 1992; Parasuraman, 1999, Nangbeto and Akosombo in De Wet, op cit, p83. p11; Itá in Bermann, 1999, WCD Regional Consultation Paper; Bhumibol in Sluiter, 17 Correa, 1999, WCD Regional Consultation 1992, p62. Paper. 33 Jing, op cit, p11. 18 Dhom dam in Parasuraman, 1999, p154; Bargi dam in Mander et al, 1999, Contribut- 34 Hoa Binh dam in Srettachau et al, 2000, ing Paper for WCD Thematic Review I.3 WCD Regional Consultation Paper; Sirind- Displacement, p64. horn dam in Sluiter, op cit, pvii; Batang Ai dam in ADB, 1999a, p5. 19 Bartolome and Danklmaier, 1999, Contribut- ing Paper for WCD Thematic Review I.3 35 Miguel Aleman and Cerro de Orro dams in Displacement. Nahmad, 1999, WCD Regional Consultation Paper; Panama in Huertas and Pacheco, 20 Balsalm, 1940a, and 1940b. 1999, WCD Regional Consultation Paper; 21 WCD Grand Coulee Case Study. Brazil in WCD Tucurui Case Study; Zambia 22 WCD Thematic Review I.3 Displacement, and Zimbabwe in WCD Kariba Case Study. section 3.1. 36 Sluiter, op cit, p62. 23 WCD Thematic Review I.3 Displacement, 37 Sluiter, op cit, p63. section 5. 38 Jing, op cit, p7. 24 Colson, 1971, cited in De Wet, 1999, WCD Contributing Paper for Thematic Review I.3 39 Cernea, 2000; WCD Thematic Review I.3 Displacement, p9. Displacement, section 1.3.

The Report of the World Commission on Dams 131 Chapter 4

40 World Bank, 1994, p2-3; Jing, op cit, p5. tion of 1987; in India it is the fifth and sixth 41 Cernea, 2000, p2. schedules under the Indian constitution; in Brazil it is the Article 231& 232 of the 1988 42 OED, 1996b, p86. constitution. National laws reflect contem- 43 WCD Thematic Review I.2 Indigenous porary indigenous rights norms, in Chile, People, section 2.1.10. Ecuador, Bolivia, Colombia, and Argentina. 44 Jing, op cit, p35. 69 Adams, 1985, cited in WCD Thematic 45 Supreme Court of India, 1999. Review I.1 Social Impacts. 46 ADB, 1999b, p20-21. 70 WCD Thematic Review I.1 Social Impacts. 47 De Wet, op cit, p18. 71 Horowitz et al, 1994. 48 Robinson, 1999, Contributing Paper for WCD 72 Ferradas, 1999, Contributing Paper for WCD Thematic Review I.3 Displacement, p4. Thematic Review I.1 Social Impacts. 49 Driver, 2000, Submission to WCD Thematic 73 WCD Thematic Review I.2 Indigenous Review 1.3 Web Conference. People, section 2.1.11. 50 Jing, op cit, p18-19. 74 Urrá I dam in Correa, 1999, WCD Regional 51 Bermann, op cit. Consultation Paper; Pak Mun dam in WCD 52 De Wet, 1999, op cit, p21. Pak Mun Case Study; Other dams in ADB, 53 World Bank, 1993, p18. 1999a, p23-24. 54 WCD Thematic Review I.2 Indigenous 75 Mehta and Srinivasan, 1999, Contributing People; WCD Grand Coulee Case Study, Paper for WCD Thematic Review I.1 Social Annex 9; WCD Tucurui Case Study. Impacts. 55 WCD Thematic Review I.3 Displacement. 76 OED, 1998. 56 WCD Thematic Review I.2 Indigenous 77 Projects included in the studies are: Batang People, section 2.1.1. Ai in Malaysia, Sing Karak in Indonesia, Lingjintan in China and Theun-Hinbon in 57 WCD Thematic Review I.2 Indigenous Lao PDR. ADB, 1999a. People. 78 Agarwal, 1996, cited in Mehta and Srinivas- 58 WCD Thematic Review I.2 Indigenous an, op cit. People, section 2.1.6. 59 Silva Orrego, 1997, p159; Opaso, 1999, 79 While most adivasi communities in the WCD Regional Consultation Paper. Narmada Valley are classified as ‘encroach- ers’, they had usufructory rights and control 60 Gapuz and Shalupirip, 2000, WCD Regional over land. Mehta and Srinivasan, op cit Consultation Paper. 80 Colson, 1999, cited in Mehta and Srinivas- 61 WCD Thematic Review I.2 Indigenous an, op. cit. People, section 2.1.1. 81 Anane, 1999 soc210, WCD Submission. 62 WCD Thematic Review I.2 Indigenous People, section 2.1.6. 82 Mehta and Srinivasan, op cit. 63 Huertas and Pacheco, op cit. 83 Colson, 1999, cited in Mehta and Srinivas- an, op cit, p.12. 64 WCD Thematic Review I.2 Indigenous People, section 2.1.5. 84 Mehta and Srinivasan, op cit, p22. 65 Soong, 2000, WCD Regional Consultation 85 Gender compares men and women. Where Paper. both benefit, but men benefit more than 66 WCD Thematic Review I.2 Indigenous women, the gender impact can be negative People, section 2.1.5. because the benefit/s in question can result in wider gender gaps. 67 WCD Thematic Review I.2 Indigenous People, section 1.1.4. 86 Tamakloe, 1994. 68 In Canada it is the Constitutional Law of 87 van Koppen, 1999. 1982; in the Philippines it is the Constitu- 88 Niasse, 1997.

132 Dams and Development: A New Framework for Decision-Making People and Large Dams – Social Performance

89 Brandt and Hassan, 2000, WCD Working 111 Area irrigated before dam is included. Paper on Cultural Heritage Management. 112 Brody, Contributing Paper for WCD Thematic 90 Gwala, 2000. Review I.1 Social Impacts, section 5.5. 91 Kinahan, 2000, in Brandt and Hassan, op cit, 113 Arrow and Lind, 1970; World Bank, 1980; p18. Belli et al, 1998, cited in WCD Thematic Review III.1, Economic Analysis, Chapter 5 92 Norr et al, 2000, in Brandt and Hassan, op and 6. cit, p35-36. 114 The extent to which other types of capital 93 Ota, 2000, in Brandt and Hassan, op cit, p52. can substitute for natural capital is debated. 94 Faught, 2000, in Brandt and Hassan, op cit, The degree of substitutability, as well as the p11. degree of irreversibility of ecosystem impacts 95 Brandt and Hassan, op cit, p59. will be important determinants of the 96 Politis and Endere, 2000 in Brandt and optimum balance between leaving a river in Hassan, op cit. its ‘natural’ state or going ahead with a dam (or determining environmental flow require- 97 Childs-Johnson, 2000, in Brandt and Hassan, ments). WCD Thematic Review III.1, op cit. Economic Analysis, Chapter 7. 98 WHO, 1999, WCD Working Paper on 115 Van Wicklin, 1999 soc184, WCD Submis- Human Health, p6. sion, p8. 99 Kariba dam in Hira, 1969, and Mungomba 116 Van Wicklin, op cit. et al, 1993; Akosombo and Aswan dams in 117 Milewski et al, 1999, soc 196,WCD Submis- Jobin, 1999, p278, 298-300. sion and Adeler and Flatby, pers comm, 100 Jobin, op cit, p300-303, 327-330, 425-427. 2000. 101 World Bank, 1999b, p2. 118 Kuriki, pers comm, 2000. 102 WHO, op cit, p21. 119 Kuriki, op cit. 103 Government of India, 2000, in WCD India 120 Niasse, 1991, p101. Country Study. 121 Bond, 2000 eco033, WCD Submission, p6-7. 104 WHO, op cit, p19. 122 Van Wicklin, op cit. 105 For a treatment of the value of information 123 Milewski et al, op cit. in the presence of uncertainty and irreversi- 124 Reparation is defined as action or processes bility see WCD Thematic III.1 Economic that repair, make amends, or compensate for Analysis, Chapter 7. damages. In a legal sense, there are three 106 China in Jing, op cit, p10; Vietnam in generally recognised forms of reparation: Sluiter, op cit. pVII; Malaysia in ADB, restitution, indemnity (or compensation), 1999a, p5, 2000; Thailand in Sretthachau et and satisfaction Johnston, 2000, Contribut- al, 2000, WCD Regional Consultation Paper; ing Paper for WCD Thematic Review I.3 India in Laxman, 1999, p208. Displacement, p14. 107 Macoun et al, 2000. 125 Johnston, op cit, p42. 108 Mncina and Ginidza, 1999. 126 Sinha, 1998, soc009, WCD Submission. 109 WHO, op cit, p12. 127 Jing, op cit, p5. 110 Area irrigated before dam is included. 128 World Bank, 1994, p2-3 cited in Jing, op cit.

The Report of the World Commission on Dams 133 Options for Water and Energy Resources Development

Chapter 5 Options for Water and Energy Resources Development

art of the Commission’s Pmandate was to look at the alternatives for energy and water services that were considered in the past when building large dams and to

consider the current options. This chapter examines the current state of knowledge on existing and emerging

options for meeting water and electricity needs. As part of the larger discussion of planning and

decision-making processes, Chapter 6 critiques the past assessment of alternatives to large dams.

The Report of the World Commission on Dams 135 Chapter 5

Options normally emerge in response contexts and provides snapshots of opportu- to demand or supply. The choices nities in specific countries, regions or available to a society at any given contexts. time also depend on factors such as natural resource endowments, tech- The investigation of options is organised nological capability, institutional around the four ‘needs’ areas that are the capacity, finance, market conditions, focus of this report: agriculture, energy, cultural preferences, awareness and water supply and flood management. Broad- education. These can act either as ly, options consist of technological, policy barriers or as enabling conditions, and institutional responses. They may be depending on whether they impede categorised further based on whether they or promote the consideration and contribute to demand-side management adoption of a particular option. (DSM), supply-side efficiency or represent Creating conditions for certain new supply options. For example, policies options to emerge as competitive and institutional options to improve man- responses to demand and supply agement of existing systems may respond to requires support. Policies, institutions supply-side efficiency, while a new dam and regulatory measures can either represents a technological option for new help or hinder innovation, moderni- supply. Previous chapters have presented sation, maintenance, continuation and analysed the contribution of large dams and sustainability of different options. to these services and the performance of large dams over time. Chapter 5 focuses on The chapter focuses on identifying the range the alternatives, locating large dams in the or mix of options available today to meet larger mix of options. water and electricity needs in different societies and in urban and rural settings. It This report confirms that selecting the most documents the large range of generic appropriate combination of options depends options currently available. However, given on giving all the options equal and appropri- concerns about a number of barriers that ate consideration in any assessment process. have led to limited assessment of options in Assessment should be based on the respec- the past, it is not enough simply to identify tive merits of available options in the given the technologies and policies that can satisfy context and should include not just a set of water and energy needs. It is also necessary technical, financial, and economic criteria, to identify the obstacles that prevent the but also full integration of social and envi- more widespread adoption and use of various ronmental criteria. The options listed below options. Obstacles may be generic to an are not exhaustive and the Commission option – such as the high cost of a technolo- does not endorse particular options. Rather, gy – or they may be specific to a particular the intention is to highlight the options and context – such as limited wind potential. issues that should be considered and explored Only a thorough and integrated examina- as part of the options assessment process. tion of the options and obstacles can yield a Much more detailed information on these and precise list of alternatives for consideration other choices can be found in the WCD in a given regional, country or local context. Thematic Reviews on irrigation, electricity, The chapter therefore indicates options that water supply and flood management options represent significant opportunities across all and the related contributing papers.1

136 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

Agriculture and Irrigation with the remainder supplied from direct river abstraction, Efforts to promote sustainable water man- groundwater and traditional agement practices have necessarily focused water harvesting systems.3 Since on the agricultural sector as the largest the 1970s, the predominant consumer of freshwater. Governments have focus has been on providing several objectives in deciding the nature and irrigation to support the green extent of inputs in agriculture. These revolution package of hybrid include achieving food security, generating seeds, chemical fertilisers and employment, alleviating poverty and pesticides. Conditions for higher producing export crops to earn foreign growth were created in such areas through exchange. Irrigation represents one of the subsidised infrastructure, agricultural inputs inputs to enhance livelihoods and achieve and electricity for pumping. economic objectives in the agricultural sector with subsequent effects for rural Irrigated agriculture has contributed to development. growth in agricultural production worldwide, although inefficient use of water, Just as strategies and approaches to rural inadequate maintenance of physical systems development are context-specific, there are and institutional and other problems have numerous and diverse alternatives to agri- often led to poor performance. Emphasis on cultural development and irrigation that large-scale irrigation facilitated consolida- need to be examined. The diversity relates tion of land and brought prosperity for to scale, level of technology, performance, farmers with access to irrigation and mar- and appropriateness to the local cultural and kets. Chapter 4 documents the socio-economic setting. Government major multiplier effects produced policies and institutions play an important by successful large irrigation Agriculture in Asia and role in the promotion of particular water schemes. However, the scale of Africa supports appropriation technologies and methods. support to rain-fed areas was hundreds of millions of Each method has different implications for limited, even though such smallholder cultivators food production, food security at local and systems supported more than who depend on land for national levels, and the distribution of costs 80% of farmers in the developing subsistence, livelihood and benefits. countries of Asia and Africa. As and food security. a consequence, there has been a The growth of modern ‘conventional’ widening income gap between irrigation since 1900 has been characterised irrigated and rain-fed areas. Even within by large water projects that harnessed rivers large-scale irrigation systems, inequities of through the construction of diversion this nature are observed, leading to the structures and canal systems. Since 1950, marginalisation of smallholders. the spread of such technology accelerated through state-sponsored large-scale irriga- The regional economic and development tion and an emphasis on large dams for context for agriculture differs markedly for water storage. Irrigated areas increased from industrial and developing countries. In the 40 million hectares in 1900 to 100 million former, agriculture tends to be capital- hectares by 1950 and to 271 million by intensive with large, highly mechanised 1998.2 Dams support 30-40% of this area, holdings requiring minimal labour. In

The Report of the World Commission on Dams 137 Chapter 5

contrast, agriculture in Asia and Africa A final option is to import food from other supports hundreds of millions of smallholder countries rather than trying to achieve cultivators who depend on land for subsist- either a higher degree of food self-sufficien- ence, livelihood and food security. These cy or security through domestic production. farmers generally do not have access to This may be possible in countries with a support mechanisms or capital resources to small farming population, for individual risk growing high-value crops in volatile crops with a high water demand, or for market conditions. The low productivity of countries with significant foreign exchange the land and labour of many earnings. However, it would be extremely subsistence cultivators is also counterproductive in countries with a large and poor rural population if it interfered There is considerable symptomatic of absence of with the income-earning potential of small scope for enhancing the support and widespread neglect farmers and their incentives to produce. viability, adoption and of their agriculture and irrigation performance of other systems. sources of irrigation Improving performance and water such as This section presents a brief productivity of existing groundwater, direct overview of some of the options irrigation systems available for agricultural devel- river abstraction and As noted in Chapter 2, there is considerable opment, with an emphasis on traditional water under-performance of large dam irrigation those most likely to be consid- harvesting systems. schemes and scope for improving the ered as alternatives to irrigation, performance of existing systems. Increasing particularly irrigation supplied by competition for water has highlighted the large dams. The presentation is grouped inefficiencies in irrigated agriculture and according to three levels of options: thus increased demands for a more effective ■ improving performance and productivity and integrated approach to managing of existing irrigation systems through existing irrigation systems, particularly improved basin and system-level man- surface water irrigation. There is also agement, on-farm technological options considerable scope for enhancing the to enhance the productivity of land and viability, adoption and performance of other water, and policy and institutional sources of irrigation water such as groundwa- reforms to improve incentives for water ter, direct river abstraction and traditional efficiency and demand management; water harvesting systems. In the past, ■ improving the productivity and liveli- emphasis on performance enhancement hood opportunities offered by alternative programmes has had mixed results, however, supply-side measures through enhancing with rapid decline in gains achieved leading to rain-fed agriculture, supporting local and the need for periodic restoration, often heavily traditional water appropriation tech- subsidised by the state. Policy interventions niques and adopting new technological and institutional reforms to support technical options such as water recycling; and interventions have been inadequate. ■ investing in conventional supply-side Improved basin and system level measures to develop new irrigation areas management based on direct abstraction from rivers As reviewed in Chapter 2, the risk and and groundwater. consequence of sedimentation of reservoirs

138 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

is site-specific, but of considerable impor- tivity of existing land, make Salinity affects tance in a portion of the larger dam popula- better use of irrigation, and approximately 20% of tion. Measures to improve the sustainability demonstrate that new irrigation irrigated land worldwide. of existing water resources systems through areas can be managed in a Controlling salinity and sediment flushing and catchment manage- sustainable manner. Application reclaiming saline land is an ment can increase the contribution and of drainage technology and urgent priority in order to longevity of irrigation systems. Enhancing maintenance of existing drain- increase productivity of infiltration and reducing surface erosion age is one way of containing existing land, make better through catchment-protection initiatives salinity. But reliance on physical use of irrigation, and may improve sustainability of reservoirs and drainage of saline effluent alone demonstrate that new irrigation systems, but must account for the is insufficient to tackle the irrigation areas can be trade-off with subsequent losses in annual problem, and an integrated managed in a sustainable water yield and the potential for lower dry approach combining manage- manner. season flows. A review of 94 catchment ment of surface water, groundwa- experiments from around the world suggests ter and agricultural practices is essential. a loss in water yield of 10, 25 and 40 mm for Salt-tolerant crops and vegetation can form a 10% increase in catchment cover respec- part of such strategies to remove excess tively for scrub, deciduous hardwood and surface water and lower water tables. The pine and eucalyptus.4 irrigation of crops or pastures in ‘series’ arranged in order of increasing salt tolerance Further, as explained earlier in Chapter 2, is a further example of saline management. increases in vegetation lead to greater If practised early enough, the integrated (or evaporation without necessarily leading to conjunctive) management of surface water, greater absorption and a reduction in surface groundwater and salinity can prevent the run-off. Thus, the balance between evapora- build-up of salts (see Box 5.1). Trials on this tion and gains from increasing absorption are currently under way in Australia and will determine whether catchment measures California. lead to increases or decreases in dry season flow.5 The utility of these options will Water quality is another important factor therefore be site-specific and depend on the affecting basin productivity. For example, techniques applied and must be developed efforts to improve salinity problems up- and evaluated in the larger context of stream may have adverse effects down- natural resources management in the catchment. In particular the contrast between the effects on soil and water Box 5.1 Conjunctive management of salinity conservation of vegetative cover and structural measures such as embankments, Conjunctive management, that is groundwater pumping with reuse of the saline effluent for irrigation, is an economic and sustainable means of salinity control. ditches and small dams needs to be consid- In the Shepparton Irrigation Region in Australia, one project covering 600 ha ered. with 15 groundwater pumps has been in operation since the early 1980s. The salinity of the diluted groundwater is kept below threshold level for the crops and the management system prevents salinisation of the root zone. Long-term Salinity affects approximately 20% of sustainability depends on the ability to obtain a salt balance in the area protected by the drainage pumps. 6 irrigated land worldwide. Controlling Source: Heuperman, 1999, Contributing Paper for salinity and reclaiming saline land is an WCD Thematic Review IV.2 Irrigation Options urgent priority in order to increase produc-

The Report of the World Commission on Dams 139 Chapter 5

stream. An innovative improvements in water savings and reliabili- approach to the problem of ty of supply. Studies from Pakistan in the managing the disposal of early 1990s demonstrated that a focus on saline effluent comes from targeted maintenance is more cost-effective the Murray Darling Basin of than deferring problems to rehabilitation Australia where disposal is programmes or lining canals.10 regulated through regional load quotas or ‘salt credits.’ Inadequate maintenance is a feature of a These credits are used to number of irrigation systems in developing ensure highly saline water is released only countries. An impact evaluation of 21 during periods of high flow, when disposal irrigation projects by the World Bank has the least impact on river water quality. concluded that a common source of poor performance was premature deterioration of The review of irrigation system performance water control structures.11 Often poor in Chapter 2 revealed a considerable lag maintenance reduces irrigation potential time between commissioning of irrigation and affects the performance of systems. dams and full development of irrigation infrastructure. Accelerating the full devel- The scope for improvements through better opment of associated infrastructure, such as maintenance is significant. The Food and canal networks, and providing an integrated Agriculture Organisation (FAO) of the package of agricultural support measures for United Nations estimates that the total area new irrigators may be a cost-effective option that can be used with improved mainte- for enhancing performance. nance is 150 million hectares.12 In India, such steps could add approximately 9 mil- In most irrigation systems, particularly those lion hectares to the irrigated area.13 In with long conveyance lengths, a dispropor- countries like Nigeria, only 52% of irriga- tionate amount of water is lost as seepage in tion from large-scale schemes was actually canals and never reaches the farmlands. For used in 1993.14 In the combined Gezira- example, 40% of the water diverted from Managil schemes in Sudan, 126 000 ha had the Indus basin in Pakistan is lost in con- to be taken out of production due to sedi- veyance. In the late 1980s it was estimated mentation and weed growth in canals.15 that improvements in supply efficiency Performance-based maintenance contracts could save some 14.8 billion m3/yr of water.7 can be used as a cost-effective management Canal lining is one such improvement. measure to maintain system integrity. Such Experience with buried plastic linings in an arrangement can be extended to cover China demonstrates good performance over operation of the main canal system in 18 years.8 The United States introduced a tandem with decentralised farmer manage- low-cost lining programme and since 1946 ment at the secondary and tertiary level. some 4 600 km of various types of linings have been installed.9 Proposals to improve system efficiency should be viewed in a river basin context But in the absence of good quality control examining the interactive effects of surface and effective maintenance the canal linings water, drainage re-use and groundwater often have not achieved the predicted recharge.16

140 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

On-farm technologies for enhancing among smallholders in developing countries, the productivity of land and water but local manufacture has brought down the Raising the efficiency of surface irrigation cost and improved viability for high-value provides considerable scope for improve- crops. Further development in this area may ments within existing irrigation systems, make such pressurised irrigation whether the water comes from dams, rivers, technology attractive to farmers or groundwater. There are large variations in in developing countries. Low Micro-irrigation methods, efficiency. As cited in Chapter 2 the average technology versions, such as such as sprinkler and drip water use efficiency that was reported for making portable drip line and systems, provide an the Aslantas project by the WCD Case bucket kits for small household opportunity to obtain Study was 40%. However, the range of field- irrigation, have recently gained higher efficiency gains level efficiency for the project varied from support.19 Another method for than those available in 25% to 55%. Efficient surface irrigation, increasing water use efficiency is surface irrigation. Field however, has produced figures of 60%.17 to change crop cultivation application efficiencies practices (see Box 5.2). are typically in the range In a number of regions, irrigation has of 70–90%. promoted cultivation of water-intensive As with other water manage- crops such as sugarcane and rice. For exam- ment initiatives, the potential ple, intensive cultivation of sugarcane in for efficiency gains in water use needs to be India and Pakistan has led to unsustainable examined within a river basin context, as in use of water. A number of technologies exist many cases losses upstream are recouped for improving water use efficiency and, downstream. hence, the productivity of water in irriga- Policy and institutional reform tion systems. Policy and management initiatives are Micro-irrigation methods, such as sprinkler fundamental to raising productivity per unit and drip systems, provide an opportunity to of land and water and increasing returns to obtain higher efficiency gains than those labour. They are often interlinked and available in surface irrigation. Field applica- require political commitment and institution efficiencies are typically in the range of tional co-ordination. 70-90%.18 The output produced with a given amount of water is increased by Agricultural support programmes tend to be allowing for more frequent and smaller developed and implemented in relative irrigation inputs, improved uniformity of isolation from irrigation systems. Typically watering and reduced water losses. These there is weak co-ordination between agen- methods have found wide usage in water- cies responsible for agricultural activities scarce regions. Nearly all of Israel’s irrigated area, 68% of Jordan’s and 40% of Brazil’s irrigated area uses micro-irrigation methods, Box 5.2 Cultivation techniques can reduce irrigation water use as do small areas in China, India and parts The Muda Irrigation Scheme in Malaysia reported a reduction in irrigation of Africa. duration from 140 to 105 days and a reduction in overall water use of 28% as a result of the shift from transplanted rice to wet seeded rice. Although the impetus for this change was shortage of farm labour, it had a considerable water The capital-intensive nature of micro- conservation benefit. irrigation technology and low cost of water Source: Fuji and Cho, 1996, cited in Guerra et al, 1998 has prevented more widespread adoption

The Report of the World Commission on Dams 141 Chapter 5

(such as extension services, land ciency and promote a shift to less water- consolidation, credit and mar- demanding crops, especially in arid and keting) and those responsible for water-scarce regions. Ideally, pricing struc- irrigation development. Price tures for irrigation should reflect the cost of incentives are also inadequate to supplying water and associated externalities, raise productivity and the and should be designed with stepped rates to outcome is a significant gap provide security for basic livelihood needs. between potential and actual yields. In the absence of better In many systems farmers are charged on a opportunities from agriculture, per hectare basis independent of the volume many farmers seek off-farm employment. of water used. This removes any incentive to Incentives to enhance production are save water. One obstacle to volumetric necessary and can result from a more inte- charges is the practical difficulty of measur- grated set of agricultural support measures ing water delivered through an open canal and the involvement of joint ventures that system to a large number of smallholders. A provide capital resources and market access solution may lie in water user organisations to smallholder farmers. Appropriate arrange- acting as intermediary bodies. They can ments need to be introduced for such joint enter into contracts with irrigation agencies ventures to ensure an equitable share of and recover the charges directly from the benefits. irrigators.

As demonstrated in Chapter 2 the extent of One of the major contributors to poor recovery of the costs of operations and performance of large irrigation systems is the maintenance in large irrigation dams in the centralised and bureaucratic nature of WCD Knowledge Base and irrigation system management, characterised by low systems in general is often limited. In a levels of accountability and lack of active recent study of 16 projects, annual irrigation user participation. Agency reform and fees varied from zero in Thailand to as high management transfer have been initiated in as $130/ha in Colombia.20 Collection rates more than 25 countries where governments varied from 50 to 100%. One rationale for are gradually reducing their roles in irriga- higher fees is that they will encourage more tion management and transferring responsi- efficient irrigation practice and a shift to bility for various levels of the systems to more water-efficient and higher-value crops. farmers’ organisations and water user associations. The major impetus for the transfer For example, in the case of Pakistan and lies in the desire to cut back public expendi- India reported above, farmers are attracted ture on operation and maintenance costs.21 to the intensive cultivation of sugarcane by its profitability. This is because water The structure of farmer involvement varies charges are low, capital costs are not recov- from transfer of assets to a range of joint- ered, and the mitigation costs of attending management models. As yet, there is no to waterlogging or salinity problems are not general evidence to suggest that irrigation borne by the farmer. Removing the subsidies performance has improved as a result of inherent with supplying irrigation and transfer alone, although there are promising drainage services may encourage adoption of examples indicating that decentralisation technologies for increasing water use effi- may be a required, but not sufficient meas-

142 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

ure to improve performance.22 Experience techniques and irrigation methods were used has shown that in order to be effective, a in different contexts. Several supply options strong policy framework is required, provid- existed, complementing each other and ing clear powers and responsibilities for the contributing to agriculture, food production farmers’ organisations.23 and livelihoods. In today’s context, local solutions and large-scale irrigation need not Water rights and trading are highly conten- be mutually exclusive. They can co-exist tious issues. Win-win situations occur for and complement each other, raising food farmers when they trade a part of their water production and enhancing livelihoods. to replace lost income while at the same time being able to finance water use effi- For these systems to function optimally, a ciency gains from their remaining water number of enabling conditions are required. allocation. In the United States, Colorado Water appropriation systems need repair and has one of the most advanced institutional maintenance, desilting and weed clearing. support networks for water markets. In Innovations and enhancement of traditional recent years some 30% of a district’s annual methods are needed to improve crop pro- water entitlements has moved through the ductivity. To optimise the productivity of rental market. The price at which farmers the restored water management systems, sell water is often significantly higher than appropriate land use, including cropping their cost of supply. patterns, mix and rotation need to be promoted. Sustaining these location-specific Enabling conditions for water markets are systems and practices will depend on protec- clear and secure entitlements along with tion of sensitive catchments, floodplains effective administrative systems and infra- and deltas. Lastly, improving community structure to regulate the trade and to moni- stakes will be an important factor for greater tor compliance. There should be clear application and long-term functioning. environmental limits on the extent of the Enhancing rain-fed agriculture and trade. These prerequisites are not present in supporting local techniques many developing countries. Chile is often Some 80% of agricultural land worldwide is cited as an example where the development under rain-fed cultivation, contributing to of water markets has helped to avoid expen- 60% of food production.25 Given the sive new water infrastructure, such as dams, number of low-income households that rely by allowing transfers of water rights from on rain-fed agriculture throughout the agricultural to urban sectors.24 Concerns developing world, the enhancement of over water trading stem from the equity and opportunities in this sector can have a major livelihood implications of permanent trades effect on productivity and livelihoods. of water (as opposed to yearly rentals) from small to large farmers or from rural to urban Over a period of time these farming practic- areas. es and irrigation methods have been margin- Improving alternative supply- alised by irrigation policies, the lack of institutional support and low levels of side measures investment and research. In recent years, Prior to the advent of large-scale irrigation, increasing attention has been paid to the a number of traditional water appropriation successes of such methods and their impor-

The Report of the World Commission on Dams 143 Chapter 5

Box 5.3 A local approach to integrated water management, the technological and capital base of the Rajasthan, India farmer.27

Water-harvesting structures to store rainwater runoff were constructed in the Agriculture in rainfed areas is supported by a Bhagani Tildeh river catchment in Rajasthan in response to water scarcity. Tracts of land inundated by the construction of embankments and dug retention number of water appropriation methods. ponds are referred to as johads. The production of crops and milk have increased many-fold in the Alwar district of Rajasthan due to the greater Many of these have been used for centuries availability of water through the construction of 2 500 johads in 500 villages. The and have been adapted to satisfy local total expenditure incurred was in the order of $3.6 million (in 1998 prices), with 28 73% contributed by local people. The johads have recharged the groundwater needs. Rainwater harvesting uses small table resulting in an average rise from 200 feet below ground level to 20 feet. dams and embankments to trap run-off, Sources: Thakkar, 1999, Contributing Paper for usually with the aim of recharging ground- WCD Thematic Review IV.2 Irrigation Options water. Surface sources such as springs and streams are also diverted into natural depressions using bunds and check dams. In tance is now being recognised by policy- floodplains, cultivation uses natural mois- makers.26 The productivity of these rain-fed ture brought in by annual floods. A variety systems can be enhanced through improved of water management techniques are used to agricultural support, local commitment to tap the soil water in wetlands and valley develop reliable water supply sources and bottoms. Many of these localised practises small-scale, low-cost technologies. A serve multiple purposes, such as checking progressive development path, where saline ingress in coastal areas and recharging support provided is tailored to increasing groundwater in floodplains. Effectiveness, local capacity, is emerging as an important sustainability and productivity are location- conceptual approach to the development of specific and depend on the extent to which such systems. For example, simple technolo- they are spatially integrated. One successful gies such as treadle pump and low-cost drip example comes from the Rajasthan area of systems can be introduced as a first stage India (see Box 5.3). innovation. In a few years – once farmers have recovered their investment costs – the In many regions rainwater is harvested returns can be re-invested in more advanced where it falls. In Argentina, Brazil, and technologies such as small motorised Paraguay, natural or artificial depressions are groundwater pumps, thus gradually building used to store rainwater for crop and livestock production. In the semi-arid areas of Argentina, Brazil and Venezuela rainwater Box 5.4 Rainwater harvesting for domestic and agricultural use in run off captured from roads is collected in China drainage ditches or street gutters and Gansu Province lies on the loess plateau in central China and is one of the driest transported to cultivation areas.29 Rainwater and poorest areas of the country, with annual per capita incomes of around $70– 80 in rural areas. In dry years, traditional rainwater collection could not always harvesting initiatives in Gansu province in provide sufficient water, and people were forced to trek long distances to rivers China have provided both domestic and or to depend on government water trucks. In response to the 1995 drought, the 1-2-1 project was launched: it provided 1 clay tiled roof catchment area, 2 irrigation water to areas previously in upgraded cement water cellars, and plastic sheeting for concentrating rainwater drought (see Box 5.4). runoff on 1 field. The project ensured that some 1 million people not only had sufficient water but also could grow their own food and earn limited income through cash crops. The total implementation cost for the project came to $12 There are hundreds of thousands of small per capita. reservoirs or water tanks around the world, Source: Gould, 1999, Contributing Paper for WCD Thematic Review IV.3 Water Supply ranging from farm dams in Africa used to provide intermediate storage, to recession

144 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

reservoirs in Cambodia that are used to As noted earlier, there is a need enhance soil moisture for cultivation. In to adequately account for the India and Sri Lanka, more than 500 000 effects of structural alternatives tanks store rain water, sometimes supple- such as dams and levees on mented by water from streams or small these natural irrigation ‘options’. rivers.30 Tank systems have provided irriga- It is also necessary to consider tion and supported agriculture and liveli- the option of managed flood hood for centuries in South Asia, although releases from existing facilities reliability problems, siltation, poor mainte- as a means of re-establishing and nance and degradation have reduced the supporting these productive area served by small tanks in India from 4.6 systems. million hectares in 1960 to 3.3 million Adopting water recycling hectares by 1987-88.31 In addition to helping irrigation, tanks can play several Reuse of irrigation drainage water can be a important roles, such as flood control, soil major supply source in areas where intensive erosion control, conservation of runoff surface irrigation is currently practised. during heavy rainfall and recharge of Without formal arrangement, farmers in the groundwater. Eastern Nile Delta in Egypt use approximately 3 billion cubic meters of saline The implications of the widespread replica- drainage water every year for irrigation after tion of these small-scale storage methods on suitable dilution or treatment. Similar reducing demand for irrigation water and examples exist in the North China plains, enhancing the sustainability of groundwater Arkansas valley in Colorado, the Pecos 34 supplies, as well as their impact on existing valley in New Mexico and Australia. surface water sources, could be far reaching. Drainage water reuse requires increased As described in Box 5.5 and Chapters 3 and levels of management skills to deal with 4, the floodplains and deltas of a number of increased levels of salinity, toxic concentra- the world’s major rivers support wetland tions and associated health concerns. ecosystems of exceptional productivity that in turn support large rural communities. Box 5.5 Wetland and floodplain agriculture Floods provide natural irrigation that 32 fertilises floodplain soils. As the floodwa- Cultivated wetlands and valley bottoms, with some form of traditional water ters recede, arable crops are grown. Some management constitute 12 200 km2 in Central Africa, East Africa and the Gulf of Guinea. Flood recession agriculture is practised over an estimated 10 400 km2 in soil moisture persists in the dry season and the Gulf of Guinea and the Sudano-Sahelian region. In Senegal, the floodplain provides grazing for migrant herds. Flood- runs 600 km downstream of Bakel and covers an area of 10 000 km2. In good years half of it will be inundated, supporting close to a million people who plain cultivation is among the most produc- depend on cultivating the floodplain. In southern African countries seasonally tive and widely practised of agricultural waterlogged lands called dambos, augmented by small human powered devices like shallow treadle pumps, support crop production and smallholder systems in central and Sub-Saharan Africa. food security. In countries like Zambia, dambos cover 38 000 km2. Similarly, Similarly, wetland and delta cultivation is wetland and delta cultivation is widely practised in East Asia. In Vietnam, the Mekong delta constitutes the principal rice producing area of the country. Rice widely practised in Southeast Asia. A form of is also grown extensively in the wetlands of Cambodia and in smaller quantities flood agriculture specific to arid environments in Laos and Thailand. is spate irrigation, which diverts seasonal flood Sources: Gulf of Guinea and Sudano-Sahel in FAO, 1995, p10; Senegal in FAO, 1997, p136; Zambia in Postel 1998, p213; flows from dry riverbeds. Large areas under Southeast Asia in Friedrich, 2000, spate irrigation are found, for example in WCD Regional Consultation Paper Morocco, Yemen and Pakistan.33

The Report of the World Commission on Dams 145 Chapter 5

Reuse of urban wastewater is a significant irrigated agriculture. The availability of source for irrigation in a number of coun- electricity, centrifugal pumps and well- tries. In Israel, 275 million m3 of wastewater drilling technology gave a major boost to are used for irrigation after treatment. This rapid growth, particularly by individual is approximately 22% of the total agricultur- farmers. Countries such as China, the al use of water.35 Strict controls are needed United States, India, Pakistan, Bangladesh, on the level of treatment required for Saudi Arabia and the North Africa region various classes of water, with more stringent have high rates of groundwater use. Cur- requirements for irrigating food grains than rently in China, 8.8 million hectares of land fodder crops. Examples from Ghana and are irrigated through groundwater wells, Kenya also show the potential for peri-urban constituting 18% of the total irrigated irrigation based on water reuse.36 Where area.38 The Ogallala aquifer in the United regulation of water quality is weak, this States waters a fifth of that nation’s irrigated practice raises significant health concerns. land.39 By the late 1990s, groundwater irrigated over half of all irrigated land in India, Investing in conventional contributing to 78% of additional irrigated supply-side measures area created between 1984 and 1994.40 The potential to expand irrigation into new areas has sharply declined due to increasing- Groundwater – employed on its own or in ly constrained resources and significant conjunction with surface irrigation – is increases in the unit development cost. often more productive than surface irriga- Conventional sources of water for irrigation tion per unit applied. The determining besides reservoir storage behind large dams factor appears to be the higher degree of include diversion from rivers and lakes and control available to farmers who are often groundwater abstraction. prepared to pay considerably more for reliable sources of supply such as groundwa- Diversion canals or lift irrigation pumping ter.41 Improved management of surface schemes supply irrigation systems by ab- irrigation systems, leading to greater reliabil- stracting water from rivers. For example, ity of supply, may similarly increase produc- river diversions and pumped irrigation serve tion and returns to water. 80% of the irrigated area in Kenya and 68% in Nigeria.37 The lack of over-season storage Continued withdrawal of groundwater at implies that the capacity of the system to current levels is, however, becoming unsus- provide multiple crops depends on the tainable in many places. The Ogallala reliability of river flow. The size of run-of- aquifer, for instance, is being depleted at the river schemes can vary from a few hectares rate of 12 billion m3 annually. Falling water to hundreds of thousands. Run-of-river tables, increased pumping costs and histori- diversions can be used to supplement cally low prices have led to a reduction in storage-based systems. In Sri Lanka, for the area irrigated by the Ogallala of 20% 42 example, diversion weirs were built to over a 10-year period. Groundwater capture drainage water from upstream dam- depletion has been a growing concern in the based irrigation projects for reuse. North China plains for over three decades, with water levels falling by 30 metres since Groundwater abstraction has played an the 1960s.43 Efficiency improvement meas- important role in the global expansion of ures are as important to groundwater systems

146 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

as to surface water irrigation. Effective options that would respond to Sustainable use of regulation is also necessary to curb over- different development needs, groundwater can be pumping, which leads to lowering water sustain a viable agricultural achieved through tables and can limit access of poorer farmers. sector, provide irrigation and controlled abstraction offer livelihood opportunities and associated recharge Sustainable use of groundwater can be to large populations. First, measures. Recharge can achieved through controlled abstraction and policy and institutional be achieved by associated recharge measures. Recharge can support for innovation, spreading surface water be achieved by spreading surface water over modernisation, adaptation, over large areas, using large areas, using recharge wells, and water maintenance and extension of recharge wells, and harvesting techniques as described above. traditional irrigation and water harvesting Floodplains perform a significant natural agricultural systems was techniques. recharge function. As floodwaters soaks lacking in the past. Increasing through, underground reservoirs are re- recognition of this has led a charged, and these supply water to wells number of actors to place priority on im- beyond the floodplain. Recharge can also be proving rain-fed agriculture and developing a suitable approach for controlling saline small-scale irrigation capacity. An extension intrusion and land subsidence and for of this priority is the need to protect (or reducing pumping costs. Most of the artifi- restore) the natural functioning of deltas, cial recharge systems used to date have floodplains and catchments in order to focused on small-scale systems or municipal sustain and enhance the productivity of water supply uses. traditional systems in these areas.

Currently, almost half of the large dams in Second, the institutional framework needs the world provide irrigation services. The to be redefined by transferring management spread and contribution of dams to irriga- to decentralised bodies, local governments tion and food production and the environ- and community groups (water users associa- mental and social implications of their use tions or other appropriate bodies) for were described in previous chapters. As recovering tariffs and maintenance. Strong- discussed in Chapter 4, irrigation projects er commitment is required to transform can have significant multiplier effects on irrigation bureaucracies into more efficient, the local economy in terms of contributing service-oriented organisations capable of to the development of agricultural process- managing water and land in an integrated and ing and related industries. The WCD sustainable manner. Irrigation water needs to Knowledge Base does not elaborate on the be appropriately priced so that charges are comparative effects of the different options based on volume used, taking into account the in terms of engendering such multiplier need to support basic needs and serve both effects, although this will be an important equity and conservation. Third, emphasis must consideration for options assessment. be given to developing a package of agricultural support measures that are mutually reinforc- Obstacles and enabling ing and develop intersectoral linkages in the conditions local economy so as to spur rural development. A number of policy, institutional, and These efforts also need to counteract the regulatory factors hinder the emergence and tendency of current policies and extension widespread use of an appropriate mix of services to impose high transaction costs and

The Report of the World Commission on Dams 147 Chapter 5

risks on smallholder farmers in options to food production and livelihood The priority for a developing countries. Measures to security. For this purpose, analysis that sustainable and enhance security of tenure are extends beyond the immediate costs and equitable global energy also required. benefits of these options to the secondary sector is for all societies economic impacts on specific social groups to increase the efficiency Farmers also need access to is necessary. of energy use and the international markets through use of renewable reduction in barriers and sup- sources. High- Energy and Electricity portive domestic policies. In consumption societies response to structural adjust- The range and scale of energy resources and must also reduce their ment programmes and interna- technologies for electricity demand-side use of fossil fuels. tional agreements on world management and supply have expanded trade in agriculture, many dramatically in the last quarter-century due developing countries have liberalised their to advances in individual technologies and agricultural policies, including cutting tariffs greater success in adapting existing and new and subsidies. However, tariff and non-tariff technologies to local settings. While coun- barriers to OECD markets – such as the tries have different energy resource endow- large production and export subsidies for ments, there is no supply global crisis on the farmers in the US and European Union – horizon as is anticipated for fresh water. The limit the ability of developing countries’ to world’s renewable sources and fossil fuels are diversify their agricultural sectors and gain sufficient to meet foreseeable global de- the benefits of increased international trade, as mands for electricity generation over the well as increasing rural poverty where local next 50-100 years using existing or near- farmers cannot compete with cheap imports.44 term technologies.45 Moreover, the range of energy systems and technologies that may be To sum up, future assessment of alternatives called upon to convert primary energy sources will need to clearly consider the following: into electricity has dramatically expanded in ■ improvements to the efficiency and the last few decades. The priority for a sustain- productivity of existing irrigation systems able and equitable global energy sector is for before planning and implementing new all societies to increase the efficiency of ones; energy use and the use of renewable sources. ■ adaptation and expansion of local and High-consumption societies must also traditional water management solutions; reduce their use of fossil fuels. ■ more co-ordinated management of Key factors in the expansion of options surface and groundwater resources; and include the improved capacity of developing ■ improvement of the productivity of rain- countries in design and manufacturing, fed agriculture. growing experience in adapting new technologies to rural and decentralised settings, To make progress in this area will require and enhanced cost-competitiveness of the concerted efforts in policy formulation and new technologies due to volume production institutional reform. – ranging from wind-turbines to compact As an aid to this assessment it will be useful fluorescent bulbs. These have given devel- to more fully understand the contribution of oping countries the opportunity to ‘leapfrog’ alternative irrigation and agricultural over older options when extending services

148 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

to rural and urban areas. Innovation and The world’s demand for electric- change were spurred by the oil price shocks ity has doubled over the past 22 in the 1970s and 1980s, which prompted years.47 People are using more major academic, government and industry electrical services in the tech- programmes to develop alternatives. As oil nology-driven digital economy, prices in the late 1980s and 1990s fell back and are using electricity more in line with historic real prices and the widely in post-industrial, perceived shortages and security threat transitional and developing receded, alternatives were set aside. Recog- economies. Yet the supply and nition of the causes and scale of the threat use of electricity is highly of global climate change in the 1990s have skewed between industrial and developing refocused work on alternatives, and galva- countries and between the rich and the poor nised the thinking on sustainable develop- in developing nations. There are enormous ment, including the role of the power sector. opportunities for demand-side management A tripling of oil prices in early 2000 pro- in industrial economies. Government efforts vides a further reminder of the need for to reduce greenhouse gas emissions and make continued long-term research and develop- a contribution to climate stabilisation targets ment of alternatives to fossil fuel technolo- are indications of moves in that direction. gies. There is also considerable scope for efficien- Some observers suggest that the world has cy improvements in developing entered a period where the revolution in economies, where they would Rural populations are electricity technology – coupled with the moderate the required investment often low-income and revolution in digital technology – will pave in new supply. In addition, almost live in settings where the way for a profound transformation in the 2 billion people, both urban and centralised energy delivery of and access to electricity services rural poor, have no access to services are expensive, early in the 21st century.46 It is certainly electricity at all.48 Rural popula- both in national or broadly accepted that the long-term trend is tions are often low-income and consumer terms. towards a global energy system that is less live in settings where centralised Decentralised, off-grid carbon-intensive and less reliant on finite energy services are expensive, systems are an energy resources. Nonetheless, there is both in national or consumer important option in considerable debate on the means and terms. Decentralised, off-grid these areas. timing of the transition, its shape in differ- systems are an important option ent regions of the world, and its contribu- in these areas. tion to addressing the larger global equity issues of disproportionate resource use. Figure 5.1 provides a simple schematic There is also considerable inertia and representation of the electricity sector today, resistance to change in the existing system, showing generation, transmission, distribu- but the direction implied is clear: a shift tion and end-use components. As shown, towards ‘cleaner fossil fuels;’ a significant, there are three general ways to improve the accelerated shift towards the use of renewa- delivery of electricity services: ble energy sources for electricity generation ■ demand-side management options, and a focus on improving efficiency in the concerned with efficiency on the user delivery and use of electricity services. side of the electricity meter;

The Report of the World Commission on Dams 149 Chapter 5

■ supply-side efficiency measures, con- nity to reduce the need for electrical genera- cerned with how efficiently electricity is tion and consequently the need for dams. generated by the centralised or local The discussion also has broader dimensions. supplier and transmitted and distributed Demand-side management is about consum- to users; and ers using less electricity and using it more ■ new supply options, which replace efficiently in the residential, industrial, existing generation options or supply commercial or government sectors. The incremental growth in demand beyond major entry point for these improvements is what can be achieved by options in the the replacement of energy inefficient first two categories. appliances. Enabling conditions that affect DSM uptake of improved appliances include New supply options may be further divided the replacement cycle (whether it is a few or into grid and off-grid options. The latter many years), standards, comparative cost includes small isolated ‘mini’ grids and and availability, consumer awareness and stand-alone supply to individual customers affordability. Generally investments in DSM and homes. These options are outlined that promote consumers’ use of efficient below beginning with demand-side manage- appliances will be more than offset by the ment and then passing on to supply-side avoidance of investments in new supply and options (efficiency and new supply) and finally environmental and social costs of generation. covering options for rural electrification. Recent investigations as part of the UNDP’s World Energy Assessment, to be finalised in Demand-side management 2001, indicate significant potential for In the context of this report, demand-side electricity efficiency improvements in all management (DSM) represents an opportu- countries.49 The technical potential in

Figure 5.1 Schematic of Electricity Options

Electricity

Demand Supply

Tariff structure Load management Interruptable loads End-use efficiency Transmission and Main grid Isolated network Home systems Distribution Power plants Power plants in • solar home Loss reduction • thermal (coal, mini-grid systems and improvement oil, gas) • diesel • micro turbines System reliability • nuclear • small hydro • fuel cells • transmission • hydro • biomass • distribution (storage, • wind run-of-river, • cogeneration pumped storage) • wind • solar • biomass • co-generation

150 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

countries with a high per capita consump- ■ The structure of the international energy tion, such as the United States, may be up industry remains biased against DSM. to 50%, including modification of consumption and improved conservation behaviour. The latter bias stems in part from the Others see less potential. Recent successful diversity of producers in the industry which programmes in Asia and elsewhere also inhibits the development of a lobby struc- illustrate that there is great potential for ture similar to that of the energy supply DSM, particularly in the modern sectors of industry which is dominated by some 50 developing countries.50 Most efficiency global players.51 measures and technologies are cost-effective at today’s electricity prices, and the use of full The fundamental enabling condition to environmental and social costing of electricity improve the prospects for DSM is that supply options makes them even more so. governments lead the way and clearly define the support roles Despite the promise, actual global invest- for utilities. Effective policies Most efficiency ments in energy efficiency and the savings that can reduce transaction costs measures and from them continue to be small compared include continuation and expan- technologies are cost- with the potential. And while utility-govern- sion of information and product effective at today’s ment partnerships in DSM grew strongly in promotion, training programmes electricity prices and the 1980s in North America and Europe, the on energy efficiency service for the use of full move to open markets has led to lower suppliers and consumers, mini- environmental and electricity prices and a perverse disincentive mum efficiency standards, social costing of for DSM in terms of tariffs. At the same labelling and endorsement of electricity supply time, many power utilities have dramatically high-efficiency equipment and options makes them cut their DSM programme budgets. selective government procure- even more so. ment. Other steps will require Several market and institutional barriers innovative financing and con- continue to obstruct the accelerated adop- tracting mechanisms, removal of subsidies tion of efficient end-use appliances and for conventional supply options and crea- realisation of the potential of DSM: tion of power-sector regulatory frameworks ■ Utilities closest to the consumer often view and market incentives that encourage DSM. efficiency in terms of loss of market share, Ultimately, building capacity to design and and in trying to meet a growing demand deliver efficiency policies and programmes tend to think at first of new supply options. in government agencies and utilities is ■ Subsidies are still given for energy supply required. and consumption (especially for energy- intensive consumers), and there is a lack Supply-side options of credible commitment to energy Before electricity reaches the consumer, efficiency from governments and inter- there are two broad ways to improve electri- national agencies. cal services: investment in supply-side ■ Governments find it easier and more efficiency and new generation. The latter politically attractive to approve new option involves two further issues – substi- supply options than to get consumers to tuting current generation with new sources use energy more wisely. (for example replacing fossil fuels with wind

The Report of the World Commission on Dams 151 Chapter 5

power), and meeting increments of demand systems and direct current transmission growth in economies that are expanding. technology, including back-to-back AC/DC converters, are paving the way for operation Supply-side efficiency and grid connection of a diverse array of From the point where the electricity is options on electrical systems, particularly generated to its end-use destination, numer- renewables. These technologies enable ous technical and non-technical losses may connection of small generators with inter- occur. The technical losses include con- mittent power (for example wind) to the sumption at the power station, step-up grid and allow interconnection of grids at transformer losses, and transmission and different voltage levels. distribution losses. There is considerable scope to reduce conveyance losses in many A recent and significant trend in the power countries and these require urgent attention, sector is the regional interconnection of especially where programmes have not been electricity grids. This ranges from bilateral introduced. arrangements for electricity sales between two neighboring countries, to cooperative Losses between what is sent out from the power pool arrangements, to full spot- power station and what is recorded as power market and open pools involving several consumed on the customer’s meter can be as countries. Power pools and arrangements are high as 35-40% in some countries – and in place in Europe and in North America. more in exceptional cases. Making existing They are now emerging in Central America, transmission and distribution parts of South America, Asia and Africa. systems work better can lead to China is in the process of transforming into Major advances in considerable savings, and often five regional energy pools under the Elec- electronic control defer the need for new supply. In tricity Law (1997) and early stages of power systems and direct India, for example, 35% of pool formation are underway in other parts current transmission electricity generated is lost in of Asia. technology, including the transmission and distribution back-to-back AC/DC system before reaching the In parallel with the trend towards linking converters, are paving consumer. Yet it is considered grids, is the rapid expansion of regional the way for operation technically feasible to reduce pipeline networks – mainly for gas but also 52 and grid connection of a these losses to around 15%. for oil. By linking and therefore expanding diverse array of options Moreover, the improvements are power markets these two trends are likely to on electrical systems, easier to implement than DSM. increase the ability of already installed particularly renewables. Investment in improving the capacity to meet demand, increase system technical efficiency of transmis- operating reliability, and reduce vulnerabili- sion and distribution efficiency ty to drought in grids with a mix of hydro- systems may be undertaken in a few years and thermal power. These changes will and typically involve action by a single affect the future competitiveness of different agency. electric power generation technologies.

Advances in transmission technology also New generation options show particular promise for improving Excluding hydro, which represents 19% of transmission and distribution efficiency. the world’s electrical supply, all renewable Major advances in electronic control sources including biomass, wind, solar,

152 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

geothermal, ocean energy sources and co- globally) and oil-fired gas generation currently constitute around turbines for peaking and 1.4-1.6% of generation globally. standby. Nuclear power, which was the fastest growing source Every electricity-generating technology – globally between 1976 and including coal, oil, natural gas, nuclear, 1996, is faced with high invest- biomass, wind, solar, geothermal and ocean ment costs, limited public energy – has technology-specific characteris- acceptance, and unresolved tics that make it suitable or unsuitable for concerns on high-level waste different roles in the system. There are also disposal and decommission- technical limits to the amount of power that ing.53 The outlook for nuclear any intermittent source (such as hydropow- power remains highly contro- er, wind or solar) can supply to the grid versial, and countries such as before reliability of supply is affected. Sweden and Germany have Numerous technical criteria influence the plans to eventually eliminate utility preference for a particular generation nuclear generation. The use of source, especially in terms of the role of the co-generation and combined option in the system (namely, peak, base, heat and power (CHP) facilities reserve or intermittent power supply). Other is growing. CHP plants provide space and considerations include comparative cost, water heating along with electricity in cold reliability, flexibility, efficiency, availability, climate settings, with very high efficiencies. experience and familiarity with the technology and its operation. Public policy and regula- With current trends in power sector reform tion are also important factors in deciding and financing, private developers are technology choices in the power sector. showing limited interest in large hydropower projects except for uniquely competitive While conventional generation technolo- low-cost sites and medium and small high- gies are mature, each is undergoing techno- head hydropower in hilly and mountainous logical advances, with manufacturers and areas with few other energy resources. suppliers working to improve efficiency, reduce costs and construction times, and Wind power is the fastest growing of the minimise emissions and environmental renewable energy options and is competitive impacts. In the face of competition and with other conventional options when a rapid market change, power utilities and back-up generation source is available and producers favour proven low-risk options when government support (for example with short construction times, and prefera- carbon credits for offsetting GHG emis- bly off-the-shelf technologies. Gas-fired sions) is provided as an incentive. Global combined cycle systems, which combine annual growth in installed capacity of wind high efficiency and flexibility with a com- turbines averaged 40% between 1994 and paratively low initial investment cost, have 1998.54 In 2000, installed capacity stood at become the preferred choice where gas is more than 13 000 MW in 50 countries available. When it is not available, the (though this is currently small in terms of choice depends on the energy resource base. total electrical generation, equivalent to Typically the choice would be coal-fired 0.4% of installed global capacity). Europe plant for baseload (40% of generation has 70% of the grid-connected wind capaci-

The Report of the World Commission on Dams 153 Chapter 5

ty, North America about 19% and Asia on numerous factors such as competition for about 10%. About 45% of the European water and land with other users, concerns wind capacity is installed in Germany. over the loss of biodiversity to plantations, and technologies for biomass conversion to In the last two decades, the cost of wind convenient energy services. Consequently, power under good wind conditions dropped the greatest potential for biomass is seen in by 75%, bringing it within reach of avoided decentralised local systems. fuel costs of modern fossil- fuelled power plants.55 The Among advanced technologies in research Among advanced European Wind Energy Associ- and development, microturbines and fuel technologies in ation estimates that, fuelled by cells show the greatest near- and mid-term research and further cost reduction resulting promise. Fuel cells are electrochemical development, from volume production, the devices that convert hydrogen and oxygen microturbines and fuel installation of new wind directly into electricity and heat. A number cells show the greatest capacity could rise rapidly at of companies are currently investing signifi- near- and mid-term 20-30% annually so that by cant amounts in fuel cell research and promise. 2020 a total of 1.2 million MW development (R&D) and expect the com- of wind capacity could be mercialisation of the technology for use in installed world-wide, providing 10% of the vehicles and in grid and off-grid electricity world’s electricity.56 Much of the future supply by 2005. While natural gas is expect- development is expected to occur offshore. ed to be the main source of hydrogen initially, in the future hydrogen could be The cost of solar photovoltaics (PV) has produced at remote hydropower sites, wind dropped 80% in the past two decades and farms, solar stations and ocean power plants. will need to fall by a further 50-75% in order to be fully competitive with coal-fired Electrification in rural areas electricity. Photovoltaic technologies have In rural areas, electricity can be supplied reached a global production level of 120 from existing grids under traditional rural MW. Although PV production is expected electrification or from decentralised local to keep rising, this technology will not grids. Perhaps the greatest expectation for significantly contribute to grid power change is the deployment of new and requirements in the short term due to high renewable technologies for rural electrifica- cost. The long-term potential is consideration in decentralised mini grids and small ble, however. Another renewable technolo- simple devices for households. There is an gy, solar thermal systems, can already almost opportunity to leapfrog the distribution compete with conventional thermal in stage and reach some of the 2 billion people settings with high solar insolation levels. who currently have no access to centralised electricity grids. In addition to wind and solar, biomass and ocean energy systems (wave, tidal energy, At the moment, policy-makers generally and ocean thermal) have application for prefer centralised electrification options grid power. Biomass options are commercial because the technologies are already proven, where biomass fuel is readily available. But they retain the flexibility to cater for both for large-scale applications, the extent to urban and rural demand, investment credit which this potential can be used will depend (domestic and foreign) and established

154 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

credit mechanisms backed by government Some of the key elements of a Rural electrification guarantees are available, and the technolo- rural electrification policy based on decentralised gies are accepted by key institutions, such as include: approaches has been planning bodies, utilities, local government ■ financial support for commu- imposed on utility agencies and others. There are other subjec- nity institutions for owner- planners for social tive reasons in favour of centralised electrifi- ship and implementation of objectives, but there is cation including the vested interests of technologies based on local a strong preference in various groups. Although grid extension will resources or strategies; many areas for remain important, it is unlikely that central- ■ subsidisation of most aspects community-based ised approaches will continue to consume of rural electrification for decisions on the the bulk of the financial resources available meeting cost of delivery or options. for rural electrification. the inability of low-income consumers to pay for the Rural electrification based on decentralised services at cost recovery levels; and approaches has been taken up by utility ■ consideration of sustainability of any planners for social objectives, but there is a rural electrification programme to ensure strong preference in many areas for commu- its success, taking into consideration nity-based decisions on the options. Decen- economic, environmental and social tralised electrification options are attractive aspects. for a number of reasons: ■ They are simple and flexible ways to Obstacles and enabling extend energy services to remote areas conditions that have a low gestation period, may cost less, and involve low adverse envi- Numerous barriers prevent the rapid spread ronmental impacts. They include simple of the new renewable technologies and household lighting systems and mini- DSM innovations including: the direct and grids powered by diesel generating sets, hidden subsidies that have been built up small steam or gas turbines, micro- over time for conventional energy technolo- hydro units, windmills coupled to gies, a lack of commitment from governments, and the lobbying power of the generators, modified engines using conventional energy industry. biofuels coupled with generators and photovoltaic systems. Over the life cycle of an energy system, tax ■ Community-based solar programmes are and fiscal policies have an impact on the gaining momentum and can be used for competitiveness of each electricity genera- rural industry, to meet needs like crop tion option. Since many developing coun- processing or water pumping and for tries had access to concessional financing for household lighting, with a range of capital but not operating costs, hydropower financing mechanisms to recover capital often had a clear advantage over the longer- and operating expenses. term financing for other options. For emerg- ■ Rapid improvements in technology and ing renewable technologies, the cost of organisational learning have made prototypes is high, but dramatic cost reduc- systems based on renewable energy tions would ensue due to mass production if increasingly attractive and available. the technology were widely accepted.

The Report of the World Commission on Dams 155 Chapter 5

Regulatory change has had a major impact cross subsidised by higher income con- on utility practices, and power sector reform sumers, can expand access. can create enabling conditions for the ■ Governments can also ensure that power adoption of decentralised and renewable sector reform is founded on a transparent generation technologies for grid supply. The regulatory framework in which consum- reforms will also affect rural electrification. ers participate. In many countries, especially in the South, power sector reforms are extremely contro- Water Supply versial. Concerns are raised about the lack Domestic, municipal, and industrial con- of transparent and accountable regulatory sumption accounts for less than one-fifth of mechanisms that contribute to corruption as water use worldwide, and only about 5% in well as about arrangements that favour Africa, Central America and Asia. Major private power producers, impacts on tariffs, shortages occur in developing countries and the consequent hardships for the low- where more than 1 billion people have no income population. access to clean water, and supplies are unreliable for many more. Distribution of Governments can take several steps to water supplies is also inequitable, with the reduce the barriers to reform of the electrici- urban poor using unreliable and poor-quality ty sector and the adoption of DSM, supply- sources and paying excessive amounts to side efficiency measures and non-fossil fuel water vendors. Urban demands are often based generating technologies given priority over rural demands. ■ In strategic options assessment, it is important to identify and make transpar- The issues and options differ between ent the level and nature of subsidies industrial and developing countries: through the life-cycle cost of each option ■ In industrial countries, coverage rates are from extraction onwards. generally high, demand pressures are ■ Fiscal incentives or subsidies can stimu- moderate, and the focus is shifting to late implementation of emerging tech- reducing consumption and increasing nologies until prices drop enough for conservation in order to reduce or defer them to be commercially competitive. development of new supply sources. ■ Tariff policies influence the ability of ■ In developing countries, the low levels of utilities to upgrade and improve existing service coverage coupled with high assets and invest in new supply options demand pressures have focused attention and are currently being used to allow on new supply sources. Poor performance consumer preference to enter the equa- of existing water supply systems and tion at the bottom line (for example weak management capacity are wide- European and other initiatives towards spread and a large proportion of urban ‘green’ power pricing). poor and disadvantaged communities’ lack access to water supply. ■ In developing countries, where availability and cost of electrical services are of Still, the exhaustion of supply sources near greater concern, tariff blocks that in- urban centres and deterioration of water crease as consumption increases and life- quality are issues of concern across both line tariffs for low-income consumers, developed and developing countries.

156 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

A number of opportunities exist to meet retrofit water-saving devices. In various kinds of water supply needs. In Denmark, per capita consump- developing countries, extending services to tion dropped 24% over 10 years unserved populations in rural and urban due to widespread adoption of areas and revitalising sources that currently water-efficient technologies, supply water to urban areas are priorities. including toilets, showers, and Local resources and strategies may be more washing machines.58 appropriate, effective and affordable in providing discrete supplies to communities. In many locations, including In situations where water needs cannot be the United States, South Africa met through local sources, other solutions and Europe, block tariff structures with low are available. Demand-side measures, such lifeline tariffs and progressively rising rates as influencing consumption patterns, are for high levels of consumption have been relevant in industrial countries and among successful in controlling or reducing water high-consumption urban water users in demand. For these to be effective an effi- developing countries. New sources of supply cient yet cheap system of water metering is are limited, however. essential.

Demand-side management A significant proportion of high-quality domestic water is used in conventional DSM is gaining widespread acceptance in sewerage systems to transport waste and many industrial countries. Its potential to treat it although coverage varies. However, reduce consumption is proportional to the in Japan conventional sewer- prevailing level of use. DSM has a high age covers only 39% of the In many locations, potential in the United States, where population, compared with including the United average per capita consumption is about 400 80% in Europe.59 A number of States, South Africa and litres per capita per day. Developing coun- low-cost and alternative Europe, block tariff tries generally have low average per capita sanitation systems that have structures with low consumption rates, yet there is considerable low water requirements are lifeline tariffs and variation and potential for savings among available, such as offset pit progressively rising high volume users. In New Delhi, for latrines and septic tanks used rates for high levels of example, daily family use for those sup- in parts of South America. In consumption have been plied with piped water ranges from 700 India, 10 million people use a successful in controling litres for low-income families to 2 200 two-pit pour flush toilet, which or reducing water litres for affluent families.57 Tariffs are has the potential of replication demand. heavily subsidised, with little incentive for on a large scale.60 conservation. Community management of Washing and sanitation are heavy users of water supply systems has a positive impact water in domestic and commercial settings. on both coverage and efficiency. This Institutional standards, regulations and system works best when using discrete water sanctions are necessary to promote water- sources in rural and urban areas, and it can efficient technologies, including regulatory reduce demand for external water resources. standards for appliance and equipment For example, in Bwera, Uganda, a commu- manufacturers and subsidies to consumers to nity manages a large-scale system where a

The Report of the World Commission on Dams 157 Chapter 5

gravity-flow water supply serves over 60 000 northeast Thailand are typical of the people.61 increasing number of traditional rainwater harvesting technologies that have been Education and information are key elements upgraded to provide affordable and sustaina- in a long-term commitment to more effi- ble supplies.64 With the rising cost of cient use of water. Priority areas include conventional water supply, rooftop catch- raising awareness about efficient technolo- ment systems are spreading in Argentina, gies and practices, financial incentives and Barbados, Brazil, Costa Rica, Dominican savings, metering, and the rationale under- Republic, Chile, Mexico and Peru.65 Rain- lying any new tariff structures. water harvesting is not restricted to rural areas (see Box 5.6); it is also used by mil- Supply-side alternatives lions of residents on the peripheries of new The loss of water from leakage, illegal cities, like those in Tegucigalpa, Honduras. connections, and measurement problems is A related measure for increasing water high. In Asian cities, this can amount to supply in rural areas is the placement of 35-40%, with individual cities reaching as artificial arrays in areas of high fog density high as 60%.62 Stabilising and reducing to capture cloud moisture. losses from piped systems can help extend services. For example, figures in the United In places with short, intensive rainy seasons, Kingdom suggest a 29 litre per capita per day much of the rainwater runs off already decrease in system loss as a result of manda- saturated surfaces. Even minor earthworks tory leakage targeting set by the regulatory such as contour bunds and desilting existing authority. Proposals for a new dam in village ponds can increase short-term above- Yorkshire were deferred by introducing a surface storage enabling infiltration to occur leakage reduction programme and other over a longer period of time. Where sub- conservation measures.63 To succeed, surface storage is available these methods leakage reduction programmes require can raise the water table and make ground- strong institutional capacity and regulation water available longer into the dry season. to enforce standards. Reducing system- In Chennai, India, the metropolitan water operating pressure may offer a relatively supply utility tackled the problems of simple initial step to reduce leakage. groundwater depletion and seawater intrusion through a conservation-based strategy. Rainwater harvesting through rooftops, A series of check dams increased groundwa- tanks, and other methods are an alternative ter levels by 5-10 meters.66 These were source of domestic water supply. Major rural supplemented by controls on private water supply initiatives in Gansu, China and in extraction and compulsory rainwater harvesting for new buildings.

Box 5.6 Rainwater harvesting is spreading to urban areas As previously discussed in this chapter,

In Germany subsidies are available to encourage households to construct maintenance of natural vegetation can have rainwater tanks and seepage wells. Due to savings in monthly water charges and important effects in terms of improving other concessions, investments are paid back in 12 years. In Japan, 70% of the facilities in Tokyo’s Ryogoku Kokugkan sumo wrestling arena are supplied by water quality, but with variable effects on stored rainwater. water supply. In South Africa, the Working Sources: Gould, 1999, Contributing Paper for WCD Thematic Review IV.3 Water Supply; Down to Earth, 1998, p23 for Water programme in the Western Cape region supports the eradication of alien

158 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

vegetation, thereby enabling the restoration scrutiny, particularly as there will be no of the indigenous fynbos vegetation. Fynbos return flows within the basin as would occur provides less cover and vegetative mass and with other abstractions. As noted above, thus the programme aims to provide both water can be transferred from agricultural employment and water yield gains.67 Efforts use through alternative and more proactive by New York and New Jersey to purchase programmes, such as water trading, that and protect the Sterling Forest from devel- effectively cross-finance water efficiency opment brought renewed attention to the gains in irrigated agriculture. water quality benefits of forest management around single-purpose facilities near urban Enabling conditions areas. Sterling Forest protects a catchment As with the other sectors, a Meeting the needs of that provides a series of reservoirs that commitment to policy, institu- those currently not supply water to more than 2 million peo- tional and management reform is served in both urban ple.68 In the latter case, catchment manage- essential to implement demand- and rural areas is a ment serves to avoid the need to invest in side management of water priority and requires water treatment facilities or, in the extreme supplies and alternative supply concerted efforts. It case, new supply sources. Such options are approaches. Initiatives include requires an appropriate highly location specific. the following: mix of demand-side management measures The recycling of wastewater can be another ■ In the United Kingdom, to enhance water use significant source of supply. By 1999, United States and Australia, efficiency and reduce enough water was recycled in the Bay Area regulations mandate explora- wasteful consumption, of California to meet the needs of 2 million tion of all economically increases in supply people. The target there is to increase justifiable demand-manage- efficiency and the capacity to serve 6 million people by 2020.69 ment initiatives before a development of new Agriculture used 32% of the recycled water, licence for new abstraction is supply sources. 27% went to groundwater recharge, 17% granted. supported landscape irrigation, 7% went to ■ Management efficiency is industry, and the remainder for environmen- fundamental to conservation and im- tal and other uses. proved planning. No single public or private management approach is appro- Desalination contributes to water supply in priate for all contexts. Stronger institu- 120 countries, with 60% of the 11 000 tional capacity and accountability are desalination plants being located in the required to improve the performance of water-scarce Middle East. Though costs of public agencies. desalination have been reduced dramatical- ■ ly, they are still high and the technique is Effective regulatory mechanisms need to very energy intensive.70 Meaningful poten- be in place to safeguard access and tial exists, but a major technology break- affordable cost of water for the urban and through is required to increase desalination’s rural poor. global contribution significantly. Meeting the needs of those currently not Inter-basin transfer of water is often offered served in both urban and rural areas is a as a solution to local water scarcity. The priority and requires concerted efforts. It impacts of such transfers require careful requires an appropriate mix of demand-side

The Report of the World Commission on Dams 159 Chapter 5

management measures to enhance water use be neither achievable nor desirable. The efficiency and reduce wasteful consumption, more appropriate objective is to predict, increases in supply efficiency and the manage, and respond to the flood situation development of new supply sources. In the in order to prevent widespread losses and latter case, due consideration is required to obtain the best outcome in each situation. non-conventional alternatives such as reducing conveyance losses, rainwater Following major destructive flood events in harvesting, catchment management and the last few decades, significant changes in water recycling. flood policy have occurred around the world, redefining interventions to some Integrated Flood extent. For example, Management ■ coastal flooding of 1953 led to the Delta works in the Netherlands; Floods differ widely in their nature, in the ■ the 1988-89 floods in Bangladesh led to characteristics of the affected floodplains, the Flood Action Plan and the National and in their implications. As documented in Water Management Plan; and earlier chapters, in some contexts millions ■ the Upper Mississippi floods of 1993, the of people are dependent on annual floods for Rhone floods in 1993, the 1997 floods in their livelihood, while in others similar the Rhine and the 1998 flood in China floods threaten life and property. A number drew attention to the role of non- of options have emerged to predict, manage structural catchment measures. and respond to floods and at the same time obtain the most socially beneficial and Intervention strategies in flood management economically sustainable outcomes. Institu- have gradually shifted from a focus on tional and policy support is needed to structural responses to flood control to achieve integrated flood management, introducing or expanding the role of non- which consists of flood alleviation and structural responses as part of integrated utililisation, mitigation and risk manage- strategies for floodplain management.71 ment rather than a strict reliance on structural flood control based on dykes, levees This section looks at the strategies and and dams. options available for integrated flood management and control. These strategies Floods are a complex phenomenon, and the consist of three broadly complementary success of flood intervention strategies approaches, namely; depends on a number of factors. Chapter 2 ■ considered the performance of dams as a reducing the scale of flood through a structural response to flood control and number of structural and non-structural highlighted the basis for concerns and the means; shift towards flood management as an ■ isolating threats through structural, approach in place of flood control. Vulnera- technological and policy alternatives; and bility to flood events determined by charac- ■ increasing people’s capacity to cope teristics of the population and land-use effectively with floods. systems at risk and their capacity to cope and recover, are key issues in flood manage- As with the role of large dams, associated ment strategies. Absolute flood control may structural measures and the relationship

160 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

Table 5.1: Complementary approaches to flood stream uses should be consid- management ered. In assessing the likely Reducing the scale Isolating the threat Increasing people’ s effectiveness of source controls, of floods of floods coping capacity ■ Better catchment ■ Flood embankments ■ Emergency planning pre-flood conditions (such as management ■ Flood proofing ■ Forecasting frozen or saturated ground) must ■ Controlling runoff ■ Limiting floodplain ■ Warnings ■ Detention basins development ■ Evacuation be considered. ■ Dams ■ Compensation ■ Protecting wetlands ■ Insurance Flood flows can be stored in the lowlands, as detention basins are normally dry except when required for flood between floods and natural floodplains have storage. In some cases, lakes on rivers can be been discussed in earlier chapters, and used, such as the Dongting lake in Hunan therefore the focus here is on alternative province in China. Traditional tank systems measures for flood management. in flood-prone villages of Madhubani district, in Bihar, India, are used for floodwater diver- Reducing the scale of floods sion and storage.73 Natural wetlands are also Reducing the scale of floods implies manag- important assets for flood storage and agricul- ing the quantity and quality of surface water tural fields can be used for micro-storage.74 runoff. Catchment management measures include: Isolating the threat of floods ■ infiltration measures, such as infiltration Along with dams, earthen embankments, trenches, detention basins, infiltration levees, dykes and bunds have been the ponds, retention ponds and wetland dominant flood-control option in most of areas to reduce runoff; and the world. One key characteristic of this ■ forest protection, reduced impact logging option is that it interferes with natural practices, avoidance of clear felling drainage patterns in the area protected. This techniques and less intensive agricultural area may still be flooded from local precipi- practices to reduce soil erosion and tation, with minor tributaries causing local landslides that lead to channel siltation, flooding, inland flooding or drainage con- raising flood levels proportionately. gestion. In the absence of adequate drainage, the protected area will suffer crop loss Small-scale storage of runoff and improve- due to waterlogging.75 Significant morpho- ments in drainage are other approaches to logical changes can occur following the flood mitigation, particularly at the local level. construction of embankments, as sedimenta- Check and warping dams (built for erosion tion of channels and bank erosion lead to control) can lead to groundwater recharge and raised river beds and the need for even store initial quantities of run-off during storm higher levels of protection (as noted in events. Wire mesh dams are also used for this Chapter 2). purpose in the highlands of Trinidad where the mesh traps water-borne debris during flash Structures such as dwellings may be modified floods, blocking much of the flow.72 in a variety of ways to reduce the risk of floodwater penetration: waterproofing walls; If afforestation is considered as part of a fitting openings with permanent or temporary packet of measures, its effects on the full doors, gates, or other closure devices; fitting range of hydrological function and down- one-way valves on sewer lines; or building

The Report of the World Commission on Dams 161 Chapter 5

boundary walls around the house structure. access to resources as a means of address- Other possible measures include sump-pumps ing flood vulnerability). 77 that begin operating in basements when water levels rise, and contingency plans and facilities Emergency planning and management has designed to be operated when a flood is three phases: preparedness, response and anticipated. Contingent flood proofing recovery. The capacity of individuals, house- depends upon a reliable flood warning sys- holds, groups, and communities to cope with tem.76 This approach also includes raising flooding depends upon their knowledge, dwelling places. For example, in the floodplain resources, organisation and power: kampungs (traditional villages) of Malaysia, ■ their knowledge about how to identify that houses are built on stilts to raise them above a flood threatens, how to mitigate effects of anticipated flood levels. Similar adaptations to floods, what to do before, during and after floods are found throughout Southeast Asia a flood, the causes of flooding and appro- along rivers, estuaries and coastlines. priate mitigation measures; ■ Increasing people’s coping the resources at their command, including their skills and physical assets, and capacities the support of others that they can call New integrated approaches for flood allevia- upon; and tion, mitigation and flood risk management ■ the extent of their organisation, includ- emphasise: ing within households, within neigh- ■ integrated catchment and coastal zone bourhood groups, and within whole management, and wise planning and use communities, as a way of pooling knowl- of floodplains and coastal zones; edge, skills, resources, and planning and ■ empowering local communities to make co-ordinating activities to achieve choices about land development and optimum use and power in relation to flood alleviation; other groups in society. ■ reducing the impacts of humans on the A flood management strategy will need to environment by promoting flood disaster cover flood warnings, flood mitigation, any resilience (see Box 5.7); necessary evacuation and post-flood recov- ■ valuing and preserving the best of ery. A clear commitment by national or indigenous adaptations and improving federal governments to the emergency local capacities to respond; and planning and management process will ■ addressing problems of equity (for enhance its effectiveness. example alleviating poverty and lack of Enabling conditions Box 5.7 Flood resilience Enabling conditions that will promote an

‘Resilience’ may be taken as the opposite of vulnerability and may be enhanced integrated approach to flood management by promoting access to knowledge and resources achieved through develop- include: ment processes and poverty reduction programmes. Alternatively, traditional social and cultural systems may enable resilience. The vulnerability of poor, rural ■ Promoting public involvement and Malaysians to floods in eastern Peninsular Malaysia is reduced by close kinship devolution of decision-making to the systems that exist in the floodplain villages. Source: WCD Thematic Review IV.4 lowest possible level enables integrated Flood Control, Section 4.2; Malaysia in Chan, 1995 catchment management. This is essential as integrated flood management strate-

162 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

gies at catchment level are most appro- ■ Improving system management can defer priate to local conditions. the need for new sources of supply by ■ Funding should have a multi-functional enhancing supply and conveyance approach. Integrated management of the efficiency. Needless loss of power and catchment will increasingly result in water can be avoided through reductions multi-functional options being adopted. in water leakages, improving system maintenance and upgrading control, ■ Institutional design is critical to the transmission and distribution technology success of a flood hazard management in the power sector. policy, and co-ordination across various institutions has been a crucial factor. ■ Basin and catchment management through vegetative and structural meas- ■ Flood hazard management and emergen- ures offers an opportunity across all cy response agencies have a key role in sectors to reduce sedimentation of enhancing local communities’ coping reservoirs and canals and manage the capacities by involving them in decision- timing and quantity of peak, seasonal making on all matters relating to floods. and annual flows, as well as groundwater The overriding message is that local owner- recharge. The multi-functional nature of ship of flood alleviation strategies and the hydrological system; the types and options is necessary for long-term success. importance of downstream uses of water; and the on-site costs and benefits of the measures themselves will determine the Findings and Lessons attractiveness of different interventions. This chapter has examined the options for ■ A number of supply options have fulfilling energy, water and food needs in emerged that are locally and environ- today’s circumstances and the barriers and mentally appropriate, economically enabling conditions that determine choice viable and acceptable to the public, or adoption of particular options. Many including water recycling, rainwater options currently exist – including demand- harvesting and wind and solar (off-grid) side management, supply efficiency and new power. supply options. These can all improve or expand water and energy services and meet The ability of various options to meet evolving development needs across all existing and future needs or to replace segments of society. An overview of options conventional supplies depends on the for all the sectors covered suggests the specific context, but in general they offer following general findings and lessons: significant potential individually and ■ Demand-side management options collectively. More specifically the different include reduced consumption, recycling sectoral alternatives are as follows: and technological and policy options ■ In the irrigation and agriculture sector, that promote water and power efficiency preference is for improving the perform- at the point of end-use. DSM has signifi- ance and productivity of existing irriga- cant untapped and universal potential tion systems and alternative supply-side and provides a major opportunity to measures that involve rain fed, as well as reduce water stress as well as achieve local, small-scale and traditional water other benefits such as the reduction of management and harvesting systems, greenhouse gas emissions. including groundwater recharge methods.

The Report of the World Commission on Dams 163 Chapter 5

■ The priority for achieving a sustainable floods through structural, non-structural, and equitable global energy sector is for all technological and policy alternatives, and societies to increase the efficiency of increasing people’s capacity to cope with energy use and the use of renewable floods. sources. High-consumption societies must also reduce their use of fossil fuels. Decen- Numerous market, policy, institutional, tralised, small-scale options based on local intellectual and regulatory barriers hinder the renewable sources offer the greatest near- emergence and widespread application of an term and possibly long-term potential in appropriate mix of options in response to rural areas. needs in the power and water sectors. The ■ In the water supply sector, meeting the barriers to be overcome include capacity and needs of those currently not served in both resource constraints, the dominance of urban and rural areas through a range of conventional approaches and interests in efficient supply options is the priority. development planning, a lack of awareness Further efforts to revitalise existing sources, and experience with non-conventional introduce appropriate pricing strategies, alternatives, inadequate access to capital and a encourage fair and sustainable water lack of openness in the planning system. marketing and transfers, recycling and These are further analysed in the next chapter. reuse, and local strategies such as rainwater While they are context-specific, hidden harvesting also have great potential. subsidies and other incentives to conventional ■ In the case of floods, as absolute flood options may limit the use and rate of adoption control may be neither achievable nor of even superior alternatives. To better enable desirable, it is necessary to manage floods the selection and use of the broader range of to minimise flood damage and maximise options will require that options are compre- ecological benefits. An integrated ap- hensively and fairly evaluated by all stakehold- proach to flood management will involve ers throughout the planning, decision-making, reducing a community’s vulnerability to and financing process.

Endnotes

1 Sources are cited in the sections below only 8 Huasham et al, 1995 in WCD Thematic when they do not appear in these Thematic Review IV.2 Irrigation Options, Annex 8. Reviews or where it is otherwise necessary for 9 Mitchell, 1995, in WCD Thematic Review clarity. IV.2 Irrigation Options, p118. 2 Postel, 1999, p41; FAOSTAT, 1998. 10 Murray-Rust and Vander Velde, 1994. 3 WCD Thematic Review IV.2 Irrigation 11 OED, 1990, p4. Options, Section 1.3. 12 FAO, 1995, p280. 4 Bosch and Hewlett, 1982. 13 WCD India Country Study, Section 3.3.1. 5 Bruijnzeel, 1990. 14 FAO op cit, p233. 6 Postel, 1999, p93. 15 FAO, op cit. 7 WCD Thematic Review IV.2 Irrigation 16 Seckler, 1996. Options, Section 4.3.4. 17 Cornish, 1998, p20.

164 Dams and Development: A New Framework for Decision-Making Options for Water and Energy Resources Development

18 Cornish, op cit. 48 UNDP et al, 2000. 19 Frausto, 1999, Contributing Paper for WCD 49 UNDP et al, 2000. Thematic Review IV.2 Irrigation Options, p18. 50 Rumsey and Flanigan, 1995; Worrell, 1999 20 FAO et al, 1999. working draft. 21 Vermillion, 1997. 51 UNDP 2000, Chapter 6 p1. 22 WCD India Country Study, Section 3.3.6; 52 Sant et al, 1999 eco013, WCD Submission. Vermillion, op cit. 53 IEA, 1998 and balances of non-OECD 23 Bandaragoda, 1999; Vander Velde and Tirmizi, countries. 1999. 54 Kowalski and Schuster, 2000, p165. 24 Brehm and Quiroz, 1995; Hearne and Easter, 55 WCD Thematic Review IV.1 Electricity 1995 cited in Hearne and Trava, 1997. Options, Section 3.4. 25 WCD Thematic Review IV.2 Irrigation 56 EWEA, 1999, Contributing Paper for WCD Options, Section 3.4.1. Thematic Review IV.1 Electricity Options, 26 Agrawal and Narain, 1997; Thakkar, 1999, Annex H. Contributing Paper for WCD Thematic 57 WCD Thematic Review IV.3 Water Supply Review IV.2 Irrigation Options. Options, Section 6.5 27 Frausto, op cit. 58 White et al, 1999, eco018, WCD Submission, 28 WCD India Country Study; Agrawal and p9. Narain, op cit; Barrow, 1999. 59 Lane, 1999, Contributing Paper for WCD 29 Ringler et al, 1999, p10. Thematic Review IV.3 Water Supply Options. 30 Agrawal and Narain, op cit. 60 WCD Thematic Review IV.1 Electricity 31 WCD Thematic Review IV.2 Irrigation Options, Section 2.4.2. Options, Annex 1. 61 WCD Thematic Review IV.1 Electricity 32 Acreman et al, 1999, Contributing Paper for Options, Section 3.12. WCD Thematic Review II.1 Ecosystems. 62 McIntosh and Yñiguez, 1997, cited in WCD 33 WCD Thematic Review IV.2 Irrigation Thematic Review IV.3 Water Supply Options, Options, Annex 5. Annex 1. 34 WCD Thematic Review IV.2 Irrigation 63 Yorkshire Water, 1997, eco082 WCD Submis- Options, Annex 6. sion A7.1. 35 Shevah, 1999 in WCD Thematic Review IV.2 64 Gould, 1999, op cit. Irrigation Options, Section 4.3.2. 65 Ringler et al, 1999, p10. 36 DFID, 2000. 66 SANDRP, 1999 opt080, WCD Submission, p20. 37 FAO, op cit, p234. 67 Preston, 1999. Contributing Paper for WCD 38 Ministry of Water Resources and Electric Thematic Review IV.3 Water Supply Options. Power, PRC, 1987, cited in Postel, 1999, p56. 68 Stapleton, 1996, p2-5. 39 National Research Council, 1996, cited in 69 Dickinson, 1999, Contributing Paper for WCD Postel, 1999, p77. Thematic Review IV.3 Water Supply Options. 40 Thakker, op cit. 70 Costs range from $1.50-5.00 per cubic metre. 41 Dhawan, 1998, cited in WCD Thematic 71 WCD Thematic Review IV.4 Flood Manage- Review IV.2 Irrigation Options; Molden et al, ment Options, Section 1.2.3. 1998. 72 WCD Thematic Review IV.4 Flood Manage- 42 WCD Thematic Review IV.2 Irrigation ment Options, Section 4.4.2. Options, Section 3.2.3. 73 WCD India Country Study, Annex 5. 43 ADB, 1999c. 74 Delaney, 1995 in WCD Thematic Review IV.4 44 Smith, 2000, Contributing Paper for WCD Flood Management Options, Section 4.4.2. Thematic Review IV.2, p17, 30. 75 WCD Thematic Review IV.4 Flood Manage- 45 UNDP et al, 2000. ment Options, Section 4.4.5. 46 Flavin, 1999, Contributing Paper for WCD 76 WCD Thematic Review IV.4 Flood Manage- Thematic Review IV.1 Electricity Options, ment Options, Section 4.4. Annex H; Economist, 5 August 2000. 77 WCD Thematic Review IV.4 Flood Manage- 47 IEA, 2000. ment Options, Section 1.2.4.

The Report of the World Commission on Dams 165 Decision-Making, Planning and Compliance

Chapter 6: Decision-Making, Planning and Compliance

he previous chapters suggest Tthat the main challenge for water and energy resource developers in the 21st century will be to improve options assessment and the

performance of existing assets. This will require open, accountable and comprehensive planning and decision-

making procedures for assessing and selecting from the available options. It also calls for monitoring programmes,

evaluation procedures and incentive mechanisms that ensure compliance with project commitments, especially

in the area of environmental and social performance. To do this we need a

The Report of the World Commission on Dams 167 Chapter 6

better understanding of why large dams were tion have meant that many dams were not proposed and developed, and why failures in built based on an objective assessment and performance and impacts on ecosystems and evaluation of the economic, social and affected people are still not properly environmental criteria that apply in today’s accounted for, monitored or context. resolved. This chapter draws on Once a proposed dam the WCD Knowledge Base to Conflicts over dams have heightened in the project passed characterise the critical problems last two decades. This results from dissatis- preliminary technical encountered in the past, analyse faction with the social and environmental and economic feasibility their underlying causes and impacts of dams, and their failure to achieve tests and attracted chronicle recent developments targets for costs and benefits. It also stems interest from that point the way forward, this from the failure of dam proponents and government or external is explored in the remainder of financing agencies to fulfil commitments financing agencies and the report. made, observe statutory regulations and political interests, the abide by internal guidelines. In some cases, momentum behind the As a development choice, large the opportunity for corruption provided by project often prevailed dams often became a focal point dams as large-scale infrastructure projects over further for the interests of politicians, further distorted decision-making, planning assessments. dominant and centralised govern- and implementation. Whereas substantial ment agencies, international improvements in policies, legal require- financing agencies and the dam-building ments and assessment guidelines have industry. Involvement from civil society varied occurred, particularly in the 1990s, it with the degree of debate and open political appears that business is often conducted as discourse in a country. However, there has usual when it comes to actual planning and been a generalised failure to recognise affected decision-making. Further, past conflicts people and empower them to participate in remain largely unresolved due to a number the decision-making process. of reasons, including the poor experience with appeals, dispute resolution and recourse Once a proposed dam project passed prelim- mechanisms. inary technical and economic feasibility tests and attracted interest from government The key to improved performance in the or external financing agencies and political future lies in screening out undesirable dams interests, the momentum behind the project projects as part of a process that considers often prevailed over further assessments. In the full range of options for water and any event project planning and appraisal for large dams was confined primarily to techni- energy power services, and responds posi- cal parameters and the narrow application of tively to changing priorities. These efforts economic cost/benefit analyses. Historically, must find ways to ensure that performance social and environmental impacts were left in living up to existing institutional arrange- outside the assessment framework and the ments governing the planning and project role of impact assessments in project selec- cycle is improved. tion remains marginal even in the 1990s. The influence of vested interests in the The chapter groups these topics under three decision-making process and the narrow, headings: decision-making, planning and technical approach to planning and evalua- compliance.

168 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

Decision-making and the State-led decision-making Political Economy of Large Governments were the proponents for Dams practically all large dams and many large dams were built by government agencies Large dams arise from a series of decisions themselves.1 Centralised agencies or utilities taken from the beginning of the planning have traditionally managed the water and process through to the final approval of a energy sector within government. Like most project and financial closure. At each stage major development projects, decision-making different actors are involved, including processes around large dams have been government agencies, public or private centralised and technocratic in virtually all utilities, interested parties from the region, parts of the world, particularly through the financing agencies, consulting and construc- 1970s. The exception may be certain large tion companies and equipment suppliers. dams built as part of regional development Affected people and NGOs are increasingly projects where local political interests have involved as well, often through people’s played important roles in promoting projects – movements against dams. Each of these groups often in conjunction with their representa- promotes its own self-interest throughout the tives in central government. process, ranging from profits and political power to property rights and livelihoods. This Indeed, the degree to which decision- section looks at the interplay of these forces in making surrounding a dam was politicised the context of rivers, dams and the develop- and the level at which the decision was ment of water and energy resources. made varied tremendously with the project. Large and spectacular dams have often been Similar pictures emerge for the industrial- seen as symbols of development and nation ised and the developing worlds. Planning building, a potent demonstration of man’s processes are controlled by single-purpose ability to harness nature’s forces and a government agencies or public utilities and tangible ‘deliverable’ for politicians, usually the decision to build is taken as the outcome funded from the public purse. The WCD Case of a fairly limited set of political interactions Studies demonstrate that for very large dams at political levels commensurate with the the decision to build often was taken by heads size and importance of the dam. In the case of State, whereas smaller facilities were of developing countries, the selection of typically guided through the process by the alternatives for meeting water and electric relevant agencies or utilities (see Box 6.1) power needs was, and is, frequently constrained by preferential access to interna- Dam-building in industrialised tional finance and the pre-existing interna- countries tional expertise in large dams rather than alternatives. Recently, restructuring and In industrialised countries, alliances be- reform of the energy and water sectors in tween local political interests and powerful, many countries – both industrialised and single-interest agencies and utilities respon- developing – has changed the role of gov- sible for water and power development ernment in decision-making and planning, drove planning and decision-making on with private investors and corporations large dams. In the United States, the taking both financing and ownership roles political desire to settle and develop the in these projects. land and resources of the western states

The Report of the World Commission on Dams 169 Chapter 6

Congress, which further scrutinised the project plan.2

Outside the United States, the reconstruction of Europe after World War II led to the construction of many large dams. The Marshall Plan ushered in the era of foreign aid with the transfer of $17 billion to help rebuild Europe.3

During the cold war era, centralised, state- driven consolidation of resources through interventions such as the building of large dams was the hallmark of communist encouraged the construction of large dams. regimes. Most of the political and economic At the same time, however, laws governed decision-making processes for the large dams the planning and approval process. They built in eastern and central Europe were required agencies and utilities to perform a top-down and technocratic. Besides the long series of surveys, hold public hearings, central government, other stakeholders and and conduct inter-agency reviews, including the general public were not in a position to cost-benefit analysis. The appropriation of express their concerns or represent their funds ultimately required approval by the interests in the decision-making process.4

Dam-building in developing Box 6.1 WCD Case Studies: political decisions to build large dams countries From the WCD Case Studies, the predominant role of the State can be seen The success of the Marshall plan in Europe throughout. In the Glomma and Laagen Basin the Norwegian government was active in licensing hydropower projects initially to promote development in led to great optimism that the key to nation- isolated river valleys, then to feed power based smelting industries and other al development was investment in capital heavy industries in the period after the Second World War. Hydropower development was also promoted to support specific districts in periods of stock. The International Bank for Recon- depression and high unemployment. struction and Development (IBRD), created Similar government intervention is demonstrated at the Grand Coulee project in to help finance the reconstruction of war- the United States where a presidential decision was made to proceed with the Grand Coulee project in 1932. The project formed part of the federal govern- torn European countries, became a focal ment’s campaign to bring the country out of economic depression, provide point for these efforts and, alongside bilater- construction jobs to eight thousand people, reclaim land for irrigation and reduce price manipulation by private power companies, thereby making publicly al development banks, helped export the generated electricity more widely available at low cost. model of centralised nation building for The planning, implementation and initial operation of the Kariba project was economic development. Dams fit well with done by the Inter-Territorial Power Commission of the then Central African Federation (the former colonies of Northern and Southern Rhodesia now Zambia this model of foreign aid and were often the and Zimbabwe) in the 1940s. The priority was to deliver power to the copper first visible sign of IBRD (later called the industry owned by multi-national corporations. World Bank) presence in a country.5 On the Orange River in South Africa, the proposal to build a major dam and water diversions scheme was called for by Prime Minister Hendrik Verwoerd following the Sharpeville massacre in 1960 that undermined confidence in the 6 government and led to outflows of foreign capital. One primary motivation was Role of foreign assistance to demonstrate national capacity to build major projects and to restore international confidence in the country’s development and investment potential. Both the multilateral and bilateral develop- Source: WCD Case Studies ment banks played a significant facilitating role in getting Asia, Africa and Latin

170 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

America started in the dam business. The ability of individual states to directly access World Bank began financing large dams in the foreign assistance and provided incentives for 1950s, committing on average over $1 billion doing so. per year to this purpose (Figure 6.1). For the period from 1970 to 1985 this amount had Since then World Bank loans to India have risen to $2 billion per year. Adding in finance doubled or tripled each decade. By one by the Asian, Inter-American, and African estimate loans for irrigation, drainage and Development Banks, as well as bilateral flood control are 14% of World Bank loans funding for hydropower, suggests total financ- to India.8 The India Case Study reports ing for large dams from these sources of more that, in total, foreign assistance provides than $4 billion annually at the peak of lending about 13% of public sector outlays in the during 1975-84. irrigation sector, with the World Bank Group accounting for almost 80% of this Bilateral and multilateral development assistance. Thus, in India the World Bank financing agencies have helped finance did not provide the initial impetus behind studies needed for dam construction, and lent money for the construction of the Figure 6.1 Development assistance for large dams, 1950-1999 dams themselves. They identified development goals through strategic sectoral planning documents, provided resources 25 and technological capacity to conduct feasibility studies, and created basin-wide institutional frameworks to plan and 20 implement dams. Although the proportion of investment in dams directly financed by bilaterals and multilaterals was perhaps less 15 than 15%, these institutions played a key strategic role globally in spreading the technology, lending legitimacy to emerging 10 dam projects, training future engineers and US$Billions, 1998 prices government agencies, and leading financing arrangements.7 5

The extent and nature of this influence varied from country to country and from 0 region to region. The India Case Study locates the orientation of Indian planners 1950-54 1955-59 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 and engineers towards dams as the princi- World Banka Bilateralb IDBc pal response to water resource develop- ADBd AfDBe EBRDf ment in the 1950s and 1960s when large numbers of dams were first built. This Source: aSklar and McCully, 1994 eco029, WCD Submission and World Bank, predated the World Bank’s major involve- 2000; bOECD, 2000a; cIDB, 1999; dLagman, 2000; eAfDB, 1998; fEBRD, ment in India. The Bank began lending in 1996, 1999, 2000a, 2000b. earnest to India in the 1970s at a time when Notes: Data for bilateral agencies also includes financing by the Commission of the European Community and includes only all hydropower investments from policy reforms removed restrictions on the 1975 to 1997.

The Report of the World Commission on Dams 171 Chapter 6

the tendency to choose dams as the response Only in the late 1980s and early 1990s has to water and energy needs, but rather this lending activity tailed off in the face of provided continued and increasing external increasing public scrutiny and criticism by backing to the large number of dams which civil society (see Figure 6.1). The decline were built from the 1970s onwards. followed unfavourable independent reviews of two high profile projects that were As in the case of India, the WCD China supported or considered by the World Bank Case Study shows that dam building was – Sardar Sarovar in India and Arun III in well advanced prior to the entry of foreign Nepal. A number of other factors contribut- donors. Brazil also follows this pattern. ed to the shift away from large dam projects. Comparison of statistics on large hydropow- They include: er dams commissioned in Brazil between ■ continued criticism of the pervasive 1950 and 1970 and the finance provided by ‘approval culture’ of the World Bank and the World Bank and the Inter-American its willingness to promote large infra- Development Bank (IDB) show that just structure projects; over 10% of the 79 large dams listed in the ■ internal evaluations of the Bank that International Commission On Large Dams documented ever-increasing ‘appraisal (ICOLD) database received financial optimism’ despite evidence of poor assistance from these donors. However, the economic and financial performance by figure rises to over 30% of the 47 dams for projects in the water supply and irriga- the 1970-1990 period. Foreign assistance, tion sectors; thus, did not drive the selection of dams as ■ failure to meet the Bank’s poverty an option but did provide significant fi- alleviation goals; and nance during peak dam-building periods. ■ growing recognition of the severity of the social and environmental impacts of The picture differs for smaller countries. In dams.11 Colombia, multilaterals helped fund the first large dam and 40% of the subsequent 50 More recently, a gradual shift towards an large dams appearing in the ICOLD data- increased role for private sector finance in base. Multilaterals have played a particularly hydropower and, to a lesser extent, water strong role in countries that have not built supply, have also led the banks to move into many dams and do not have local planning a facilitation role with the emphasis on and construction expertise and capacity. In public-private partnerships and risk guaran- Costa Rica, which relies on hydropower for tees. Part of the financing has now been roughly 90% of its power generation, the taken over by export credit guarantee World Bank and IDB had directly supported agencies in donor countries that finance and over half of the installed hydropower capacity underwrite risks taken by home-country 9 by the mid-1990s. In Tanzania, bilateral engineering firms and equipment suppliers agencies and the World Bank have supported participating in projects abroad. essentially all the large hydropower dams.10 In these smaller countries the role of financing Role of industry and bilateral agencies and the firms they employ to funding undertake preparatory studies, design Ultimately it is the country government projects and build dams can be significant. that is responsible for taking the decision to

172 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

build a dam. However, governments are and developing countries alike. The associ- naturally influenced by international ations play an important role in building expertise and financing opportunities (see capacity of member countries by collecting Box 6.2). Once a government is politically and disseminating technical and other committed and construction has begun, the information and holding annual meetings to nature of large construction projects makes promote formal and informal professional it extremely hard to change course, even if exchange. there are cost overruns, unforeseen negative impacts, or benefits are less than predicted. Decision-making on shared The public purse generally carries the risk of rivers poor economic performance, and there has The flow of water through States or prov- historically been no consequence or liability inces sharing a basin links them inextricably for building under-performing dam projects. to a finite and common resource (see Box 6.4). Yet water resources and energy plan- For industrialised countries with a history of ning has frequently been undertaken at the dam-building and expertise in related level of administrative or political units that equipment, bilateral overseas aid has often do not coincide with the watershed. As a become a vehicle for supporting local industry by exporting this expertise through aid programs tied to the purchase of services Box 6.2 WCD Case Studies and submissions: foreign involvement in or equipment from the donor country.12 dam projects

Conflicts of interest have inevitably resulted While the WCD Case Study dams built in the United States and Norway relied between the financing agency’s interest to exclusively on national capacity, the Case Study dams in developing countries reveal the involvement of foreign firms in master plans, inventories, feasibility provide contracts for home-country compa- studies, design, construction and financing. In the case of Tarbela the World nies and the borrower or grant recipient’s Bank even co-ordinated the Indus Water Treaty signed between India and Pakistan that gave Pakistan the opportunity to build Tarbela. interest in providing appropriate and In Turkey, the comprehensive development of the water resources in Ceyhan affordable development. In the case of Basin was outlined first in a 1966 study by a foreign consulting firm financed by bilateral agencies these conflicts of interest the United States Agency for International Development (USAID). This document has guided investments in the basin for the last thirty-five years. The US may be exacerbated in smaller, poorer Bureau of Reclamation, through USAID, undertook the initial study of the water countries where the donor plays a more resources of the Tocantins Basin in 1964, where Tucuruí was later built. For the Pak Mun project in Thailand, French engineering firms conducted the initial central role in financial matters (see Box feasibility studies in the 1970’s and early 1980s. 6.3). In all, the World Bank provided financing for four of the Case Study dams (Kariba, Tarbela, Aslantas and Pak Mun). Kariba was partially financed by the copper companies for which much of the power was destined. The decision of Professional associations such as ICOLD, multi-national aluminium producers to invest in the Carajas region of Amazonia the International Hydropower Association was subject to the decision to proceed with the Tucurui hydropower complex. Financing for the project came from internal sources and French banks. (IHA) and the International Commission A recent NGO report provides details on the role of 12 European companies in on Irrigation and Drainage (ICID) have also the design, construction and supply of equipment to 84 large dams, many of played an important role in setting standards them major dams in developing countries. The report also lists the further involvement of these companies in technical studies of a larger sample of dams. within their technical disciplines and Many of the projects listed are financed by home-country bilateral agencies, promoting professional capacity related to export credit agencies and commercial banks, as well as by the multilateral development banks. The report documents the billions of dollars that have gone the building of large dams and their associ- to the European ‘dam building’ industry from projects in developing countries. ated infrastructure. These are international Source: WCD Case Studies and Lang et al, 2000 associations made up of members from eco041, WCD Submission government and industry from industrialised

The Report of the World Commission on Dams 173 Chapter 6

Box 6.3 Nordic influence in the Pangani Falls Redevelopment Project, means of water storage, dams play an Tanzania important role in the management of the

A 1985 Canadian study provided Tanzania with a national energy development resource and its allocation to different uses plan that led to the decision to redevelop the old Pangani dam, raising its within and between countries. In the installed capacity from 17 MWs to 66 MWs. The Finnish International Develop- ment Agency (FINNIDA) funded the $2.5 million feasibility study in 1989-90, context of shared rivers, dams are a technol- which was carried out by Finnish and Norwegian consultancy firms. Given the ogy that allows an upstream riparian to close relationship between the Finnish firm and FINNIDA, the firm not only wrote the terms of reference for the feasibility study but later was also given partially ‘privatise’ the river by storing and contracts to procure supplies and supervise construction (jointly with its using water and thereby effectively exclud- Norwegian partner). In the event, the feasibility study confirmed that the dam was the best option to meet sector needs and the EIA concluded that no ing downstream riparians from access to the adverse effects existed that would prejudice the project. water. In the downstream context, when As the Finnish, Swedish (SIDA) and Norwegian (NORAD) aid donors planned to faced with dwindling supplies from up- finance the project, SIDA hired a Swedish firm which reviewed and confirmed the results of the feasibility study. The three Nordic donors subsequently stream, dams provide downstream riparians approved grants to Tanzania to cover the costs of the project. While the aid was with a practical means of replacing lost dry not ‘tied’, no competitive bidding was undertaken for contracts, rather, checks were made to ensure that prices offered by selected firms were competitive. A season flows by storing wet season flows. Norwegian firm supplied the turbines, a Swedish firm the generators and control equipment and a number of Finnish firms were involved in the civil works and transmission lines, including the parent company of the consultancy firm that Relative power relationships within basins undertook the feasibility study. determine to a large extent how individual While the final 1991 project document stated that the hydrological risks to the countries interact and whether other project were small, there was sufficient concern over the availability of water riparians are consulted concerning dam supply to the project to make the Nordic donors insist on a water basin management board as a condition of the funding agreement. This decision has projects. A regional power that holds an engendered conflicts between local, national and donor interests. The water upstream position is in a better situation to board was to institute water fees to limit irrigation withdrawals and ensure an adequate supply of water for power generation at Pangani. With work underway implement projects without consultation, in the early 1990s concern mounted as precipitation in the basin and flows at the site fell well below the 1981-92 averages. This was compounded by a lack of and this has been the case in Turkey, India information on the extent of upstream withdrawals which fed traditional and China. In other cases powerful down- smallholder agriculture by the Chagga people on the slopes of Mount Kiliman- jaro, as well as a series of large-scale projects financed by other international stream neighbours whose existing resource donors. The political repercussions of charging smallholders in order to limit their water use so that electricity could be generated for consumption by industry and Box 6.4 Co-operation in shared river basins urban households soon manifested themselves. By 1994 resistance to the board was evident in local opposition to the tariffs. As it turns out smallholder As shown in Chapter 1, a significant proportion of the irrigation by the Chagga is a well-studied example of a centuries-old traditional world’s rivers cross international boundaries. In system for the management of common property – replete with a local water addition to these international basins, there are many management ‘board.’ This Council of Furrow Elders is formed by elders of a others that cross provincial or state boundaries within specialised clan – the Wakomfongo – that plan and direct the construction of a country where these states have a mandate to furrows as well as co-ordinate water distribution and maintenance of the furrows. manage water resources. Examples include India, Along with elders of other clans the Council administers the furrow system and Australia and the United States. resolves any problems that arise. Co-operation between riparian states is not new. The potential for negative effects on the food security of traditional farmers as a Since AD 805 approximately 3 600 water related result of the tariffs imposed by the official water board was acknowledged in treaties were signed between nations. Although the further studies in 1995 by the original authors of the feasibility study. Yet the majority of these relate to navigation and national plan to transfer political control over water from local to centralised authority boundaries, approximately 300 are non-navigational went ahead, laying the foundation for future struggles between the local people and cover issues related to water quantity, water and the Tanzanian utility that operates the dam. The water board, which must quality and hydropower. Of these, many are limited to mediate in this regard, is constituted by five government representatives and relatively narrow aspects and do not extend principles three representatives from each of the regions traversed by the Pangani river. for integrated resource management throughout the No provisions were made for representation of the Chagga or other traditional basin. As pressure on resource use intensifies, an water users on this board. The result is that Nordic development assistance had increase in conflicts over water may be expected and the paradoxical effect of undermining local resource management. greater co-operation will be required. Source: Mung’ong’o, 1997; Usher, 1997a, eco026, WCD Submission Source: WCD Thematic Review V.3 River Basins

174 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

base may be affected by water resource This situation leaves a number of key development upstream may hold the devel- international rivers lacking a basin-wide opment plans of upstream States in check. agreement that defines a process for estab- This has been the case historically, for lishing equitable water use and therefore example, with Egypt and Ethiopia.13 with no framework for good faith negotiations with other riparian States. In the In many cases, one of the key obstacles to absence of such agreements some States reaching international agreements lay in have taken unilateral action, continuing to looking at water as a finite commodity and build dams without adequate information trying to allocate it on a proportional basis exchange or consideration for impacts to different uses in different countries. In elsewhere in the basin. While this may cases of water scarcity, this approach often constitute disregard for emerging interna- does not give the flexibility needed to meet tional practice and the standards governing the multiple claims along the river course. peaceful relations between riparians, it also In these circumstances it has proved helpful reflects the political economy of the up- to extend sharing agreements to include the stream-downstream relationship. As long as benefits generated by the water. The divi- the political and economic costs of engaging sion of benefits under the 1968 Columbia in such behaviour are small relative to the Treaty between Canada and the United economic benefits gained there is little States on the Columbia River reported by incentive to engage in collective discussion. the Grand Coulee Case Study is a case in Clearly, as the demand for water rises and point. becomes ever more scarce, dams built on these international rivers are likely to Arrangements for water sharing at provin- increasingly affect regional relations. cial level are facilitated by the ability of the federal government to impose overarching Planning and Evaluation regulatory frameworks, financial incentives and sanctions to ensure that provinces In general project planning and evaluation collaborate. A similar supra-national body is for large dams has been confined primarily often lacking between nations and the most to technical parameters and the narrow transparent decision-making on internation- application of economic cost-benefit analy- al rivers therefore lies within the frameworks ses. Decisions of this nature were typically of the many international protocols and taken with little participation or transparen- agreements that clearly lay out the planning cy. In particular, those to be negatively stages at which information should be affected by a dam were (and are) rarely exchanged and consultation occur. Efforts to involved in this process. establish accepted international principles have been negotiated through the UN for The primary concern with planning process- over 25 years, leading eventually to the UN es is that once a proposed dam project has Convention on the Law of Non-Navigable survived preliminary technical and econom- Use of International Watercourses. However ic feasibility tests and attracted interest from it looks unlikely that the Convention will financing agencies and political interests, enter into force due to the reluctance of the momentum behind the project and the States to ratify it. need to meet the expectations raised often

The Report of the World Commission on Dams 175 Chapter 6

prevail over further assess- dam projects are linked to cases where A number of key ments. Environmental and affected people played no role in the plan- international rivers lack a social concerns are often ning process, or even in selecting the place basin-wide agreement ignored and the role of impact or terms of their resettlement. In addition, that defines a process for assessments in selecting op- governments have frequently committed establishing equitable tions remains marginal. Once themselves unquestioningly to large infra- water use between operations have been initiated structure projects, whose merits have not riparian States. there is a generalised lack of been tested by public scrutiny, without effort to monitor, assess and hearing alternative views on the choice of respond to operational concerns and chang- development objectives for a village, region ing values surrounding dams. Again, the or country. As pointed out in Chapter 4, the political economy of large dams and the involvement of displaced people has the dominant power of a small number of actors advantage of enabling them to contribute to often drive these planning and evaluation the benefit stream of a project and thus to processes. In some cases, such as Norway, achieve different outcomes. Quebec, Brazil or Nepal, a high level political choice made in favour of hydropower has Participation and transparency in decision- driven subsequent choice of technology (large making processes involving large dams – dams) and project development. again like most development projects – was neither open nor inclusive through the Participation and transparency 1980s. Of the 34 dams in the Cross-Check Survey that involved resettlement of dis- The WCD Knowledge Base shows that the placed people, only 7 required participation most unsatisfactory social outcomes of past as part of the decision-making process. While there has been a growing emphasis Figure 6.2 Trends in provisions for participation and information on transparency and participation in deci- disclosure sion-making involving large dams, especially in the 1990s, actual change in practice

Global sub-sample: 105 dams remains slow. 70 60 Additional results from the Cross-Check 50 Survey illustrate that while participation has 40 increasingly been required in the planning % of dams 30 documents of large dams and for various 20 activities, around 50% of projects still do 10 not plan for the public participation of 0 affected people. The trend for requirements for transparency through information 1950 1950s 1960s 1970s 1980s 1990 disclosure for large dam projects is similar to Year from start of commercial operation (by decade) that for public participation (see Figure 6.2).

Information disclosure Participation of affected people The Commission’s review identified the following recurring concerns and criticisms about how the public, and particularly Source: WCD Cross-Check Survey. affected people, have been involved:14

176 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

■ insufficient time, resources and informa- Box 6.5 Even late participation leads to a consensus resettlement tion have been made available for public plan: Salto Caixas dam, Brazil

consultations; The 1 240 MW Salto Caixas hydropower project in Brazil was completed in 1999 ■ the spectrum of participants was usually and was the fifth to be built on the Iguacu River. It was the first hydropower project in Brazil planned under the environmental regulations stipulated in the very narrow, ignoring rural communities, 1988 Constitution. The measures taken to comply with the new environmental indigenous groups and women, and legislation at Salto Caixas amounted to about one-quarter of the total project cost of approximately $1 billion. However, the EIA was only undertaken after the affected people’s organisations whose project was approved and land had been acquired, putting political pressure on effective participation may be constrained the process. This resulted in an EIA study of ‘poor quality.’ both culturally and linguistically; On the social side, public pressure, based in part on unsatisfactory resettlement outcomes for previous dams, led to the establishment of a ‘resettlement ■ where opportunities for participation of committee’. This committee created a forum to address conflicts and meet affected peoples and NGOs representing requests of the affected people. A negotiation process involved the committee of affected people in developing an acceptable relocation programme for affected groups have been provided, they displaced people. Views of local people were also incorporated into the often occur late in the process and are monitoring stages of the project.

limited in scope. Moreover where Source: dos Santos, 1999, p153–154; substantial differences arise, those Verocai, 1999, Contributing Paper for WCD Thematic Review V.2 Environmental and Social Assessment, p7 seeking to modify plans and decisions often must resort to legal or other action outside the normal planning process; ■ there was a generalised failure to involve affected people in the design and imple- Box 6.6 Public participation and project acceptance: three scenarios mentation of project monitoring and from Austria follow-up; and In the early 1980s, nine multi-purpose dams had already been built in the Austrian section of the Danube River. Two more dams, Freudenau in Vienna, and ■ the government agency staff leading the Hainburg downstream, were planned on the main river and some low-head discussions had often been trained only projects envisaged on the Mur river. in one sector (such as engineering) and The decision to build the Hainburg hydropower and navigation dam was made explicit in 1983. Public participation was restricted to holders of property and this reduced the scope for promoting a water rights to be directly affected by the planned intervention, thus excluding multi-disciplinary approach. environmental activist groups and other civil society organisations. With strong support from the general public, these civil society groups occupied the site of the project, and ultimately managed to stop the project. Subsequently the site From the experiences recorded in the WCD was protected as a National Park. Knowledge Base there are recent examples Turning to the Freudenau hydropower dam, from 1986 to 1988, the provincial government of Vienna and the power utility promoted ideas and proposals for that show where participation has reduced mitigating some of the potential impacts of the project (during construction and conflict and made outcomes more publicly after completion). The public responded with great interest and as a result selected proposals and the detailed project plans were made fully accessible to acceptable (see Box 6.5 and Box 6.6). These the public. Information meetings on these documents were attended by more contrast starkly with those where projects than 15 000 people. In 1991 a referendum was organised around the final project proposal. About 44% of the entitled Vienna inhabitants participated, and 75% have been pushed through by central authori- supported the project. ties without consultation resulting in drawn- The Freudenau dam was completed in 1997, however the project is not cost- out and acrimonious conflicts over compensa- effective as a consequence of the design modifications required to gain public support. The contrary was the case for the Fisching and Friesach projects on the tion, resettlement and benefit sharing. Mur river – where following occupation of the dam site by protesters, joint- planning sessions with stakeholders led to ‘significant improvements in environmental friendliness,’ with associated cost savings. The failure to provide a transparent process Source: Hainburg and Freudenau dams that includes effective participation has in Nachtnebel, 2000, p109-111; prevented affected people from playing an Zinke, 1999, p6-9; Fisching and Friesach dams active role in debating the project and its in Brunold and Kratochwill, 1999, p176-17 alternatives. As a result they are unable to

The Report of the World Commission on Dams 177 Chapter 6

assist project planners to provide a develop- and decision-making approaches of the day. ment response that meets their needs and Many sectoral planning studies from which allows them to add to the benefits to be projects emerged were narrow technical and derived from the project. Without doubt this economic studies, aimed at least-cost supply has magnified the negative impacts of such solutions for providing a single service such projects and alienated affected communities as irrigation water or electric power. When leading to active opposition to projects and dams were contrasted with alternatives, they considerable uncertainty for project propo- were typically only compared to other nents. As observed in previous chapters of this potential dam projects or, in the case of report, the outcome is often not only poor hydropower, with alternative large-scale performance of the social components of thermal power generation options (see Table projects but also schedule delays, cost overruns 6.1). In developing countries the pressure on and poor financial and economic performance. development aid agencies to move large amounts of capital – a considerable portion Options assessment of it as tied aid – argued for large-scale The range, scale and type of options consid- solutions such as large dams. Administrative ered in development plans in the past were efficiency is a related factor leading to a limited by the boundaries of the planning preference for financing large projects.

Table 6.1 WCD Case Studies: options assessment Project Alternatives considered in preliminary planning Comparison with Criteria and parameters used for selection alternatives at appraisal Aslantas Existing run of river source for irrigation was considered The hydropower Least-cost analysis for power supply. insufficient to support wider agriculture development. component was compared Alternative dam locations in the basin considered. with a thermal alternative. Grand Coulee The agreed objective was regional development through irrigated agriculture. Economic analysis Alternatives considered over a 15-year period related to gravity and pumped systems for the delivery of water. The 1932 Butler Report recommended the pumped option financed predominantly by hydropower revenues from the dam. Glomma and Government policy for hydropower established in early Alternative hydropower Least cost and environmental and social Laagen Basin 1900s. Oil and gas discovered in early 1980s, but sites considered. ranking of sites in the 1980s protection and hydropower policy remained in place until recently. development plans. Kariba Kafue Gorge in North Rhodesia considered as an A set of thermal power Least-cost analysis for power and influence alternative but rejected after protracted debate with alternatives of South Rhodesia’ s political interest. Southern Rhodesia. Orange River Reservoir storage was seen as the only way to achieve more reliable water supply for Political year round irrigation. Hydropower was a secondary benefit therefore thermal alternatives were not considered. Subsequent studies led to raising the VanderKloof dam to increase output. Pak Mun Thermal alternatives located elsewhere considered. A Alternative thermal Least cost analysis and power system higher dam option was rejected due to resettlement and power (gas turbine). studies. environmental concerns. The revised project approved by the Cabinet included irrigation benefits. Tarbela Additional storage was considered the only option for A number of alternative Economic comparison of the sites combined replacing water of the eastern rivers allocated to India. dams sites including with Government preference for larger Kalabagh and Gariala storage and power potential of Tarbela Tucurui Tucurui responded to objectives for development of the mining-metallurgical sector and to supply power to urban areas in the Amazon Region. There was no explicit options assessment Source: WCD Case Studies.

178 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

Projections of demand many circumstances this has militated The needs for power, food and water are against a gradual approach of adopting typically identified through sectoral demand smaller, non-structural options and has forecasts, which have frequently overstated pushed decision-makers into adopting large- sectoral needs. The WCD Glomma and scale dam projects because they seem to be Laagen Case Study reports that in Norway the only adequate response to the large gap gross power consumption in 1990 was 75% between existing supply and forecast de- of that forecast in 1970. In Slovakia, water mand. A further complication is the long- supply needs as assessed in 1985 were lead time of large dam projects, which may expected to rise to 408 and 465 litres per take 10 years or more from initial develop- capita per day (lcd) in 1990 and 2000 ment of a project idea to the commissioning respectively. While demand initially rose of the structure. Changes in market condi- above projections (433 in 1990) it has since tions during construction have left propo- reversed course and had fallen to 294 lcd by nents stranded with costs or projects that 1997.15 are not financially or economically viable. Of principal concern is that it is frequently Failure to adequately account for the rate of the agencies that are responsible for building development of new supply and the effect of supply infrastructure that are also charged policy reform, when it is outside the limits with undertaking demand forecasts, leading of the planning exercise, may also lead to to a potential conflict of interest. what effectively amounts to overstated demand. In the Slovakia case cited above, a Available options doubling in the price of water and the As shown in Chapter 5 there is currently a development of alternative sources of supply wide range of alternatives available for by industry contributed to reducing actual fulfilling water and energy needs, although demand. Projections for demand (and the actual number available will depend on hence prices) for crops and other agricultur- local circumstances. The number of alterna- al products that are widely traded can be tives has, however, not always been so large. subject to market boom and bust cycles when For instance, alternatives to hydropower a series of independent decisions in different prior to the 1950s included conventional countries or provinces lead to over-produc- fossil fuel and biomass generation options. tion relative to demand. As documented in Nuclear power arrived in the 1960s and in Chapter 2 the prices for agricultural com- the past decade the range and scale of modities have fallen over time, instead of renewable electricity supply options has remaining constant or rising as assumed in dramatically expanded. Alternatives for many projections for irrigation projects. The municipal and industrial water supply have same market information, technical assist- tended to be site specific and depend on ance packages or consultants frequently whether there are groundwater aquifers, influence these planning exercises, thus, natural lakes and rivers to draw from with while difficult to foresee, such over-esti- sufficient quantity and quality of water. On mates are not inevitable. the other hand, many of the irrigation water supply and flood management options that Overstating future demand has led to a are being considered today have been perceived need for a large incremental available for a long time. The principal response to meet rapidly growing needs. In change here is a more receptive policy

The Report of the World Commission on Dams 179 Chapter 6

context and the increasing cost of develop- competitive solutions. Interests promoting ing new water supplies. non-structural alternatives have rarely offered an adequate political counterbalance Demand-side management options (DSM) to interests promoting a dam option. In are a more recent phenomenon. Efficiency many cases the weight given to the infra- and conservation became concepts in policy structure option by the key actors obstructed and planning in the 1970s and 1980s, but proper consideration of other viable alterna- serious attention to demand management tives. As a result, such options continue to programmes has tended to depend on a be viewed as secondary to large projects. perception of crisis. The oil price shocks in China provides an example of a country that the early 1970s focused attention on DSM has mixed both the small- and the large-scale. in the electricity sector in many western It has the world’s largest programme for the countries. Water scarcity and the threat of development of small-scale rural and appropri- water shortages is a driving force for more ate technology, while at the same time it has efficient water use in many countries, but built half the world’s large dams. the response has not been universally translated into concrete action in fostering The hindsight provided by Chapters 2, 3 water-efficient practices. and 4, however, does suggest that in cases where dams have failed to deliver or led to Obstacles to consideration of options large negative social and environmental impacts, a more comprehensive assessment Political economy or intellectual barriers of options may have been warranted. often pre-determined what options were Whether failures to adequately assess all considered in a given context. Political options implicitly lead to the selection of a economy barriers include efforts made by dam over other equally attractive or even groups, primarily those holding economic superior options is difficult, if not impossible, control and political influence, to protect to answer. Certainly, the options currently their own interests and to impede similar available (as described in Chapter 5) reflect efforts by other stakeholders. In practice, not only continued technological develop- these barriers were immensely varied and ment over the last 50 years, but also are a wide-ranging. They included soft and subtle product of more recent efforts to find locally actions such as withholding information appropriate, small-scale solutions that have necessary for making informed decisions benign social and environmental impacts. from other stakeholders and from the Thus, in many cases alternatives may not have decision-makers. At the other end of the been available previously and may have spectrum there have been overt and even appeared more expensive given the methodol- violent measures such as use of State and ogies employed at the time, or were excluded police power to protect favoured options. due to the influence of vested interests. With few exceptions, an inclusive institutional and policy structure capable of Parameters for project 16 dealing with a spectrum of options has been appraisal slow to emerge in developing countries. Cost-benefit analysis (CBA) emerged Small-scale infrastructure alternatives often between the 1950s and 1970s as the domi- have not received integrated planning nant economic tool supporting decision- support, impeding their ability to emerge as making on dam projects. Initially it was

180 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

limited to a number of parameters, most of would be to lower the economic profitability them internal to the dam owner and rela- of dams. A number of the weaknesses of tively easy to assign values to. Efforts in the CBA may lead to understatements of the last two decades to expand the scope of CBA net project benefits. At the same time, it is to cover social and environmental issues have clear that quite a number of the weaknesses rarely led to comprehensive social and envi- can have important impacts in terms of ronmental valuation, and have usually been lowering net project benefits. Improved limited to incorporating the costs of resettle- application of CBA would assist in identifying ment and environmental mitigation. projects that are not economically viable.

Review of multilateral bank appraisals and Over reliance on CBA and the implicit the performance of CBA more generally pursuit of economic welfare maximisation leads to the following conclusions on the also handicap decision-making where dams adequacy of CBA as applied to the appraisal have other (or additional) objectives as: of large dam projects: ■ CBA does not examine wider economic ■ projections of project costs are systemati- impacts – such as economic multiplier cally understated; impacts; and ■ social and environmental impacts are ■ CBA does not explicitly identify who not valued explicitly or are only indirect- gains and who loses from a project. ly accounted for through mitigation or resettlement budgets; Although CBA is typically a prerequisite to ■ difficulty in predicting inter-annual the analysis of macroeconomic and regional volatility of hydrological flows, growth in impacts, as well as to distributional analysis, demand and final design capacity (hydro- it is not designed to examine the potential power, irrigation and other benefits); of a project to achieve objectives in these ■ difficulty in predicting market conditions areas. Given the continued ‘partial’ ability and farmer behaviour over time (irriga- of CBA to capture even the extent to which tion benefits); efficiency objectives are achieved, and given that equity, macroeconomic and purely non- ■ employing social discount rates that are economic objectives are often integral too high; objectives of water resource development ■ sensitivity and risk analysis is inade- projects, CBA alone is not a sufficient basis quate; and for the evaluation of large dam projects. ■ the effect of uncertainty and irreversibility of investment is ignored. Influences from the larger political economy also filter through into the process of In other words, the historical and actual undertaking CBA. In some cases, early practice of dam project appraisal often political or institutional commitment to a violates the conditions under which it project became overriding factors, leading could, in theory, provide a reliable measure subsequent economic analyses to justify a of the change in economic welfare produced decision that had in fact already been taken. by a dam project. It is worth emphasising that it is not a foregone conclusion that the Decisions made to build dams solely on the net effect of fixing all of these problems basis of such an analysis are questionable

The Report of the World Commission on Dams 181 Chapter 6

given the failure to undertake Similarly, the adverse social implications of options assessment and to large dam projects have rarely been a factor include external impacts, in the initial assessment and therefore have particularly social and envi- not generally influenced the decision- ronmental costs. An alterna- making process to reach a least social cost tive approach to a decision alternative. The experiences of affected support system based on CBA people around the world as reviewed in is to use a method that Chapter 4 confirm the extent to which recognises that projects often impacts remain inadequately assessed and have multiple objectives and efforts at mitigation, development and not simply economic welfare maximisation. resettlement unsatisfactory. Experience to date with these multi-criteria approaches suggest that while economic Following the United Nations Conference criteria remain important, these decision on the Human Environment held in Stock- frameworks have the benefit of allowing holm in 1972, environmental agencies and disaggregated information on social and ministries were formed at a rapid rate with environmental impacts to enter directly into approximately 60 being created by 1988 and the decision analysis. Such decision support at least another 40 by 1992. The World systems appear particularly appropriate and Bank adopted its first dam-related policy in useful in the case of large dams when 1977 (on dam safety). During the 1980s the implemented within a participatory, trans- Bank developed policies and guidelines that parent multi-stakeholder approach. focused on the social and environmental dimensions of dams and water resources. Addressing social and environmental impacts17 Environmental Impact Assessment (EIA) was adopted and formalised in many coun- Social and environmental issues have tries during the 1980s, although many historically been among the least addressed developing countries only approved EIA concerns in dam-related decision-making. legislation in the 1990s. EIA has become The Commission has focussed on these the major tool for addressing social and because they are two of the key issues that environmental impacts and the Commission determine whether a dam proves to be an has reviewed an extensive literature on this effective development project that enjoys subject as well as hearing directly from those general acceptance by the public. The affected through the regional consultations. environmental risks associated with large dam projects have not been generally The WCD Knowledge Base demonstrates incorporated as key factors in the decision- that EIA consists mostly of measures to making process. Enforcement of existing compensate or mitigate the planned impacts regulations is often weak, initial assessment and render them acceptable when the has not been comprehensive and it has decision to proceed has already been taken. frequently been incorrectly assumed that This is reflected in the tendency for EIAs in impacts could be effectively mitigated (see the 1990s to focus increasingly on mitiga- Chapter 3). Generally, monitoring of tion plans. Added to this is the fragility of impacts and assessments of the effectiveness newly established environment ministries of environmental mitigation measures have that may be unable to ensure compliance been absent. with many of the plans or clearance condi-

182 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

tions.18 There are well-documented cases, planning mode, based on voluminous even in the 1990s, of decisions to proceed assessments and reports, to an implementa- with financing or construction before an tion mode during project construction effective EIA is completed (see Box 6.7). creates severe institutional and human resource challenges and in many cases the Political pressures and tight schedules are as measures are either not implemented or relevant today as in earlier decades and EIA have fallen short of the efficacy envisaged in results often have no significant influence the planning documents. The reality is that on the choice of a dam as the preferred dams create huge management challenges option. The EIA process is also not well for the implementing ministries and agen- suited to this purpose as it was meant solely cies. Where institutional capacity in the for identifying impacts and associated environmental area is weak the accompany- mitigation measures rather than as a tool for ing measures needed for sustainable out- including environmental and social consid- comes often prove difficult to manage, erations in the final project choice and particularly when compared with the design. Many governments and financing physical act of designing and building the institutions have adopted EIA in the last dam. This in turn may lead to public dissat- two decades, however the quality of assess- isfaction with dams when affected people ments and their ability to genuinely influ- perceive that promises have not been kept. ence outcomes is still under-developed. Most dam proponents see EIA as an admin- Operation, monitoring and istrative hurdle to be cleared, or a require- decommissioning ment to secure funding. This means that a huge political, technical and financial After large dams are commissioned there are investment in the project has often already a number of management and operational been made before the EIA is launched. If issues that require technical studies and impacts are severe, it is often too late to involve either decisions at the management change design, and project cancellation may level or decision processes that are public in involve loss of face and financial loss. Further, EIA operates under considerable constraints due to the political and adminis- Box 6.7 Environmental Impact Analysis (EIA): too little, too late trative pressures imposed by project sched- Even with improved environmental and social guidelines EIA still frequently fails ules as it is seen as ‘delaying’ the project. to influence decision-making. The Theun Hinboun project in Laos was initiated in the early 1990s. The initial EIA financed by NORAD concluded that the dam EIAs are also often done with inadequate would have minimal adverse impacts and significant benefits. Most of those who baseline data on demographic trends, socio- reviewed the document disputed these findings and NORAD undertook supplementary studies. These were completed one year after construction cultural systems and ecosystem functioning. began, so they had no impact on the decision making process or the design of This leads to unsatisfactory outcomes. the dam. In the WCD case studies, an EIA was conducted only for Pak Mun at the planning stage as it was a World Bank requirement. However the EIA was done As an impact management tool, EIA has ten years before the final project was approved – and examined a different evolved towards a tool for also setting up an project design for a different location than the one finally approved. Further, the EIA was never revised or updated. EIA’s were only required in Thailand from ongoing environmental management system 1992, one year after Pak Mun was approved by the Thai Government. or programme when construction begins, Source: Theun Hinboun dam in Norpower, 1993, p1-7 as cited in Usher and Ryder, 1997 eco026, p80-81, WCD Submission; WCD Pak Mun Case Study involving appropriate experts, ministries and field activities. The transition from a

The Report of the World Commission on Dams 183 Chapter 6

nature. These are required for the following WCD Case Studies suggest that provisions reasons: for intensive monitoring of physical, social ■ to support routine day-to-day operations and environmental effects of projects were such as reservoir operations and releases; often weak or entirely absent. Where monitoring was present, it was often restrict- ■ to change operations to conform to new ed to hydrology and engineering parameters regulations that are introduced such as related solely to the physical integrity of the economic, technical, environmental or dam structure. social regulation on dam safety, operation of reservoirs in flood events or The operation of large dams is subject to changes to environmental flow require- many unforeseen and unforeseeable influ- ments; ences over time that transform and redistrib- ■ to change operations when a new dam is ute benefits and impacts. Patterns for the introduced in the river basin that would release of water from reservoirs will normally impact on the operating rules of the change over time in response to demograph- existing dam; ic and land use changes in the river basin, ■ to adapt the operation to changing needs shifts in water use priorities, as well as in the services provided by a dam over changes in the agricultural economy and the time, particularly when the project is markets for electricity. Physical changes in multi-purpose, for example a change in river morphology or reservoir sedimentation as well as changes in the value that society power markets that increases the value of places on ecological and social impacts of peaking power generation or a shift to dams will influence how the dam is operated recreation priorities in controlling at different periods of time over its life. reservoir water levels;

■ for renovation, upgrading or expansion As shown in Chapter 2, there are good of the existing facilities; and practice examples of adaptive management ■ for relicensing processes in some coun- to meet this changing context, drawing on tries, or for decommissioning. sophisticated decision-support and forecasting software and in some cases accommodat- One of the most disturbing findings of ing stakeholder participation. Many devel- Chapter 2 was the lack of monitoring of the oping countries that continue to focus on impacts of dams and the complete failure to building rather than optimising operations conduct proper ex-post evaluations of have not yet adopted tools and policies for performance and impacts. That such large adaptive management and optimisation. investments have rarely been evaluated once they have been in operation for a What happens to dams at the end of their significant period suggests little obligation lives? Dam decommissioning may be neces- on the part of powerful centralised agencies sary due to safety concerns, dam owners’ and donors to account for the costs and concerns about lower profits, or concerns benefits incurred. Perhaps more critically it about social and environmental impacts. signals a failure to actively engage in learn- Decommissioning can mean actions ranging ing from experience in both the adaptive from stopping electricity production to dam management of existing facilities and in the removal and river restoration. Several design and appraisal of new dams. The hundred dams have been deliberately

184 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

removed, mostly in the United States, and Reasons for this include: most of them small. Provision is not always ■ incompleteness, incoherence and ambi- made in advance for who should pay for the guity of national legal and regulatory removal or for safety or other improvement frameworks; measures. As reviewed in Chapter 2, rising ■ difficulties of accurately defining the operations and maintenance costs may also specification of social and environmental raise questions regarding the maintenance requirements and integration of these and safety of large dams. At present, decom- components into the implementation missioning costs are difficult to predict due agreements and schedules of projects; to the uncertainty surrounding the various ■ lack of transparency and accountability, parameters affecting the costs and the frequently with opportunities for corrup- limited amount of practical experience with tion at key points in the decision-making decomissioning. One proposal is to ensure process; that decommissioning funds are set aside at ■ lack of meaningful participation at key the time of dam commissioning or during points in the decision-making process the project’s license period. Such decommissioning funds are accepted practice for ■ low levels of internal and external nuclear power plants in countries such as monitoring that reduce feedback into the United States. Decision-makers in the decision-making; developed world are increasingly looking at ■ weak or non-existent legal recourse and how best to handle the end of the dam life appeals mechanisms to an independent cycle. In contrast many other countries do judiciary, particularly for negatively not yet have firm licensing periods for their affected and vulnerable groups; and dams (see Box 6.8).

Box 6.8 Licensing processes and duration Compliance19 There is considerable variation in the licensing procedures for dams. In some Dam projects are expected to comply with countries dam sponsors must obtain only one licence. In other countries the dam sponsors must obtain a licence for each phase in the planning and project the legal framework and guidelines of the cycle. For example in Hungary a dam project sponsor must first seek approval of country and the organisations involved in the EIA, then obtain a permit to complete the activities required to prepare the project for construction. Two further licenses are required for construction and financing and constructing the dam. Where operation. Some countries exempt government operators and only require environmental and social problems have licences for private operators. Licensing procedures are often restricted to hydropower dams, with irrigation dams largely exempt from formal licensing. occurred in the WCD Knowledge Base, the There is considerable variation in the term of dam operating licences. In Spain principle cause is the lack of legal require- licences are granted for 70 years, in Norway 60 years and in the USA for 30–50 ments for particular standards at the outset years. In other cases dam licences are granted for short renewable periods of time. For example in Hungary and Vietnam licences can be granted for or a lack of appropriate recourse mecha- unspecified periods of time but they are subject to regular inspection and nisms to adequately reflect people’s rights in review. Where dams are built and operated by the private sector the duration of the licence period will need to reflect a reasonable payback period, typically set the face of a powerful national decision. at 30 years in build, operate and transfer (BOT) agreements. Reviews conducted This section shows that regulatory frame- within the licence period have the advantage of facilitating the monitoring of operations and providing opportunities for adapting operations against the works are often weak, and the necessary background of changing societal values and expectations. Contemporary provisions are not made in planning docu- concepts of adaptive management, transparency and accountability suggest that there should be some scope for regular review, such as every five to ten ments. Even when they are present, govern- years. ments and donors alike ignore them all too Source: WCD Thematic Review V.4 Regulation frequently.

The Report of the World Commission on Dams 185 Chapter 6

■ lack of human, financial and organisa- performance that could feed back to better tional capacity. inform decision-making based on past experience. Nor do they often provide National legal frameworks and recourse for those who may have been policy provisions harmed by a particular project. There were few policy, legal and regulatory In many cases it has only been strong frameworks governing large dam building concerted civil society movements that before the 1970s, particularly for social and have generated sufficient momentum to environmental issues. However many ensure that constructive negotiations occur, countries updated their policy and regulato- and dam projects are not imposed on ry frameworks in the 1980s and 1990s to displaced communities without consulta- give a stronger emphasis to environmental tion. and social concerns, public participation, efficiency and cost-recovery. There is now a The Cross-Check Survey demonstrates that, broad body of regulation potentially or since the 1950s, a growing number of explicitly applicable to large dams at the projects have required dam safety, economic international and national levels, referring cost-benefit analyses and financial plans to both the public and private sectors. (see Figure 6.3). But economic appraisal techniques such as risk and distributional Existing regulations in most countries tend analysis were still mandated for only 20% of to focus on project appraisal and implemen- large dam projects even in the 1990s. Sensitiv- tation with insufficient focus on options ity analysis is more common and has become assessment planning in the early stages of standard for donor-financed projects. Dam the decision-making process where funda- safety is a key issue for the world’s ageing dams mental choices are made. Few require and many national bodies have taken up the regular assessments and evaluation of challenge, assisted by the engineering networks of ICOLD. Figure 6.3 Trends in the implementation of economic and financial analyses Even requirements for large dam Global sub-sample: 105 dams projects in the environmental 100 field are far from universal

80 although they are increasingly required. EIA is recorded for less Cost benefit analysis 60 Financial analysis than 40% of dams commissioned Distribution analysis 40 in the 1990s (see Figure 6.4). % of dams Risk analysis Strategic environmental assess- Sensitivity analysis 20 ments and baseline surveys occur

0 at similar levels.

1950s 1960s 1970s 1980s e 1950 after 1990 Corruption befor Year from start of commercial operation (by decade) Corruption is a world-wide phenomenon that affects both poor and rich countries. It may Source: WCD Cross-Check Survey.

186 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

take many forms, from inducements to Box 6.9 Allegations of corruption favour certain contractors during bidding, through to manipulation of water alloca- In early 2000 the Chinese government released information that corrupt officials had embezzled $60 million (500 million yuan) from resettlement funds for the tions, offsetting farmer repayments, or Three Gorges dam project. An official was sentenced to death for embezzling manipulating domestic electricity connec- almost $1.5 million from the project. tions locally.20 At whatever level, vested In Lesotho a trial started in June 2000 against major international corporations involved in construction on the Lesotho Highlands Water Project (LHWP). interests can distort the decision-making Companies from France, Sweden, Germany, the United Kingdom and Canada process, undermining development. Deci- have been accused of paying bribes. If the accused companies are convicted, they face debarment from future projects with the European Union. sion-makers may be inclined to favour large In the United States, economists from the United States Corps of Engineers infrastructure as they provide opportunities accused senior management of deliberately manipulating economic analyses to for personal enrichment not afforded by promote billion dollar investments to be managed by the Corps. smaller or more diffuse alternatives. The Source: China in Agence France Presse, 21 January 2000, 10 March 2000; LHWP in Sunday Independent, 11 June 2000; consequences frequently directly affect the United States in Grunwald, 29 February 2000 poor or the environment. Allegations of corruption have tainted many large dam projects in the past but have seldom resulted in prosecution in court (see Box 6.9). transparency processes are therefore increasingly available for ensuring that dams are built The OECD countries, and the major inter- for societal good, not for personal gain. national financing agencies, have recognised the pervasive extent of corrupt prac- Multilateral and bilateral tice and its negative consequences. Through financing agencies the 1990s they have moved to assist countries in tackling corruption by making Overseas development financing agencies, bribery payments illegal in their country of particularly the multilateral and bilateral origin, debarring contractors convicted of bribery from future contracts and tightening up due diligence on bribery opportunities.21 Figure 6.4 Trends in the implementation of environmental and social assessments As of August 2000, twenty-three countries had ratified the 1997 OECD Convention on

Combating Bribery of Foreign Public Global sub-sample: 105 dams Officials in International Business Transac- 60 tions. Its principal objective is to eliminate 50 Environmental impact assessments bribes to foreign officials, with each country Social impact assessments 40 taking responsibility for the activities of its companies and what happens in its own 30 territories.22 % of dams 20

10 Transparency International, an international 0 NGO, has also been active in promoting

1950s 1960s 1970s 1980s workable and transparent, ‘integrity pacts’ for e 1950 after 1990 large infrastructure tenders. These have met befor Year from start of commercial operation (by decade) with growing acceptance and success in Latin America. A range of legal measures and Source : WCD Cross-Check Survey.

The Report of the World Commission on Dams 187 Chapter 6

agencies have played an impor- eventualities. Yet the WCD Knowledge The policies of the tant role in funding and secur- Base shows that achieving satisfactory social development banks are ing large dam projects. They and environmental outcomes requires more concerned with have adopted a broad set of constant adaptive management. The short project planning, design policies, criteria and guidelines term and inflexible nature of the agreement and financial since the 1980s as a result of between the borrower and the bank is an management than with lessons learned from experience obstacle to achieving this result. Further, options assessment or and public criticism. For exam- the mitigation measures often receive less with the operational ple, the World Bank has adopt- prominence in comparison with finance phase of a large dam ed ten safeguard policies relating issues. project. to such environmental issues as forestry, pest control and envi- Numerous developed countries have bilater- ronmental assessments; and such social al aid agencies and export credit agencies issues as indigenous people, cultural property which have also funded or supported the and resettlement. The result of these devel- financing of dams and dam-related 23 opments is that on paper the World Bank projects. Bilateral aid agencies vary in the has a comprehensive set of policies dealing stringency of the requirements they have for supporting large dam projects. Yet while with large dam projects. More recently the they are relatively small participants in the International Finance Corporation (IFC) and large dam sector, their funding for specific the Inter-American, Asian and African aspects of the master planning or project Development Banks have adopted similar feasibility studies can be critical in bringing guidelines. other financiers to the table. Despite these changes, the banks’ policies, Export credit agencies (ECAs) are increas- like the national regulatory systems are more ingly financing specific portions of large- concerned with project planning, design and scale infrastructure projects in developing financial management than with options countries. ECAs provide loans, guarantees assessment or with the operational phase of and insurance to domestic corporations and a large dam project. In addition, they have businesses for their activities overseas to paid more attention to monitoring the support and promote export trade from their planning and construction phases than the respective countries. They finance the high operation of the project, which is often left value, electrical and mechanical equipment to national governments. Post-implementa- components and are an increasingly impor- tion monitoring is generally discontinued at tant source of financing for private sector most five years after project commissioning. involvement in large dams. Even then, the main focus has been on comparing the project proposals with the Unlike the major development financing project outcome. Weak treatment of social agencies, ECAs generally lack policies on and environmental impacts at appraisal environmental and social issues and do not leads to weak assessments of outcomes at necessarily adhere to internationally accept- evaluation. ed standards and guidelines. Experiences from the Three Gorges dam in China, Ilisu This approach assumes that the planning dam in Turkey, Maheshwar dam in India phase can anticipate and cover all future and San Roque dam in the Philippines

188 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

underline the need for ECAs to examine ■ The lack of sanctions for non-compli- closely the social and environmental im- ance, either at national or international pacts of the projects they support. The level. In many cases local affected absence of common standards among ECAs communities were unable to defend their leads to ad hoc competitive decision-making interests when faced with a strong (see Box 6.10). centralised government especially in countries with weak legal safeguards and The policies of multilateral banks have recourse mechanisms.24 challenged the capacity of their borrower ■ The dependence, in many cases, on the countries to actually implement their good faith of sovereign States and public requirements. Bank staff have had to either pressure to resolve disputes, adjudicate exercise their own discretion to adapt the claims and ensure compensation for policies to the realities of each country or those who have suffered wrongs. The ignore cases of non-compliance by their absence of legal sanction or, where this borrowers. In either case the bank’s toler- exists, difficulty in accessing it made it ance of the staff’s and the borrower’s non- easier for developers (especially govern- compliance with the policies can breed ments) to escape the consequences of cynicism about the willingness to comply. non-compliance. The costs involved in There are no sanctions for staff members, or seeking legal remedies were often prohib- countries, for non-compliance. Performance itive for those who may have been criteria for staff have tended to be related to negatively affected. approvals and disbursement targets. The multilateral banks – and in particular The WCD Knowledge Base offers many the World Bank – have the most sophisti- examples of the failure of project propo- cated set of policies, operational procedures nents, contractors and operators to fulfil and guidelines amongst the international commitments, whether explicit (project donor community and are under regular specific agreements and contracts) or scrutiny by civil society. In examining actual implicit (applicable policies, laws, regula- practice and compliance with standards and tions and guidelines). The WCD Case the realisation of the outcomes that these Studies provide an indication of the types of breaches observed (see Box 6.11). The three basic reasons for lack of compliance have Box 6.10 Export Credit Agencies: competing for business versus been: common standards ■ The tendency for large projects to After the US Export-Import Bank declined support for the Three Gorges project in China, citing lack of information on environmental and social mitigation, other proceed under a restricted decision- ECAs, with lower thresholds of social and environmental acceptability, stepped making process negotiated between forward to issue loan guarantees to corporations. This phenomenon is especially relevant to the financing of large dam projects where ECAs are supporting governments, lenders and contractors projects declined by other funding agencies on environmental grounds. In June with little public oversight, little partici- 1999, the G-8 ministerial meeting issued a statement recognising the importance of common standards among the ECAs. Later in the year the OECD pation by affected parties and limited Working party on Export Credits and Credit Guarantees agreed to a voluntary disclosure and public access to informa- environmental information exchange on larger projects but fell short of agreeing on new criteria for ECA support. tion. In many cases lack of clear moni- Source: Udall, 2000, WCD Contributing Paper to toring procedures also limited public Thematic Review V.4 Regulation, p1-3. scrutiny.

The Report of the World Commission on Dams 189 Chapter 6

imply, the WCD Knowledge Base has agencies will have encountered similar emphasised the experience of these banks. difficulties and also fallen short of the Given that the banks have often fallen short outcomes implied by the standards set by of realising such high standards for planning the banks. and decision-making, it is legitimate to expect that the other donors and in-country Findings and Lessons Conflicts over dams have heightened in the Box 6.11 WCD Case Studies: a compliance report card last two decades, as awareness of their In most of the WCD Case Studies there are examples of agreements made not impacts and performance has grown and the being respected and commitments only partially implemented: debate over costs and benefits has spread. ■ For Grand Coulee dam $54 million for past losses and $15 million per While conflict has sparked innovation in year in compensation was awarded by the courts to the Colville tribe in 1994, 50 years after the dam was built. The settlement cited high level some contexts and by some stakeholders in government correspondence indicating an initial intention to compen- the debate, in others it has deepened and sate the tribes for loss of salmon in accordance with existing treaties, this was abandoned by the late 1930s. entrenched conflict. The Global Review of ■ Construction of Kariba dam complied with the laws of the day – however large dams and their alternatives has exam- it was planned and built prior to most regulations being in place. In ined the performance of large dams using a addition, the laws under colonial rule in Southern Rhodesia (now Zimbabwe) did not include provision for just legal redress for displaced number of different lenses – technical, Africans, a clear contravention of prevailing international standards. financial, economic, environmental and ■ At Tarbela dam nearly 2 000 families had not been adequately resettled social – and explored the options that are twenty years after displacement in terms of the 1967 criteria for compensating landowners. currently available to fulfil water and energy ■ At Tucurui, the initiation of the second phase of the project in 1999 needs. proceeded without an environmental impact assessment (EIA). Eletronorte, the utility that owns the project, maintains that Phase II does not require an EIA as it is the continuation of a project approved prior to As part of its Global Review of past experi- the setting of EIA regulations in Brazil. Local communities, concerned ence, the Commission examined the deci- about the possibility of a repeat of the social and environmental impacts of Phase I of the project, disagree with this position and have asked for a sion-making, planning and compliance full EIA. The Case Study also points out that Eletronorte did not respect processes around large dams in the WCD the Waters Code which stipulated that hydropower plants should not adversely affect the food and needs of river bank communities, public Knowledge Base to better understand what health, shipping, conservation and free circulation of fish, amongst others. factors influence these processes and the ■ In Aslantas the government agreed with the World Bank to recover a performance and results of the projects. portion of the costs of the irrigation component of the scheme from farmers over 50 years. Current recovery rates are inadequate to meet this Based on this review the findings on deci- target (see Box 2.4). sion-making include: ■ At Pak Mun, an EIA was a World Bank requirement and should have been ■ performed on the revised project prior to construction. centralised and bureaucratic State ■ In India, a national assessment of dam projects cleared in the 1980s and agencies and utilities have often promot- 1990s shows that in 90% of cases the project authorities have not fulfilled ed and implemented dams as one of a the environmental conditions under which environmental clearance was given by the central government under the Environment Protection Act small number of conventional responses of 1986. to water and energy needs, a choice that, ■ In Norway, provisions for environmental flow releases from hydropower once taken, often has not been revisited dams have allegedly dropped below the minimum established in the licensing agreements. Yet the central authorities lack legal means to even in the face of an expanding list of monitor and sanction confirmed offenders. alternatives; ■ In China a review of Lingjintan dam showed that compliance with ■ foreign assistance has stimulated large environmental clauses in construction contracts was not satisfactory due, amongst other factors, to lack of incentives, lack of accountability and investments in dams in developing poor oversight. countries, by providing financing – more Source: WCD Case Studies than $4 billion per year during the peak

190 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

of lending in 1975-1984 – and leading dams were not built based on an objective financing arrangements; assessment and evaluation of the technical, ■ large developing countries with many financial and economic criteria applicable at large dams (including China, India and the time, much less the social and environ- Brazil) have established internal capacity mental criteria that apply in today’s context. to build large dams, although in recent That many of such projects have failed to decades they have often used external deliver by standards applicable in either finance and equipment to build larger context is therefore not surprising, but projects; nonetheless cause for concern. ■ countries building fewer dams have been Focussing on the planning cycle for large disproportionately influenced by foreign dams reveals a series of limitations, risks and assistance for large dams, making them outright failures in the manner in which more vulnerable to conflicts between the these facilities have been planned: interests of governments, donors and ■ industry involved in foreign assistance participation and transparency in plan- programmes and improved development ning processes for large dams was neither outcomes for rural people, particularly inclusive nor open and while actual change in practice remains slow even in the poor. the 1990s there is increasing recognition ■ the multilateral banks and bilateral aid of the importance of inclusive processes; agencies, alongside the dam-building ■ while the number of options have industry and international industry increased over time, options assessment associations, have played a key strategic was typically limited in scope due to role in spreading the technology to political and economic interests driving developing countries, lending legitimacy dam projects, lack of familiarity with to emerging dam projects, and fostering other options, the perceived need to the technological and human resources quickly proceed with large-scale projects required to build and maintain dams; to meet large projections in demand and ■ there has been a generalised failure to the relative ease of developing new include and recognise affected people supply relative to undertaking policy or and empower them to participate in institutional reform; decision making. ■ project planning and evaluation for large ■ the lack of agreements on water use dams was confined primarily to technical within shared river basins is an increas- parameters and the narrow application of ing concern and cause for conflict, economic cost/benefit analyses with particularly as demands grow and unilat- many sectoral studies aimed at finding eral decisions to build large dams by one least-cost supply solutions for providing a country alter supply within a basin with single service such as irrigation water or significant consequences for other electric power; riparian States. ■ where opportunities for the participation of affected people, and the undertaking The end result of the influence exerted by of environmental and social impact vested interests, and the conflicts of inter- assessment have been provided they ests that have arisen, has been that many often occur late in the process, are

The Report of the World Commission on Dams 191 Chapter 6

limited in scope, and even in ■ weak regulatory frameworks and lack of the 1990s their influence in sanctions at the national level, particu- project selection remains larly for options assessment and social marginal; and environmental requirements, and ■ the paucity of monitoring and little enforcement of existing regulations evaluation activity once a have contributed to the poor economic, large dam is built has reduced social and environmental performance of the basis for learning from many large dams; experience; and ■ large projects tend to lack public over- ■ while countries that were the sight of negotiations between govern- first to build dams are now ment, lenders and contractors, including evaluating decommissioning, removing, or limited disclosure and public access to re-operationalising ageing facilities that are information; due for re-licensing, many other countries ■ in many cases lack of clear monitoring do not yet have established licensing periods procedures limits public scrutiny and that clarify the responsibilities of the owner accountability; towards the end of the dam’s effective life. ■ there is a lack of sanctions at the international level for non-compliance with The net effect of these difficulties is that international norms regarding water use once a proposed dam project has passed in shared river basins; preliminary technical and economic feasibil- ■ within public international financial ity tests and attracted interest from govern- institutions, there are few, if any, sanc- ment, external financing agencies or politi- tions for staff members, or countries, for cal interests, the momentum behind the non-compliance; project often prevails over further assessments. Moreover where substantial differ- ■ in some countries, there is a lack of legal ences arise between proponents and those opportunities for affected groups to seek potentially affected, efforts to modify plans recourse, therefore lessening the ac- and decisions often must resort to legal or countability of the project developers; other action outside the normal planning and process. ■ most of the bilateral Export Credit Agencies are only beginning to develop But poor outcomes and mistrust are not social and environmental policies and simply a matter of narrow and technically guidelines and the lack of consistency focussed planning and decision-making. among the agencies’ guidelines has They also stem from the failure of dam resulted in projects rejected by some on proponents and financing agencies to fulfil environmental and social grounds commitments made, observe statutory receiving funding from other sources regulations and abide by internal guidelines. with lower standards. Among the findings on compliance are: ■ in some cases, the opportunity for To sum up, whereas substantial improve- corruption provided by dams as large- ments in policies, legal requirements and scale infrastructure projects further assessment guidelines have occurred, partic- distorted planning and decision-making; ularly in the 1990s, it appears that business

192 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

is still often conducted as usual when it ■ shift away from a balance comes to planning and decision-making. sheet approach to decision- Further, past conflicts remain largely unre- making in favour of broader, solved and past impacts largely unmitigated. inclusive and more timely, The WCD Global Review found that the multi-criteria approaches to influence of vested interests, legal and planning and decision- regulatory gaps, disincentives for compli- making; ance and lack of monitoring, participation ■ resolve past inequities and and transparency amongst other things, injustices, and transform have combined to create significant barriers project-affected people into to reforms that could otherwise make the beneficiaries, enabling planning and decision-making processes them to contribute to project benefits; more open, responsive and accountable. ■ conduct regular monitoring and periodic Recent examples cited in this and earlier reviews; and chapters are the basis of the Commission’s ■ develop, implement and enforce incen- optimism that these barriers are surmounta- tives, sanctions, and recourse mecha- ble and these difficulties are not inevitable. nisms, especially in the area of environ- The WCD Global Review indicates that mental and social performance. there are opportunities for reducing negative impacts and conflicts, and indeed a respon- The remainder of the report builds on the sibility, to: findings and lessons of the WCD Global ■ increase the efficiency and performance Review. It delivers a way forward that can of existing assets and systems; improve planning, decision-making and ■ better assess development needs and the compliance, capitalising on the options full range of development options; available – whether of a technological, ■ avoid and minimise ecosystem impacts; policy or institutional nature – and providing economically efficient, socially equitable ■ ensure that displaced and project- and environmentally sustainable solutions affected peoples’ livelihoods are im- to meet future water and energy needs. proved;

The Report of the World Commission on Dams 193 Chapter 6

Endnotes 13 Egypt and Ethiopia are now working towards greater collaboration through the Nile Basin 1 Many government agencies involved in water Initiative. resources development in countries that build 14 WCD Thematic Review on Negotiation, large numbers of dams maintain a construction section 3.4. workforce to build infrastructure: in the 15 Hanusin, 1999, opt052, WCD Submission, 4- United States the Bureau of Reclamation and 5. the Army Corps of Engineers and in China the 16 WCD Thematic Review III.1 Economic Water Resources Ministry. Analysis 2 Eckstein, 1958. 17 WCD Thematic Review V.2 Environmental 3 Gillis et al, 1987, p366. and Social Assessment, Section 1. 4 WCD Thematic Review V.5 Negotiation, 18 For example see the WCD India Country section 3.3. Study, section 7.4. 5 Sklar and McCully, 1994, eco029, WCD 19 WCD Thematic Review V.4 Regulation, Submission, p12-14; Gillis et al, 1987. section 3.1. 6 WCD Thematic Review III.2 Financing 20 Lovei and McKechnie, 2000, p34-37. Trends, ch. 3. 21 For example the 1996 Development Assist- 7 The total investment in dams by the multilat- ance Committee’s Rcommendation on Anti- erals and bilaterals portrayed in Figure 6.1 is corruption Proposals for Aid-funded Procure- approximately $125 billion. ment. 8 Guhan, 1995, cited in India Country Study, 22 OECD, 2000b; OECD, 2000c, website http:// Section 6.1.8. www.oecd.org/daf/nocorruption/index.htm, 9 ICE, 1994, p15-16; ICE, 1996, table 1. viewed 4 September 2000. 10 Usher, 1997b, p120-123. 23 For example the United Kingdom has the 11 Morse and Berger, 1992; Umaña, 1998, p7; Department for International Development Wappenhans Task Force, 1992. and the Export Credits Guarantee 12 Usher, 1997a, eco026, WCD Submission, Department. p120-123. 24 For example see WCD India Country Study.

194 PartingDams theand Waters:Development: The Report A New of theFramework World Commission for Decision-Making on Dams Decision-Making, Planning and Compliance

Part Two: The Way Forward

he mandate required the World Commission on TDams to propose a framework for options assessment and decision-making processes for water and energy resources development, along with a set of criteria and guidelines for the planning, design, construction,

operation and decommissioning of large dams. Part Two of the report presents a new approach to decision-making, based on the findings in the Global Review (Part One).

■ Chapter 7 presents a normative framework for equitable and sustainable development and develops an approach to negotiating outcomes for water and energy development projects based on recognising rights and assessing risks. ■ Chapter 8 sets out seven broad strategic priorities that should guide decision-making. Each one includes a set of principles that, if applied, will lead to more equitable and sustainable outcomes in future. ■ Chapter 9 develops supporting criteria and guidelines that will help decision-makers and all interested parties implement the strategic priorities set out in Chapter 8. ■ Part Two closes with Chapter 10 which stresses the need for concerted and simultaneous action and proposes entry points for the different constituencies involved in the dams debate to follow up in response to the recommendations of the Commission.

The Report of the World Commission on Dams 195 Enhancing Human Development

Chapter 7: Enhancing Human Development: Rights, Risks and Negotiated Outcomes

o improve development Toutcomes in the future we need to look at proposed water and energy development projects in a much wider setting – a setting that reflects

full knowledge and understanding of the benefits and impacts of large dam projects and alternative options for

all parties. It means that we have to bring new voices, perspectives and criteria into decision-making, and

we need to develop a new approach that will build consensus around the decisions reached. This will result in

fundamental changes in the way decisions are made.

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 197 Chapter 7

This chapter proposes a new basis interests and require negotiation. Reconcil- The debate about for assessing options and reaching ing competing needs and entitlements is the dams is a debate decisions on water and energy single most important factor in addressing about the very resources development. It links our the conflicts associated with development meaning, purpose and review of past experience con- projects and programmes – particularly pathway of tained in the Global Review large-scale interventions such as dams. development as well chapters with the WCD’s frame- as the role that the work for future practice elaborated Access to water provides a graphic illustra- state plays. in chapters 8 and 9. In developing tion of such competing needs and develop- this framework the Commission ment objectives and the reason why equity found that applying the lessons learnt does and justice considerations emerge as key not merely imply a change in process and issues. Riparian communities with long- procedure. The fault lines of the dams standing use rights and economies that debate run far deeper and touch upon many depend on local resources have an immedi- of the fundamental norms and values that ate interest in maintaining current use affect our lives as citizens and communities. patterns and assuring fulfilment of their future needs. However, in the context of In moving forward the Commission recog- national policies, meeting development nises that the dams debate is rooted in the needs may require sharing water resources. wider, ongoing debate on development. The To balance these needs societies will have to emerging global vision of equitable and negotiate a framework for equitably sharing sustainable development provides the the resource. History shows that this can be foundation for the Commission’s findings done successfully provided a transparent and and recommendations. This foundation legitimate process is followed. relates to: ■ the framework of internationally accept- Dams have often been seen as an effective ed norms on human rights, the right to way of meeting water and energy needs. development, and sustainability However, the Global Review has emphasised the wide range of problems associated ■ global trends and the emerging develop- with them. The Commission acknowledges ment paradigm; and that today’s perspective on development ■ a rights based approach where recogni- reflects the benefit of knowledge that may tion of rights and assessment of risks not have been available to past decision- provides the basis for negotiated deci- makers. Nonetheless, it is clear that the sions on dams and their alternatives. positive contribution of large dams to development has, in many cases, been From Global Review to marred by significant social and environ- Future Practice mental impacts which are unacceptable when viewed from today’s values. Along with all development choices, decisions on dams must respond to a wide The debate about dams is a debate about the range of needs, expectations, objectives and very meaning, purpose and pathway of constraints. As matters of public choice and development as well as the role that the policy they will always reflect competing State plays in both protecting the rights of

198 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

its citizens and responding to their needs advances these values significantly, we will through development policies and projects. emerge at our destination – improved The WCD Global Review showed clearly decision-making processes that deliver that large-scale infrastructure projects such improved outcomes for all stakeholders. as dams can have devastating impacts on the lives and livelihoods of affected commu- In the following sections the Commission nities and ecosystems, particularly in the presents a new policy framework for absence of adequate assessments and provi- decision-making on water and energy sions being agreed to address these impacts. development options that can be applied in During its regional consultations and national and local contexts. To improve through the WCD Case Studies the Com- development outcomes, ensure public mission was confronted with accounts by acceptance and reduce future controversy, communities and individuals on the nature this new basis for judgement needs to win and extent of these impacts. These accounts the support of the full range of key stake- give rise to fundamental concerns about the holders. It suggests that decision-making on way governments and their agencies have water and energy management will align exercised their role and responsibilities in itself with the emerging global trends on the development process.1 equitable and sustainable development.

Improving the development process and its Sustainable Human outcomes must start with a clear understanding of the shared values, objectives and Development – A Global goals of development and their implications Framework for institutional change. The Commission What are these trends, and how firm is their grouped the core values informing its direction and force? What do they imply for understanding on these issues under five decision-making? What do they say about main headings: the rights that societies, communities and ■ Equity individuals within societies, are entitled to ■ Efficiency and the responsibilities that accompany ■ Participatory decision-making these rights? ■ Sustainability There is a globally accepted framework for ■ Accountability setting universal goals, norms and standards. The foundations of the framework are the These five values run through the entire United Nations Charter (1945) and the report and are the focus of concerns raised by the evidence presented in the Global Review. Applying these values to the Box 7.1 Shared values and institutional practices – the UN Millennium evidence it has collected, the Commission Report believes that negotiated outcomes using a “The economic sphere cannot be separated from the more complex fabric of rights-and-risks approach will deliver the social and political life and sent shooting off on its own trajectory. To survive and most favourable development results. thrive, a global economy must have a more solid foundation in shared values and institutional practices– it must advance broader and more inclusive, social Reference to these values enables all stake- purposes.” holders to test decisions relating to water Source: Annan, 2000 and energy development. If the report

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 199 Chapter 7

Universal Declaration of Human Rights The application of a rights based approach (1947) (see Annex VI for full texts).2 recognises the indivisibility of civil, political, economic, cultural and social rights. It In the last two decades of the 20th century broadens the range of basic human rights the United Nations General Assembly beyond the socio-economic sphere of needs reinforced this framework with the UN to include rights to life, health care, educa- Declaration on the Right to tion, shelter, food, water, remedy, security, Development (1986) and the A rights based approach subsistence and livelihood. Rio Declaration on Environ- provides a principled ment and Development (1992) Unlike needs, which are expressed as basis for mediating (see Annex VI for full texts). aspirations for benefits, rights and entitle- development choices Taken with the earlier cove- ments are expressed in law, allowing for among competing nants and conventions on human their attainment or redress through the interests. rights, they cover a broad spec- justice system. A country may use its legisla- trum ranging from human rights, tive process to ensure that appropriate through social development and environment, rights-based policies are given legal expres- to economic co-operation. sion and to establish institutional mechanisms to uphold rights. The legal system also Human rights provides a means for resolving potential conflicts in cases where rights give rise to Reference to the framework of human rights competing entitlements. A rights based adopted by the international community in approach thus provides a principled basis for 1948 advances the process of planning and mediating development choices among decision-making in important ways. It competing interests. articulates such rights as self determination and the right to consultation in matters that The right to development affect people’s lives, the right to democratic representation of people’s views on such In 1986 the UN General Assembly adopted matters, the right to remedy and the right to the Declaration on the Right to Develop- an adequate standard of living, freedom from ment (DRD). 3 It marked a significant step arbitrary deprivation of property, freedom from by the international community in develop- violence, freedom of thought, conscience and ing a normative framework that specifies religion and freedom of opinion and expres- responsibilities in applying a human rights sion. More generally it includes the right to a approach to development. It moved beyond social and international order in which these the sphere of individual human rights to rights can be fully realised. address relationships between different interest groups in society and their interac- All people are accorded human rights tion with the state. without discrimination by virtue of their humanity. Reference to the human rights The Declaration on the Right to Develop- framework means those policies that deny ment sets out a number of relevant the rights of some to fulfil those of others concepts: cannot be adopted. Thus any policy or law ■ Development is a comprehensive process adopted must contain the intention to aiming at the constant improvement of respect the rights and entitlements of all. the well-being of the entire population;

200 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

it affects economic, civic, social, cultural State authority may also be and political rights. limited through adherence to ■ The promotion of, respect for and the framework of international enjoyment of certain human rights and conventions that, in certain fundamental freedoms cannot justify the circumstances, supersedes denial of other human rights and funda- strict sovereignty. mental freedoms. ■ The creation of conditions favourable to Sustainable the development of peoples and individ- Development – the uals is the primary responsibility of their Rio Principles States. The Declaration of the United ■ National development policies aiming at Nations Conference on the Human Envi- the constant improvement of the well- ronment (Stockholm, 1972) for the first being of the entire population and of all time accepted that the environment was individuals should be formulated on the fundamental to human well-being, and that basis of their active, free and meaningful its management and care in the interest of participation and fair distribution of advancing wider human goals was a central benefits resulting therefrom. task of States and the international commu- ■ The right of peoples to exercise full and nity. Articulation and codification of complete sovereignty over all their environmental rights took somewhat longer. natural wealth and resources. The United Nations Conference on Envi- ronment and Development adopted the Rio ■ The right to self determination. Declaration on Environment and Develop- ■ Equal opportunity for access to food and ment in June 1992. The Declaration con- housing. tains 27 principles, usually known as the Rio Principles4 . Several of these are of immedi- The DRD sought to clarify the role of the ate relevance to water and energy resources State in exercising its rights, responsibilities, management. duties and obligations in planning and ■ Principle 1 states that ‘Human beings are implementing national development poli- at the centre of concerns for sustainable cies and programmes. It reflects the recog- development. They are entitled to a nition that every society acts as an organised healthy and productive life in harmony polity in which the State is accorded powers with nature’. and responsibilities. At the same time States are subject to conditions that can be sum- ■ Principle 3 recognises the right to marised under the heading of good govern- development, but insists that it be met in ance criteria, such as those in the Declara- an equitable way that considers future tion on Human Rights including the rule of generations as well as present partici- law, accountable bureaucracies and freedom pants in development. of information. The legitimacy of the State ■ Principle 4 insists that sustainable in exercising its role is premised on the development requires environment to be assumption that it acts in accordance with integrated with the development process these criteria. Without good governance, and form a central feature of the aims of the legitimacy of the State and ultimately its that process. Environment, on its own, ability to take decisions are compromised. is an insufficient goal.

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 201 Chapter 7

■ Principle 10 underlines that all con- States according to their capabilities. cerned citizens must be involved in Where there are threats of serious or handling environmental issues, and must irreversible damage, lack of full scientific participate in the decision-making certainty shall not be used as a reason for process. This participation must be postponing cost-effective measures to accompanied by effective access to prevent environmental degradation. relevant information and by opportuni- ■ Finally, Principle 22 recognises the vital ties to seek redress and remedy in case role of indigenous people and other local agreements are not respected. communities in environmental manage- ■ Principle 13 states that States shall ment and development, and entrusts ensure compensation for victims of states with ensuring their effective environmental damage and give priority participation in the achievement of to the further development of law sustainable development. regarding liability in such cases. ■ Principle 15 states that the precautionary The Rio principles in conjunction with approach shall be widely applied by Agenda 21 thus highlighted not only the linkages between environment and development but also the importance of local Figure 7.1 The WCD policy framework communities having a significant role in shaping national development strategies.

Normative Development Framework The Global Review provided extensive Universal United Nations Rio Declaration evidence to illustrate that governments, in Declaration of Declaration on on Environment constructing dams, have often found them- Human Rights the Right to and Development Development selves in conflict with basic principles of good governance that have been articulated in the three international instruments referred to above. This situation still pre- WCD Core Values and Shared Understanding vails today. The level of conflict surround- Core Values ing large dams, yesterday and today, is Equity sufficient to illustrate that dams frequently Efficiency Participatory Decision-Making trigger disagreements about the respective Sustainability rights of governments and their citizens. Accountability The UN Declaration of Human Rights, the Rights and Risks Approach Right to Development and the Rio Princi- A Tool for Negotiated Decision-Making ples together make up an internationally accepted framework of norms empowering a Agenda for Implementation concept of development that is economically viable, socially equitable, and environ- Strategic Priorities and Policy Principles WCD Criteria and Guidelines mentally sustainable. It is a powerful framework with a central bearing on the dams debate. Figure 7.1 illustrates how the Commission draws on these internationally accepted norms in the remainder of this

202 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

report to develop a new policy framework At the same time, the decision criteria used and corresponding guidance for water and by governments do not always match those energy resources development. of organised groups of citizens. Govern- ments are too often inclined to invoke Trends and Challenges in urgent development needs as a reason for Applying the New restricting rights, while civil society groups believe that full respect for rights and the Development Framework search for alternatives represents the surest When invoking this emerging universal way of promoting equity and justice in normative framework, one must not over- development. state its completeness, its complete acceptance, or the ease of applying its provisions For many parts of the developing world, in practice. Nevertheless, recent trends in access to capital, technology and develop- global public policy suggest that increasing ment opportunities determines the extent to attention is being paid to the gap between which local and national economies are able aspiration and realisation. It is significant to develop. Similarly the political economy that the focus of the United Nations Devel- of power, vested interests and access to opment Programme’s (UNDP) Human resources that characterise each society have Development Report 2000 is on human rights a large influence on its commitment to and human development (see Box 7.2). The equitable and sustainable development. framework also strengthens the notion, now gaining currency in a variety of arenas, that This is not to suggest that the problems there is a body of common concerns based on reside in the poorer countries. Pressure for a range of international conventions and water and energy development – in both accords that transcend national sovereignty.

The adoption of a rights based approach Box 7.2 Human rights and human development does not on its own resolve the practical challenge of meeting human needs. During The UNDP Human Development Report 2000 focuses on human rights as the fundamental framework within which human development must be pursued. It its regional consultations the Commission contends that societies are on the threshold of a significant advance in the listened to a wide range of views and reason- recognition of, and respect for, human rights. But this will require six fundamental shifts from the thinking that dominated the 20th century: ing on this matter. Meeting rapidly growing ■ From state-centred approaches to pluralist, multi-actor approaches – needs for water and energy – particularly in with accountability not only for the State but also for media, corpora- the developing economies of the South – tions, schools, families, communities and individuals. imposes difficult choices on governments. ■ From national to international and global accountabilities – and from the international obligations of States to the responsibilities of global actors. Failure to respond to these needs carries ■ From the focus on civil and political rights to a broader concern with all significant economic and political risks. rights – giving as much attention to economic, social and cultural rights. Food security, blackouts, empty water taps ■ From a punitive to a positive ethos in international pressure and and floods are among the most immediate assistance – from reliance on naming and shaming to positive support. and sensitive public service issues for which ■ From a focus on multiparty elections to the participation of all through society holds government accountable. In inclusive models of democracy. ■ the past, large-scale dam projects seemed to From poverty eradication as a development goal to poverty eradication as social justice, fulfilling the rights and accountabilities of all actors. offer both apparently straightforward and Source: UNDP, 2000 highly visible options for responding to these pressures.

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 203 Chapter 7

North and South – is not only caused by the activities that offered a clear economic imperative of meeting basic human needs, return. This does not mean that econom- but is also driven by wasteful consumption ic viability is no longer seen as impor- in the richer countries or among the well-off tant. If anything it has greater weight in the poorer countries5 . along with a greater sanction on poor economic performance at company or Whatever judgement is made, it is a reality country level. But we are giving greater that governments face very real dilemmas in value to non-monetary or non-monetised trying simultaneously to satisfy urgent needs aspects of development such as the need and advance the realisation of fundamental to conserve biological diversity, protect rights, even if the goal of fulfilling all cultural values, or consider the needs of people’s needs and entitlements is not future generations. questioned. Notwithstanding this, the ■ We are moving from assessing public Commission believes that interest in general terms to a focus on fulfilling development needs improving equity in the spread of costs Governments face very requires respect for fundamental and benefits from development. The real dilemmas in trying rights, and not a trade off growing disparity between rich and poor simultaneously to satisfy between them. We believe that within and across many nations has urgent needs and an equitable and sustainable fuelled doubts about traditional develop- advance the realisation approach to development ment paths.6 The emerging consensus on of fundamental rights, requires that a decision to build the need for greater transparency and even if the goal of a dam or any other options must participation in development decision- fulfilling all people’s not, at the outset, sacrifice the making is likely to speed up this transi- needs and entitlements rights of any citizen or group of tion considerably. The focus on equity is not questioned. affected people. extends to recognising intergenerational equity as an important factor in dealing In developing its framework and recommen- with resource access and use. dations the Commission has sought to draw on the broader trends and developments ■ An increasingly robust foundation of that reflect the changing context and international covenants, charters, international development discourse. Not declarations and conventions supports all countries will recognise themselves in the sharpening focus on equity and these statements and the trends are far more confirms the growing importance of advanced in some areas than in others. equity considerations in development. In Nevertheless, the Commission believes that particular, a body of international the trends described below are not limited instruments relating to human rights is to any one region or group of countries, but emerging, together with institutions to have broad relevance. From the perspective oversee their further development and of this shared experience we draw attention application. This will strengthen argu- to the following elements of the evolving ments in favour of greater transparency, development paradigm: participation in decision-making and ■ The world appears set to move beyond accountability for compliance. Pressure the growth paradigm, which judged from the human rights community has progress largely in narrow economic had an impact on governments and more terms, putting a strong premium on recently on corporations.

204 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

■ The definition of public interest is sector and civil society. shifting from one that placed a premium This is not to say that the on overriding interests of economic role of governments has growth to one that places more weight become less important; on the rights and interests of people and but it has changed and communities affected by a development. continues to change, with The level of sacrifice that affected people implications for the way are expected to endure for an often ill- decisions are taken and defined notion of the greater public good implemented. has been increasingly challenged. Simi- ■ The private sector has, by larly, the recognition that affected contrast, considerably communities, through their sacrifices, are expanded its role, undertaking functions in fact contributors to development that were until recently the exclusive projects implies a shift in focus from remit of government. Apprehensions compensatory approaches to establishing about this trend are sharp in some parts equitable benefit sharing mechanisms. of the world and reflect growing concern ■ We have also witnessed a shift from about the diminishing power of citizens technology-driven development choices to control corporate activity through towards a more integrated approach to local and national institutions. Pressure managing scarce resources with technol- is therefore growing on corporations to ogy being but one factor among others in become more accountable to widely managing demand and supply of services supported standards of social and envi- more effectively. Furthermore, the ronmental behaviour. Such pressure is recognition that traditional practises and likely to mean that corporations will face technologies can achieve great levels of steadily rising costs and risks if they fail efficiency in meeting local needs coupled to comply with existing rules, regulations with the advent of new technology and standards. options has increasingly challenged the ■ The role of civil society organisations notion that large and centralised systems has also expanded and their legitimacy in are always the most effective and effi- representing and defending interests, in cient way of meeting demands for water participating as full actors in decision- and energy. making on development and in monitor- ■ The emerging paradigm provides a new ing compliance is increasingly accepted. basis for governance and democratic Civil society organisations are playing an decision-making. This stems from a increasingly important role in influenc- substantial redistribution of roles and ing public opinion and mobilising it responsibilities in the public and private against infringement or non-compliance sectors and civil society. Many parts of with new and emerging standards of the world have seen a considerable behaviour, especially on the social and migration of national government environmental front. authority, in three directions: upward to regional superstructures or international This changing context implies a broadening bodies, downward to provincial and local range of concerns that have a legitimate government; and outward to the private right to be considered and, therefore, of

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 205 Chapter 7

actors involved in reaching key Rights … development decisions. It will The Global Review highlighted the need for be increasingly difficult to take a more practical and specific approach to decisions on the narrow basis of addressing the five values of equity, efficien- the needs of infrastructure cy, participatory decision-making, sustaina- development. Instead, such bility and accountability. These values form projects need to be considered the foundation of a rights-based approach to as part of the broader process of equitable decision-making about water and economic, social and environ- energy resources management. mental transformation.

Various types of rights may be relevant in Rights and Risks – the context of large dam projects. These an Improved Tool include constitutional rights, customary for Decision-Making rights, rights codified through legislation, property rights or the rights of developers Both the findings of the WCD Global and investors. They can be classified on the Review and the implications of the norma- basis of their legal status, their spatial and tive framework summarised in this chapter temporal reach, or their purpose. In the demonstrate that the traditional ‘balance spatial and temporal dimensions, one can sheet’ approach of assessing costs and distinguish the rights of local, basin, region- benefits of a project is an inadequate tool for al and national entities, the rights of ripari- effective development planning and deci- an countries, or the rights of present and sion-making. The case of dams clearly future generations. Regarding the purpose or illustrates that development choices made subject of rights, one can distinguish rights on the basis of such trade-offs neither to material resources such as land and water, capture the complexity of considerations and rights to spiritual, moral or cultural involved, nor can they adequately reflect goods such as religion and dignity. the values societies attach to different options in the broader context of sustainable This approach highlights the range and development. complexity of relevant rights and responsibilities and the reality that rights intersect Given the significance of rights-related and overlap. Mechanisms for conflict issues as well as the nature and magnitude of resolution, adjudication and independent potential risks for all parties concerned, the arbitration must begin with the assessment Commission proposes that an approach of these rights, entitlements and claims. based on ‘recognition of rights’ and ‘assess- This approach assumes that at the assessment of risks’ (particularly rights at risk) be ment stage, all claims are subject to a fair, developed as a tool for guiding future open and transparent review. It is based on planning and decision making (see Figure an understanding that no party’s rights will 7.2). This will also provide a more effective extinguish another’s. In fact, where rights framework for integrating the economic, compete or conflict, negotiations conducted social and environmental dimensions for in good faith, offer the only process through options assessment and the planning and which various interests can be legitimately project cycles. reconciled. This suggests an approach to

206 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

water and energy policy that provides for legitimate stake in a decision are best placed negotiated processes within a legal and to assess the risks they are prepared to take procedural framework, including arbitration, to achieve a benefit. Such conflicts are recourse and appeal mechanisms to ensure exacerbated by the absence of an agreed equitable adjudication in cases where negotiated settlements are not achievable or Box 7.3 Voluntary risk takers and involuntary risk bearers are contested. Public and private developers of large dam projects have long understood that the sector involves managing risks of a technical, financial and even political Clarifying the rights context for a proposed nature. Decision-makers have not always acknowledged the differences between ‘taking risk’ and ‘imposing risk’ and between voluntary risk takers and project is an essential step in identifying involuntary risk bearers. The private sector regards dams as high-risk projects. those legitimate claims and entitlements As ‘voluntary risk takers’, private companies manage their increased exposure to risk by requiring higher financial rates of return. Their risk management that might be affected by the proposed procedures are relatively highly developed, using contractual agreements and project – or indeed its alternatives. It is also sophisticated third party recourse and arbitration mechanisms. the basis for effective identification of Governments and regulators plan and manage the provision of services to the nation, and therefore also take risks. They must weigh the risks inherent in stakeholder groups that are entitled to a undertaking dam projects against the risks of not undertaking them. There are formal role in the consultative process, and risks attached to other options, and to the ‘do nothing’ option, given growing demand for power or food, and societies are constantly balancing these eventually in negotiating project-specific different risks and opportunities. agreements relating, for example, to benefit There are those, however, on whom risk is imposed. The ‘involuntary risk sharing, resettlement or compensation. bearers’ who are forced to bear risks include people to be displaced by the project. These people may face years of uncertainty and direct risks to livelihood even before the project is approved and before resettlement or land purchase. They may be unable to obtain finance for investments in farm infrastructure or …and Risks equipment, and local government may not maintain or develop services for communities on the verge of displacement. The risks to displaced communities The notion of risk adds an important are compounded in cases where they have no say in the decisions but are dimension to understanding how, and to obliged to bear the consequences. In these circumstances they often depend entirely on the capacity of government or the developer to manage the what extent, a project may impact on such resettlement or compensation process on their behalf . rights. Traditional practice is to restrict the Indigenous peoples face specific cultural, social and livelihood risks. Evidence definition of risk to the risk of the developer collected by the Commission illustrates that they often bear disproportionate risks associated with projects, as they were not included in decision-making or corporate investor in terms of capital processes concerning resettlement, let alone the earlier processes of assessing invested and expected returns. These needs and selecting options. Downstream communities that depend on existing river flows to maintain their resource base are also often not given any say in voluntary risk takers have the capacity to deciding the nature of projects. Yet they face the risk of losing access to define the level and type of risk they wish to resources, or having their capacity to maintain a sustainable livelihood undermined. Often these communities do not have access to information on the take and explicitly to define its boundaries nature of the risks that they face until the project is approved or completed.7 and acceptability. By contrast, as the Global The case of future generations and the ecosystem is somewhat different. These Review has shown, a far larger group often ‘risk bearers’ cannot speak for themselves, even if the risks they face are acknowledged. Future risks can be linked to present risks. The loss of natural have risks imposed on them involuntarily resources can undermine livelihood opportunities for both current and future and managed by others. Typically, they have generations. Similarly, the loss of biodiversity in the present means that it is either not available or diminished for future generations. The lower priority no say in overall water and energy policy, generally accorded to these types of risks is compounded by the absence of the choice of specific projects or in their tangible safeguards, or the failure to implement and enforce those that do exist. In such cases, as with other involuntary risk bearers, adopting a precautionary design and implementation. The risks they approach is particularly relevant in order to avoid impacts. It is also essential to identify appropriate inputs by interested parties to the options assessment face directly affect individual well-being, process and to the planning and project cycles. livelihoods, quality of life, even their The precautionary approach articulated in the Rio principles forms part of a spiritual world view and very survival. structured approach to the analysis of risk, and is also relevant to risk management. Decision-makers faced with scientific uncertainty and public concerns have a duty to provide answers where risks and irreversibility are considered This has often led to conflict because it unacceptable by society. ignores the principle that those with a

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 207 Chapter 7

approach and process for assessing and 7.3). Most important, involuntary risk negotiating acceptable project outcomes – bearers must have the legal right to engage outcomes that include rejecting a dam in with risk takers in a transparent process to favour of a more acceptable alternative. ensure that risks and benefits are negotiated Where unequal power relationships prevail on a more equitable basis. To that end, a and no process for good-faith adjudication framework is required that permits a trans- among competing interests exists, the result parent, balanced and participatory process of is often protracted conflict, escalation and, decision-making relating to key stages in the eventually, ‘win-lose’ outcomes in which less planning process. In chapters 8 and 9 we privileged groups are further disadvantaged. develop the practical application of this This all-too-common reality points to a approach in greater detail to illustrate how this serious failure of the options assessment and can be achieved at the different stages of planning process to protect and further basic assessment, planning and implementation. human and development rights. It must, however, be stressed that not all Dealing with risks cannot be reduced to countries possess the full range of legal and consulting actuarial tables or applying a institutional structures, nor sometimes the mathematical formula. In the end, as in the human and financial resources, to imple- case of rights and entitlements, they must be ment such a participatory approach to identified, articulated and addressed explic- decision-making effectively. It becomes, itly. This will require the acknowledgement therefore, a high priority to assist those of risk to be extended to a wider group than countries and communities to put the governments or developers in order to necessary structures in place and to develop include both those affected by a project and the necessary capacity. the environment as a public good (see Box Negotiating Agreements Fig 7.2 From rights and risks to negotiated agreements: a on the Basis of Rights and framework for options assessment and project planning Risks

RIGHTS RISKS In its Global Review the Commission encountered considerable experience and Recognition of Rights and Assessment of Risks lead to identification of stakeholders good practice in implementing a rights-based approach, with examples from many countries. It is also clear, however, that the Forum established for needs and options approach requires a legal and procedural assessment and project planning No consensus framework that provides for a free and informed negotiation process. This frame-

Successful Independent work must provide for arbitration, recourse mediation and/or review and and appeal mechanisms to ensure equitable arbitration mediation adjudication in cases where negotiated settlements are not achievable. Specific agreements are No agreement leads to negotiated and become part of selection of an alternative project compliance framework project option, arbitration or While it is easy to point to negotiated judicial review solutions as the answer to sound decision- making or to avoiding dam related conflict,

208 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

a range of difficult challenges remains to be recognises there is no universal formula. faced. Structuring a negotiation process that The most appropriate decision-making will lead to an optimal and widely supported process will depend to an extent on the type decision on an option for water and energy of development under review, the political development involves answering three and cultural setting of the development, and questions: other constraints relating to the urgency of ■ Who should participate in the decision- the need and the likelihood of negative making process? impacts. ■ What decision-making processes should On the other hand a process that is too be followed? complex can needlessly delay decisions and ■ What criteria can be applied to assess the deprive potential beneficiaries of the fruits process and its outcomes? of any of the development alternatives under consideration. The goal must be a Who should participate? process that gives all key stakeholders a The rights-and-risks approach is both an voice and a full opportunity to participate in entry point to the options assessment decision-making, seeks the broadest reason- process and a basis for the subsequent able consensus, and is transparent in the development of specific project options. The criteria used for reaching a decision. Such a recognition of rights and the assessment of process is likely to ensure the demonstrable risk identify the interested and affected public acceptance that projects require if parties who possess rights or entitlements as they are to achieve development. However, well as risk takers and bearers. This opens no process will work unless all of the parties the way for a negotiated approach that enter the negotiation in good faith. Without enables the decision-making process to this there is the danger that any attempt to assess options and reach project agreements. make the process more inclusive will end up Those whose rights are most affected, or being a recipe for stalemate, putting the whose entitlements are most threatened, achievement of needed benefits at risk. have the greatest stake in the decisions that are taken. The same applies to risk: those Negotiated outcomes do not replace govern- groups facing the greatest risk from the ment decision-making: on the contrary, they development have the greatest stake in the rest on the State actively fulfilling its role as decisions and, therefore, must have a planner and enabler of development choic- corresponding place at the negotiating table. es, as well as its responsiblity to provide services and safeguard entitlements. When What type of process? This chapter has suggested that global trends are increasingly leading us to deci- Box 7.4 Good governance and the UN Millennium Report sion-making based on transparent processes, “Good governance comprises the rule of law, effective state institutions, full access for affected parties to relevant transparency and accountability in the management of public affairs, respect for information, identification and empower- human rights, and the participation of all citizens in the decisions that affect their lives. While there may be debates about the most appropriate forms they ment of key stakeholders, and their ade- should take, there can be no disputing the importance of these principles.” quate participation in the decision-making Source: Annan, 2000 process. At the same time, the Commission

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 209 Chapter 7

a negotiation process results in a In Chapter 9, the Commission develops full agreement among the criteria and guidelines to illustrate how this parties, the government (as one can be achieved at the different stages of of the parties) need only en- assessment, planning and implementation. dorse it. There will always be cases, however, where a con- Conclusion certed effort of all parties acting in good faith has brought Large dams have increasingly been charac- agreement closer but left it terised by bitter conflict and deep feelings of unattainable. Where independ- resentment and injustice. Beginning to ent review and mediation fail to foster an correct this situation will require not only agreement, alternative options should be new processes for taking decisions, but considered or the project should go to building confidence in these processes and arbitration. All decisions must be taken their ability to deliver genuinely better within the framework set by constitutions outcomes for water and energy resource and national legislation as well as interna- development. In seeking to build this tional conventions, and remain subject to a confidence we do not, in many cases, begin citizen’s right to challenge them in the with a clean slate, but with a difficult legacy courts (see Figure 7.2). that needs to be recognised. This legacy can only be overcome if there is a rapid invest- Assessing the Process and its ment of confidence in the legitimacy of the Outcomes processes that are put in place. The purpose of engaging in a participatory This Chapter has defined the interaction process for decision-making is to deliver between the Commission’s findings and the better decisions than would otherwise global development debate. It has pointed emerge. The proof of a decision-making out that the debate on large dams is not process resides both in the process used, and taking place in isolation from the broader in the outcomes delivered. A process may be debate on the purposes and pathways of deemed successful if it has been, and has development. It can clearly be situated been seen as: within a framework relating to human ■ Fair: all key stakeholders perceive the rights, the right to development, and the process and outcomes to be fair and imperative of sustainability. legitimate. ■ Wise: the process is fully-informed, Further, the Commission has sought to making best use of available knowledge, demonstrate that an approach based on the and continuing to make best use of recognition of rights and assessment of risks knowledge over time. can lay the basis for greatly improved and ■ Efficient: the process and the solution are significantly more legitimate decision- both cost- and time-effective, making making on water and energy development. best use of available resources. This is an effective way to determine who ■ Stable: the agreement is likely to endure has a legitimate place at the negotiation and can be adapted – a sign that it table and what issues need to be included on maintains its legitimacy.8 the agenda.

210 PartingDams theand Waters:Development: The Report A New of Frameworkthe World Commission for Decision-Making on Dams Enhancing Human Development

Finally, it has concluded that only decision- issues surrounding dams. While presenting making processes based on the pursuit of greater demands at early stages of options negotiated outcomes, conducted in an open assessment and project design, such process- and transparent manner and inclusive of all es lead to greater clarity, certainty and legitimate actors involved in the issue, are legitimacy for subsequent steps in decision- likely to resolve the many and complex making and implementation.

Endnotes World Commission on Water (World Com- mission on Water in the 21st Century, 2000), 1 The WCD received submissions from a wide the World Energy Assessment (UNDP et al, range of interested parties, which have been 2000) and The World’s Water (Gleick, 2000). listed on the WCD web site. A number of 6 An overview of this discourse can be found in these were presented at the four regional Sen, 1999. Current data and evolving policy consultations, which the Commission held in frameworks for addressing poverty and equity Colombo, Sao Paulo, Cairo and Hanoi. issues have also been reviewed in the latest 2 UN, 1947. World Development Report (World Bank, 2000) which focuses on poverty and development. 3 UN, 1986. 7 Cernea, 2000, has developed a detailed 4 UNCED, 1992. assessment of these risks. The ‘Impoverish- 5 Recent reports summarising the implications ment Risk Analysis’ approach is referred to in of high levels of consumption on the natural greater detail in Chapter 9. resource base include the Report of The 8 Susskind and Cruikshank , 1989.

TheParting Report the ofWaters: the World The ReportCommission of the on World Dams Commission on Dams 211 Strategic Priorities – A New Policy Framework

Chapter 8 Strategic Priorities – A New Policy Framework for the Development of Water and Energy Resources

ased on the findings of the B WCD’s Global Review this chapter develops the Commission’s rationale and recommendations in the form of seven strategic priorities

and related policy principles for future decision-making. It builds on previous chapters, starting with

Chapter 1 which locates the dams debate in a broader context. This context includes the history of water

resources management and large dams, the big increase in dam construction during the latter half of

the 20th century, the subsequent emergence of conflict and the issues

The Report of the World Commission on Dams 213 Chapter 8

and interests that gave rise to the establish- approach for achieving negotiated outcomes. ment of the Commission. Chapter 8 takes this forward, moving from a traditional top-down, technology focused Chapter 8 draws extensively on the Know- approach to advocate significant innovations ledge Base summarised in Chapters 2 to 6 in assessing options, managing existing dams, which review the performance and impacts gaining public acceptance and negotiating and of dams, the decision-making process and sharing benefits. the available options for providing water and energy services. Much of this work The Commission sets out this constructive involved reviewing existing information, and innovative way forward for decision- but the Commission also collected impor- making in the form of the seven strategic tant new information on all aspects of dams priorities listed here and elaborated in subse- operation and management, especially their quent sections of the chapter (see Figure 8.1). social and environmental impacts. The wide range of consultations with affected parties The priorities are: is an important contribution to knowledge ■ Gaining Public Acceptance about dams in development and develop- ■ Comprehensive Options Assessment ment practice in general. ■ Addressing Existing Dams In Chapter 7 the Commission moved from ■ Sustaining Rivers and Livelihoods the review of past experience to look at ■ Recognising Entitlements and Sharing directions for the future. The chapter Benefits addresses the dams debate in the context of ■ Ensuring Compliance the wider debate on equitable and sustainable ■ Sharing Rivers for Peace, Development development and the corresponding frame- and Security work of internationally accepted norms and standards. It introduces the rights-and-risks A key message and a set of policy principles Figure 8.1 The WCD’s seven strategic priorities support each of the seven strategic priorities. They are expressed in the form of achieved outcomes. A section on the rationale explains the Commission’s thinking on each strategic Gaining Comprehensive public options priority and an elaboration looks at broader acceptance assessment issues involved in achieving the strategic priority. These strategic priorities provide Sharing guidelines for all affected parties on a new way rivers for Equitable and Addressing peace, sustainable existing forward – one that is founded on achieving development dams and security development of equitable and sustainable development water and energy through a process that successfully integrates resources social, economic and environmental consider- Sustaining ations into decision-making on large dams and Ensuring rivers and compliance livelihoods their alternatives. Recognising entitlements and sharing Chapter 9 provides an operational approach benefits for applying these priorities to the planning and project cycles.

214 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

Strategic Priority 1 Gaining Public Acceptance

Key Message

Public acceptance of key decisions is essential for equitable and sustainable water and energy resources development. Acceptance emerges from recognising rights, addressing risks, and safeguarding the entitlements of all groups of affected people, particularly indigenous and tribal peoples, women and other vulnerable groups. Decision-making processes and mechanisms are used that enable informed participation by all groups of people, and result in the demonstrable acceptance of key decisions. Where projects affect indigenous and tribal peoples, such processes are guided by their free, prior and informed consent.

Effective implementation of this strategic priority depends on applying these policy principles:

1.1 Recognition of rights and assessment of risks 1.3 Demonstrable public acceptance of all key are the basis for the identification and decisions is achieved through agreements inclusion of stakeholders in decision-making negotiated in an open and transparent on energy and water resources development. process conducted in good faith and with the informed participation of all stakeholders. 1.2 Access to information, legal and other support is available to all stakeholders, particularly 1.4 Decisions on projects affecting indigenous and indigenous and tribal peoples, women and tribal peoples are guided by their free, prior other vulnerable groups, to enable their and informed consent achieved through informed participation in decision-making formal and informal representative bodies. processes.

Rationale coupled with the significant involuntary risk imposed on the most vulnerable, is central Because of their scale and complexity, dams to the dams debate and associated conflicts. affect the existing rights of different groups and create a wide range of significant risks To be socially legitimate and produce for a diverse range of interest groups. positive and lasting outcomes, development Among those affected are indigenous and projects should provide for tribal peoples, women and other vulnerable greater involvement of all groups who have been shown to suffer interested parties. A fair, in- A fair, informed and disproportionately. This has been com- formed and transparent decision- transparent decision- pounded by negligible participation of these making process, based on the making process, based groups in decision-making processes, with acknowledgement and protection on the the result that planning processes for large of existing rights and entitle- acknowledgement and dams have frequently overlooked gender and ments, will give all stakeholders protection of existing equity aspects. The vulnerability of these the opportunity to fully and rights and entitlements, groups stems from the failure to recognise, or actively participate in the will give all respect their rights, and from the significant decision-making process. Instead stakeholders the involuntary risks imposed on them. of exacerbating existing inequali- opportunity to fully and ties, water and energy resources actively participate in Failure to recognise the rights of all affected development should be opportu- the decision-making groups, whether legally sanctioned or not, nities for achieving a high level process.

The Report of the World Commission on Dams 215 Chapter 8

of equity. The planning process should be nous and tribal peoples empowers them at sensitive to, and take account of, social and the negotiating table. economic disparities, and devise and implement mechanisms for addressing them. Negotiations conducted in good faith that lead to an agreed outcome would secure Recognising indigenous and tribal wider acceptance of development policies people’s rights and projects. International and national policy making increasingly recognise that historical and Elaboration of Policy Principles continuing wrongs committed against

indigenous and tribal peoples call for 1.1 Recognition of rights and assessment distinct measures to protect their rights. of risks are the basis for the identifica- These measures include the free, prior and tion and inclusion of stakeholders in informed consent of indigenous and tribal decision-making on energy and water resources development. peoples to developments that may affect them. To achieve this, the participation of Water and energy resource development indigenous and tribal peoples must become projects can affect the existing rights of an integral part of the decision-making community groups in many different ways process. This is increasingly being recog- and can lead to a variety of risks. Legal and nised in international and national law. customary rights take many forms, including livelihood, resources, habitat, social net- International legal instruments, such as works and cultural heritage. Recognising Conventions 107 and 169 of the Interna- this variety makes it possible to identify the tional Labour Organisation and the evolv- risks facing communities. ing United Nations Draft Declaration on the Rights of Indigenous Peoples, recognise and Identifying rights and risks and recognising support the concept of free, prior and how they affect different parties gives informed consent. Other planners an objective basis for identifying organisations reflect this stakeholders. These stakeholders must Identifying rights and risks trend including the Inter- participate fully and actively in the deci- and recognising how they American Development sion-making process and be party to all affect different parties gives Bank through its operational negotiated agreements throughout the planners an objective basis policy, which requires process, from options assessment to final for identifying stakeholders. informed consent of indige- implementation, operation and monitoring. nous and tribal peoples to The involvement of women and other resettlement and compensa- vulnerable groups in decision-making should tion measures. Similar reforms are found at be ensured at all stages of the planning and national level in a number of countries.1 implementation process. There should be clear consideration for the vulnerabilities Through acknowledging the rights of that expose women to project impacts vulnerable groups, and providing for their (displacement, changes in the resource base full and active participation in the decision- and resulting disruptions of social and eco- making process, all the risks associated with nomic resources and networks) and for the a decision can be addressed. Requiring the specific obstacles that reduce their opportuni- free, prior and informed consent of indige- ties to share benefits generated by the project.

216 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

At the needs and options assessment stage, and energy resources development. To strategic impact assessment enables identifi- achieve mutually agreed outcomes, stake- cation of stakeholders. Impoverishment risk holders should negotiate through recognised assessment conducted at the pre-feasibility stakeholder bodies. Public acceptance of the stage will enable identification of stakehold- decision reached by stakeholders through ers bearing risk voluntarily and involuntarily this process should guide progress at key for participation in the decision-making stages in the assessment, selection, planning process. and implementation of the project.

1.2 Access to information, legal and The following key principles define the other support is available to all nature of open and transparent decision- stakeholders, particularly indigenous and tribal peoples, women and other making processes. The process: vulnerable groups, to enable their ■ is democratic, accountable and enjoys informed participation in decision- public confidence; making processes. ■ safeguards the rights and entitlements of Various stakeholders have significantly vulnerable groups by addressing imbal- different capacities to participate fully and ances in political power; actively in the development planning ■ promotes women’s participation and process. Rural communities, indigenous and gender equity; tribal peoples, women and other vulnerable ■ is guided by the free, prior and informed groups are at a disadvantage in accessing legal consent of indigenous and tribal peoples; and financial resources and in their capacity to and participate in negotiating decisions. ■ is based on the willing participation of all parties negotiating in good faith In order for these groups to participate fully throughout all key stages, from options and actively in negotiations, they need assessment to final implementation, access to adequate resources, including legal operation and monitoring. and other professional support. Communi- ties also need sufficient time to examine Negotiations should result in demonstrable various proposals and to consult amongst public acceptance of binding formal agree- themselves. ments among the interested parties with clear, implementable institutional arrange- Resources committed to achieving these ments for monitoring compliance and ends must target a continuing process of redressing grievances. capacity building.

A stakeholder forum can facilitate this 1.3 Demonstrable public acceptance of all key decisions is achieved through process. This forum could be an existing agreements negotiated in an open planning institution located at the local, and transparent process conducted sub-national and national levels. Countries in good faith and with the informed participation of all stakeholders. that already have such planning institutions must ensure representation of rural commu- Participatory processes need to secure public nities, indigenous and tribal peoples and acceptance of plans and projects for water other stakeholders in them. Countries

The Report of the World Commission on Dams 217 Chapter 8

without such planning institutions should International law includes a body of con- consider creating a stakeholder forum for ventions and customary norms that increas- the purpose (see Chapter 9 for guidelines). ingly recognise the rights of indigenous and tribal peoples. Aspects of the national laws Negotiating agreements of many countries now reflect contemporary Reaching a negotiated agreement may need views of indigenous rights.2 assistance from an agreed independent third party from time to time. This assistance is Some of these changes are a direct response best provided through an independent dispute resolution body that: to indigenous peoples’ campaigns demanding social justice and development opportu- ■ is constituted with the participation and nity – including campaigns concerning agreement of stakeholders; and dams. However, these provisions have not ■ has the necessary skills, legal and admin- been very successful in protecting the rights istrative capacity for this purpose. of indigenous and tribal peoples.

This body should agree on a negotiating To the extent that historical and present process with all stakeholders at the outset. injustices continue to deny indigenous and Stakeholders should refer tribal peoples the right to self-determina- disagreements on any aspects of tion, countries increasingly recognise that Demonstrating public the negotiations to this body to they are entitled to distinct measures to acceptance, and examine them and provide protect their rights. This recognition has upholding negotiated assistance to the parties. This included prescriptions of non-discrimina- decisions, is best includes determining whether tion, cultural integrity, control over land achieved through stakeholders are negotiating in and resources, social welfare and develop- binding and formal good faith and suggesting ways ment and self-government. agreements. of reaching a settlement. Identification of indigenous and tribal Demonstrating public acceptance, and peoples upholding negotiated decisions, is best Several countries have clear laws and achieved through binding and formal agree- procedures identifying and recognising ments. They must include mechanisms for indigenous and tribal peoples. However, the hearing and settling subsequent grievances. situation is unclear in some other countries.

At its broadest, the adjective ‘indigenous’ is The Commission recognises that coercion applied to any person, community or being and violence have been used against com- that has inhabited a particular region or munities affected by dams. All project place prior to colonisation. However, the proponents – public and private – need to commit to the strict prohibition of such acts term ‘indigenous peoples’ has gained curren- of intimidation against any stakeholders. cy internationally to refer more specifically to long-resident peoples, with strong cus- 1.4 Decisions on projects affecting tomary ties to their lands, who are dominat- indigenous and tribal peoples are ed by other elements of the national society. guided by their free, prior and informed consent achieved through formal and informal representative The general trend in the United Nations bodies. and other international organisations has

218 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

been to accept that many of the so-called ■ Vulnerability to being disadvantaged in ‘tribal peoples’ of Africa, Asia and the the development process. Pacific are indistinguishable from indige- ■ Close attachment to ancestral territories nous peoples as far as international law and and to natural resources in such areas. 3 standards are concerned. The International ■ Self-identification as distinct from the Labour Organisation’s (ILO) Convention dominant group or groups in societies, 169 applies to both indigenous and tribal and identification by others as members peoples and thus includes many such of a distinct group. peoples from Asia and Africa. It ascribes the same rights to both categories without Securing free, prior and informed consent discrimination. Article 1(2) of ILO Con- The requirement for free, prior and in- vention 169 notes: ‘Self-identification as formed consent gives indigenous and tribal indigenous or tribal shall be regarded as a communities the power to consent to fundamental criterion for determining the projects and to negotiate the conditions groups to which the provisions of this under which they can proceed. The effec- Convention apply.’ tive implementation of this practice marks a significant step forward in recognising the In countries that do not explicitly define rights of indigenous and tribal peoples, indigenous and tribal peoples, the Commis- ensuring their genuine participation in sion proposes an alternative approach for decision-making processes and securing identifying them. Several international their long-term benefits. organisations and agencies have adopted or proposed this approach which uses the The concept of free, prior and informed criteria listed below to recognise indigenous consent achieved through formal and and tribal peoples.4 In terms of this ap- informal representative bodies should guide proach, the requirements for free, prior and decision-making on dams and their alterna- informed consent should apply to groups tives. Moreover, the Commission believes that satisfy the following criteria: that all countries should be guided by the concept of free, prior and informed consent, ■ Historical continuity with pre-colonial regardless of whether it has already been societies, which is determined on the enacted into law. basis of the following criteria, regardless of whether they are formally recognised Failing that, decisions should only be made as indigenous or tribal peoples or not: following a process of good faith negotia- ■ Subsistence oriented and natural tions that allows for the effective represen- resource based production systems tation of the peoples’ concerned, including ■ Presence of customary social and genuine attempts to reconcile differences political institutions through the mutually agreed dispute resolution process, with disagreements being ■ An indigenous language, often referred to a designated judicial body. different from the national language ■ An experience of subjugation, exclusion The customary laws and practices of the or discrimination, whether or not these indigenous and tribal peoples, national laws conditions persist. and international instruments will guide the

The Report of the World Commission on Dams 219 Chapter 8

manner of expressing consent. At the beginning of the process, the indigenous and tribal peoples will indicate to the stake-holder forum how they will express their consent to decisions. A final agreement on how to express consent will be reached before the start of the planning process.

220 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

Strategic Priority 2 Comprehensive Options Assessment

Key Message

Alternatives to dams do often exist. To explore these alternatives, needs for water, food and energy are assessed and objectives clearly defined. The appropriate development response is identified from a range of possible options. The selection is based on a comprehensive and participatory assessment of the full range of policy, institutional, and technical options. In the assessment process social and environmental aspects have the same significance as economic and financial factors. The options assessment process continues through all stages of planning, project development and operations.

Effective implementation of this strategic priority depends on applying these policy principles:

2.1 Development needs and objectives are economic and financial factors in assess- clearly formulated through an open and ing options. participatory process before the identifica- 2.4 Increasing the effectiveness and tion and assessment of options for water sustainability of existing water, irrigation, and energy resource development. and energy systems are given priority in 2.2 Planning approaches that take into the options assessment process. account the full range of development 2.5 If a dam is selected through such a objectives are used to assess all policy, comprehensive options assessment institutional, management, and technical process, social and environmental options before the decision is made to principles are applied in the review and proceed with any programme or project. selection of options throughout the 2.3 Social and environmental aspects are given detailed planning, design, construction, the same significance as technical, and operation phases.

Rationale Often dams take a long time to come on Dams have delivered benefits to society, but stream, delaying the delivery of benefits. have also caused serious social and environ- Because they are high cost investments they mental harm. Many of the controversies divert resources and can exclude other options over dam projects have focused attention on that may be able to deliver benefits more whether a dam was the most appropriate quickly. These options include response to a development need or objec- demand side management, tive, and whether these were correctly alternative supply side technolo- Options assessment identified in the first place. In some cases gies and improving and expanding involves determining the project objectives were not clearly stated, the performance of existing relevance of individual particularly in relation to broader national systems. There are also some new options or a mix of and local development goals. In others, the options reaching the stage where options to respond to decision to proceed with a dam was taken they can compete in the market, development needs in a before considering all options or following for example renewable technolo- specific location. strong backing from specific constituencies gies for electricity generation such that undermined options assessment. This as wind and solar power. failure to assess strategic options rigorously at an early stage has led to a number of Options assessment involves determining the disputes. relevance of individual options or a mix of

The Report of the World Commission on Dams 221 Chapter 8

options to respond to develop- Strategic Priority 1 presents a new perspec- An early focus on ment needs in a specific location. tive on identifying development needs based options assessment can The challenge is to assess a wider on recognising rights and assessing risks. It reduce delays and range of alternatives earlier in the integrates the planning function of govern- additional costs and process. Experience has shown ments in the water and energy sectors with conflicts, benefiting all that this needs to be done in a local processes to determine needs. This is those affected by a transparent and participatory consistent with a move towards a more project. manner ensuring that human, strategic planning process that identifies social, environmental, technical options to meet expressed needs. and financial considerations get equal weight in the final decision. The increased availabili- National policy statements on water re- ty of information about the expanding range sources, agriculture, energy and the environ- of alternatives provides a sound base from ment should embody guiding principles that which to draw. facilitate a more open process of needs assessment. Policy formulation should be a An early focus on options assessment will participatory process that lays the founda- exclude most questionable projects. Those tion for the involvement of affected groups that emerge will enjoy wider public support throughout later stages of needs and options and legitimacy. It can reduce delays and assessment. additional costs and conflicts, benefiting all those affected by a project. In addition to Effective participation depends on locally social and environmental advantages, appropriate processes that define the form of increased investment in options assessment participation and the method for consolidat- can result in long term economic and ing needs identified at local, sub-national financial benefits. and national level. Institutions or bodies representing communities should be clearly The outcome may not be as simple as ‘build defined. Strategic Priority 5 discusses other a dam’ or ‘do not build a dam’, but could be key attributes of participation. The needs a set of parallel and complementary inter- assessment will provide a framework for ventions that together meet the defined assessing options and linking expressed goals. Where a large dam is selected, there needs to development objectives for specific are a number of options within the project beneficiary groups. that can avoid, minimise and mitigate adverse social and environmental impacts. 2.2 Planning approaches that take into These options relate to altering the size and account the full range of development location of the project and designing objectives are used to assess all policy, appropriate operating rules. institutional, management, and technical options before the decision is made to proceed with any programme or Elaboration of the Policy project. Principles Once the planning process has clearly 2.1 Development needs and objectives are defined needs, development objectives and clearly formulated through an open intended beneficiaries, it will need mecha- and participatory process before the nisms to assess the appropriateness of identification and assessment of options for water and energy resource options and for the participation of stake- development. holder groups. Assessing options should start

222 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

early in the planning process and can be objectives. The reasons for rejecting options incorporated into master plans and sector should be clear to stakeholders. plans using strategic impact assessments and other planning tools. Comprehensive 2.3 Social and environmental aspects are options assessment must precede selection of given the same significance as technical, economic and financial factors in any specific development plan, whether it assessing options. includes a dam or an alternative. Future decision-making must increase the The range of options being examined at the significance of social and environmental outset will be broad and go beyond techni- considerations, bringing them to the fore- cal alternatives to consider relevant policy, front of the screening process as is already programme and project alternatives. It the practice in some countries. should also consider: The focus must shift from mitiga- Avoidance and ■ institutional changes and management tion and compensation to make minimisation of social reforms that could influence consump- avoidance and minimisation of and environmental impacts tion patterns, reduce demand, and affect social and environmental impacts must become fundamental the viability of other supply options; fundamental criteria guiding criteria guiding options ■ the river basin context, cumulative options assessment. This ap- assessment. impacts and interactive effects, including proach will give society a better the interaction between surface and chance to set thresholds for what groundwater resources; is acceptable and what is not, to ■ multipurpose functions of alternatives; consider long term priorities, and to reject options that are unlikely to meet avoidance ■ secondary local and regional develop- and minimisation principles. Stakeholders ment effects of alternatives; must agree on guiding principles to mitigate ■ subsidies that can distort comparison of and compensate for the social and environ- alternatives; mental consequences of options that remain ■ life cycle analysis to compare electricity on the table before taking further decisions. generation alternatives; and ■ the gestation period required before Environmental issues needing consideration benefits are delivered. include impacts on natural ecosystems and water quality and the implications of the A major consideration in selecting options different options for local, regional and is assessing institutional capacity for imple- transboundary effects. For example recent mentation. If capacity is weak for a particu- research shows that some reservoirs emit lar option, and strengthening measures or greenhouse gases. With climate change external support are not viable, then the emerging as a key factor in decisions on option should be rejected. energy options, reducing greenhouse gas emissions and maintaining climate stability Multi-criteria analysis is a mechanism for requires a concerted global response. options assessment. Selection criteria used in the analysis must explicitly reflect how Each case is location specific and informed each option affects the distribution of costs, decisions need an enhanced local knowl- benefits and impacts for each stakeholder edge base on social and environmental group and how it responds to development factors. Requirements include:

The Report of the World Commission on Dams 223 Chapter 8

■ social and ecosystem baseline studies at In the irrigation sector, enhancing existing an early stage to describe existing condi- systems by fulfilling undeveloped potential tions and resource endowment; and increasing the productivity of water ■ determination of the relative weighting of offers the best alternative to new construc- environmental and social aspects in tion. However, improving existing systems relation to technical, economic and does not necessarily help to address the financial aspects through an open process; needs of the poorest sections of society. The ■ a strategic impact assessment to deter- options assessment process needs to consider mine environmental, social, health and alternative means to increase livelihood cultural heritage impacts of alternatives opportunities and local food security. This and reject inappropriate alternatives at should include an objective assessment of an early stage; and the potential for local community based projects and other alternative or comple- ■ explicit assessment of future net green- mentary measures. house gas emissions of a project.

2.5 If a dam is selected through such a 2.4 Increasing the effectiveness and comprehensive options assessment sustainability of existing water, irriga- process, social and environmental tion and energy systems are given principles are applied in the review and priority in the options assessment selection of options throughout the process. detailed planning, design, construction, and operation phases. Planning must give priority to making existing water, irrigation, and energy systems Following a decision to proceed with a dam more effective and sustainable before taking project, decisions must be taken to deter- a decision on a new project. The potential is mine its precise location, alignment and highly location specific, therefore assess- height, the availability and sources of ment will require detailed in-country construction materials, the impact of the reviews that cut across sectoral boundaries construction process, the operational and go beyond technical responses to characteristics of the proposed dam, and the include consideration of policy options. The details of water and power distribution management of existing water and energy systems. Each of these decisions has further systems will require a more pro-active and sets of alternatives. The process adopted for integrated response in order to achieve these selecting alternatives requires the same gains. Strategic Priority 3 covers services multi-criteria approach proposed for the provided by existing dam projects which are earlier stages of options assessment. It must a subset of existing water and energy sys- give due prominence to social and environ- tems. mental considerations and to participatory processes for decision-making. Principles The energy sector can apply a range of agreed during the initial screening of measures to encourage more efficient options remain relevant when deciding on production, lower distribution losses and options relating to the project development reduce consumption. Similar opportunities and operations phases. Strategic Priorities 1, exist to use alternative supply sources and 4 and 5 provide further guidance on these conservation measures to provide water matters. supplies for disadvantaged communities.

224 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

Strategic Priority 3 Addressing Existing Dams

Key Message Opportunities exist to optimise benefits from many existing dams, address outstanding social issues and strengthen environmental mitigation and restoration measures. Dams and the context in which they operate are not seen as static over time. Benefits and impacts may be transformed by changes in water use priorities, physical and land use changes in the river basin, technological developments, and changes in public policy expressed in environment, safety, economic and technical regulations. Management and operation practices must adapt continuously to changing circumstances over the project’s life and must address outstanding social issues.

Effective implementation of this strategic priority depends on applying these policy principles:

3.1 A comprehensive post-project monitoring and developed with affected communities to evaluation process, and a system of longer- remedy them. term periodic reviews of the performance, 3.4 The effectiveness of existing environmental benefits, and impacts for all existing large mitigation measures is assessed and dams are introduced. unanticipated impacts identified; opportuni- 3.2 Programmes to restore, improve and optimise ties for mitigation, restoration and enhance- benefits from existing large dams are identi- ment are recognised, identified and acted fied and implemented. Options to consider on. include rehabilitate, modernise and upgrade 3.5 All large dams have formalised operating equipment and facilities, optimise reservoir agreements with time-bound licence operations and introduce non-structural periods; where re-planning or relicensing measures to improve the efficiency of delivery processes indicate that major physical and use of services. changes to facilities or decommissioning, 3.3 Outstanding social issues associated with may be advantageous, a full feasibility study existing large dams are identified and and environmental and social impact assessed; processes and mechanisms are assessment is undertaken.

Rationale maintenance, modernisation and renovation Most large dams that will operate in the 21st due to institutional or financial limitations. century already exist. A number of countries In many settings owners have not done have not realised the full benefits of existing systematic assessments of opportunities for large dams because of: optimising or expanding facilities to improve the services existing dams provide. ■ incomplete investments in delivery systems; Opportunities to improve the efficiency, ■ lack of integration with associated environmental and social performance of systems such as local and national grids existing dams and optimise their benefits and agricultural extension services; must be taken. One of the most striking ■ lack of equity consideration in allocation features is the persistence of social and of project benefits; environmental problems arising from past ■ poor maintenance; and projects. Often promises of compensation ■ ineffective and outdated management. and other benefits like local power supply and social amenities for resettled and host In other cases, dam owners have not made communities have not been kept. In many regular investments in monitoring, ongoing cases such promises were informal, making

The Report of the World Commission on Dams 225 Chapter 8

compliance more difficult to Finally, it is evident that many existing Retrofitting existing achieve. Governments, industry dams do not have formal operating agree- dams with more and dam owners also recognise, ments, licences or concessions, particularly efficient, modern often informally, that past in the case of publicly owned irrigation and equipment and control mistakes should not be repeated; water supply dams. Wider participation in systems has achieved yet they remain as an unresolved important management and operational significant legacy. decisions requires clear procedures and improvements in supporting legal mechanisms, especially benefits, extending The WCD Knowledge Base when such decisions transform or transfer facilities and provides many examples where benefits and impacts. The absence of optimising operations. the services provided by older licences or formalised agreements removes dams have been restored or the opportunity for public input and ac- extended in time. In many other countability. Where they do exist, licences cases retrofitting existing dams with more and other agreements often lack clear efficient, modern equipment and control performance targets, limiting public partici- systems has achieved significant improve- pation at re-licensing reviews. ments in benefits, extending facilities and optimising operations. 3.1 A comprehensive post-project monitoring and evaluation process, and a system of longer-term periodic reviews The recent trend to optimise reservoir of the performance, benefits, and operations for new and older dams using impacts for all existing large dams are decision support systems backed by more introduced. accurate and timely data on river flows is particularly relevant. While opportunities The WCD Knowledge Base shows that must be assessed on a case-by-case basis, historically, few comprehensive post-project good practice is to consider such measures as evaluations have taken place after the a ‘new supply option’ where they commissioning of large dams. This applies to present significant, cost effective virtually all regions and countries. With few opportunities. exceptions, there has been little or no monitoring of the physical, social and Experience shows that, condi- environmental effects of dams, a necessary tions permitting, this approach input for such evaluations. Where post- can increase hydroelectric project assessments have been undertaken, benefits by 5 to 10% over rule- they have occurred many decades after based operating criteria without construction usually with a narrow technical adversely affecting other water focus and little input from stakeholders. uses. This is a trend in Canada, the United States and Europe where opera- The WCD Knowledge Base reveals that tors are seeking full benefits from existing many unforeseen technical, social and assets in response to power sector deregula- environmental issues emerge during the tion and competition. In some cases, opti- commissioning phase and the first few years mising operations of a system of dams can of operation. More intensive monitoring, postpone the need for new projects. These extending from the construction phase experiences are not confined to developed through the first few years of operation, countries.5 followed by a comprehensive post-project

226 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

evaluation after 3 to 5 years involving and stakeholders who will participate in the affected stakeholders, will help to identify evaluations and set out the resources and and resolve many early problems. The means for stakeholder input and interaction. evaluation will encourage compliance with Upgrading monitoring capacity will pose all commitments and provide a milestone to challenges in many countries due verify public acceptance. The first post- to the costs and operation of project evaluation should help confirm and instruments and data systems, The first post-project strategically focus the longer-term monitor- and because agencies other than evaluation should ing programmes and provide ‘lessons learnt’ the dam owners and operators provide ‘lessons learnt’ for future decisions about planning, design may be involved. Clear responsi- for future decisions and operations of the dam. bilities that build on existing about planning, design capacities need to be defined and and operations of the Because the economic life of a dam may financial resources provided. dam. span many generations, it is necessary to Dam operators and the agencies review the project operation periodically in involved should publish monitoring results light of the needs it is intended to meet, and annually, and make results freely accessible the services it can provide. These periodic to all stakeholders. evaluations at intervals of 5 to 10 years should be comprehensive, integrated, 3.2 Programmes to restore, improve and cumulative and adaptive. Where dams are optimise benefits from existing large dams are identified and imple- part of a larger river basin and regional mented. Options to consider include development scheme, the evaluations rehabilitate, modernise and upgrade should take into account basin-level evalua- equipment and facilities, optimise tions of all project and programme compo- reservoir operations and include non- structural measures to improve the nents linked to the dam that affect the efficiency of delivery and use of environment and society (see Strategic services. Priority 4). Many industrialised countries are focusing Enabling conditions for evaluations are on rehabilitation and modernisation to context specific and measures should build restore or extend the economic life of on existing capacities. For many existing existing dams. In the last decade, many dam dams this will be the first evaluation of this owners have implemented techniques for nature and institutional resistance to optimising reservoir operations, especially transparency may need to be overcome. for dams generating electricity. They are Licensed private sector operators may regard considering other measures to improve some aspects of the operation as proprietary performance and safety such as increasing commercial information. An essential first spillway capacity to handle higher floods, step is for governments, or their regulatory extending reservoir storage and improving agencies, to clearly specify the requirements sediment flushing techniques. While new for monitoring and evaluation in the supply options may be needed in many appropriate regulations, project licences and countries, restoring or extending the life of operating agreements. existing dams and, where feasible, expanding and improving services from existing Government guidelines need to clearly dams provide major opportunities to address define roles of dam owners and operators development needs.

The Report of the World Commission on Dams 227 Chapter 8

The WCD Knowledge Base identified three surface irrigation systems attached to large general categories of improvement: dams. Other non-structural tariff, institu- ■ modernising and upgrading equipment and tional and management practices can controls, and rehabilitating or expanding improve the efficiency of the irrigation and facilities associated with the dam; water supply services provided by existing dams but will require sectoral initiatives ■ optimising operation of existing reser- that may be beyond the mandate of a dam voirs including daily and seasonal water operator. levels and release patterns for single or multi-purpose uses such as flood manage- Improving performance begins with assessment and hydro generation. This can be ing each dam for potential gains from done for a single dam, or in co-ordina- modernisation, renovation, expansion or tion with other reservoirs, lakes or water optimisation of operations. Other gains can course diversions regulating river flow in come from investment in necessary hydro- a basin; and logical monitoring equipment, computer ■ optimising the role of the dam within software, and the preparation of basin and the larger system it services. For example system-level optimisation plans. optimising the use of surface and ground water inputs in agriculture where water is This must be explicitly linked with the a limiting factor, or using load manage- options assessment phases of planning, ment practices to optimise the co- clearly showing the scope for improvements ordination of hydro generation with to existing dams. The public should have other energy sources. the opportunity to comment on a survey assessing improvement opportunities for all The potential for increasing benefits from a dams. This should be followed by more particular dam, or group of dams, in a basin detailed assessments of the specific dams depends on the specific circum- which have potential for significant stances. Opportunities in all improvements. Restoring or extending three areas noted above should the life of existing be considered. Depending on 3.3 Outstanding social issues associated dams and, where the situation the potential may with existing large dams are identified feasible, expanding be considerable. and assessed; processes and mecha- and improving services nisms are developed with affected communities to remedy them. from existing dams Other measures have shown provide major potential to improve the per- In all its public consultations, dam-affected opportunities to formance of existing dams and communities told the Commission about address development the services they provide. For the ongoing problems, broken promises, and needs. example, experience is growing human rights abuses associated with the with flushing and sluicing involuntary resettlement and environmental practices during monsoon floods impacts from dams. The WCD Knowledge to reduce sedimentation and restore live Base includes significant evidence of un- storage in certain types of reservoirs. Accel- compensated losses, non-fulfilment of erating the pace of investment in secondary promised rehabilitation entitlements, and and tertiary canal systems and drainage can non-compliance with contractual obliga- significantly improve the productivity of tions and national and international laws

228 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

(see Strategic Priority 6 and Chapter 9). States are at different stages in In order to address While the Commission is not in a position developing regulatory systems reparation issues, the to adjudicate on these issues, it has suggest- and institutional capacity government should ed ways to redress past and ongoing prob- including dispute resolution and appoint an lems associated with existing dams. will take different approaches to independent resolving this issue. However, committee with the Existing international laws have articulated where there are reasons to take participation of legal a legal premise for a right to remedy, or action and alleviate hardships experts, the dam reparations which is also reflected in the experienced by the people owner, affected people national legislative frameworks of many affected by dams constructed in and other countries.6 Reparation is defined as actions the past, there are two practical stakeholders. or processes that remedy, repair, make ways of addressing pressing amends or compensate for past failures and problems. damages. Given the nature of damages ■ Opportunities to restore, improve and resulting from loss of land and a way of life, optimise benefits from existing large redress could include remedies that: dams and other river basin developments ■ recognise the breach of the original should be used as an entry point to address obligation and its consequences; unmitigated social problems associated ■ acknowledge claims; with the dams in that river basin. ■ assess damages; ■ In situations where no current develop- ■ assign responsibility; and ments are envisaged but outstanding ■ devise and implement remedial activities social problems related to dams exist, to repair the long-term and cumulative meritorious claims for redress should be impact of these failures. prioritised and assessed on the following basis: Assessing claims and making reparations The responsibility for initiating the process ■ affected people file genuine claims of reparation rests with government. The related to economic, social and cultural affected people may also file claims with the loses and unfulfilled promises; government. In order to address reparation ■ the evidence accompanying claims issues, the government should appoint an filed shows that they continue to independent committee with the participa- suffer harm due to unmitigated tion of legal experts, the dam owner, affect- impacts, and that the impact is ed people and other stakeholders. The causally connected to the dam; committee should: ■ available mechanisms to resolve the ■ develop criteria for assessing meritorious complaint have been exhausted; and claims; ■ the nature and extent of the harm. ■ assess the situation and identify individu- Enabling conditions als, families and communities fulfilling To exercise their right to seek a remedy, the criteria for meritorious claims and affected people need access to political and ■ enable joint negotiations involving legal systems and the means and ability to adversely affected people for developing participate in prescribed ways. Affected mutually agreed and legally enforceable people should receive legal, professional and reparation provisions. financial support to participate in the

The Report of the World Commission on Dams 229 Chapter 8

assessment, negotiation and implementation instruments. The World Bank group’s stages of the reparation process. inspection panel and the International Finance Corporation (IFC) / Multilateral Affected peoples must be defined according Investment Guarantee Agency (MIGA) to actual experience of impacts as described office of the Compliance Advisor / Ombuds- in Strategic Priority 5.2, and not by the man acknowledge the responsibilities of the limited definition in original financier to comply with specific regulatory project documents and con- and operational policies governing its To exercise their right tracts. Further, damage from operations. to seek a remedy, dams may require assessment on affected people need a catchment basis extending In a number of instances, efforts to assign access to political and upstream and downstream. corporate responsibility for non-compliance legal systems and the Damage assessments should or transgressions related to social and means and ability to include non-monetary losses. environmental elements of a project have participate in Reparations should be based on led to complaints filed in a corporation’s prescribed ways. community identification and home country. prioritisation of needs, and community participation in developing The roles and responsibilities of all parties compensatory and remedial strategies. involved in planning, financing, building and operating the dam must be clearly The nature of remedies Remedies can include restitution, indemnity established in the process of hearing and (or compensation), and satisfaction. Restitu- assessing a claim by an independent com- tion can include stopping the damaging mittee constituted by the government in conduct or carrying out the original obliga- consultation with the affected people and tion. Indemnity involves the payment of other stakeholders. money for losses incurred, such as payments Financing reparations and compensation to compensate for loss of assets, property, While financing reparations may pose and livelihoods and a variety of remedial significant challenges this should not nullify actions, including resettlement plans and legitimate claims. Priority must be given to development programmes. Satisfaction financing a negotiated reparation plan includes other forms of reparation to address before funding new dam projects in a any non-material damage, including public specific location or river basin in a country. acknowledgement of damage and an apology.

Responsibility for reparations Reparations can be financed with funds Reparations may involve multiple actors from national, provincial, and / or local including states, financing institutions, government budgets, a percentage of loans international organisations, and private and grants or a percentage on current corporations. It is the State’s responsibility income from energy and water management to protect its citizens, including their right projects. Such funds could be allocated to a to just compensation. However, internation- trust fund to benefit the community over al organisations party to foreign investment the long term (see Strategic Priority 6 for agreements also have obligations and more information on trust funds). Through responsibilities to the rights and duties changes in dam operations or other means, specified in the UN’s declarations and reparations can take the form of allocations

230 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

of non-monetary resources, including land, increasing salinity in the Countries such as water, fish and access to sacred sites. Murray Darling Basin in South Africa, Senegal Australia, or the new South and Cameroon have An independent committee should be Africa Water Act which introduced artificial empowered to collect, manage, and award reallocates water rights. At flood releases to reparations. To ensure that decisions con- least five countries (United maintain downstream form to the laws of the country and to States, Japan, Australia, Brazil, floodplains of value to international laws, such committees should and France) are assessing the local people. include legal representatives selected by efficiency of existing fish government and affected communities. passes and recommending Parties contributing to the fund should be improvements to design and represented to ensure transparent use of operation. The design of the Mohale dam in their funds. Accountability of the parties Lesotho has been modified to allow larger responsible for reparation should be ensured flows, in anticipation of the results of the through contracts and legal recourse. environmental flow studies currently being completed. 3.4 The effectiveness of existing environmental mitigation measures is assessed Many dams in the United States have also and unanticipated impacts identified; been modified to allow for larger flows. opportunities for mitigation, restoration and enhancement are recognised, Countries such as South Africa, Senegal and identified and acted on. Cameroon have introduced artificial flood releases to maintain downstream floodplains The impact of large dams upon natural of value to local people. This technique is ecosystems and biodiversity is a major used on the Columbia River system in the concern. In the past few decades, some United States to reduce problems with total countries have made considerable invest- dissolved gases that can kill valuable fish. ments to alleviate these impacts. Wide- spread concern remains that dams elsewhere There are a number of barriers and con- continue to result in significant, and even straints to overcome. Continuous monitor- unnecessary, negative impacts on a wide ing is a prerequisite to identify and assess range of natural ecosystems and on the what the actual impacts are and the possible people that depend on them. These ecosys- effect of mitigation and restoration meas- tems perform functions such as flood allevia- ures. Resources for implementing monitor- tion and yield products such as wildlife, ing must be integrated in the project cost. fisheries and forest resources. They are also Clear guidelines on environmental monitor- of aesthetic and cultural importance for ing and a response to deal with impacts are many millions of people.7 needed. Other constraints will have to be addressed, particularly for privately devel- A range of measures is available to enhance oped hydropower projects that involve long- and restore ecosystems from their man- term supply contracts negotiated on previ- modified state, and many are already in use ous release patterns. Contracts for a speci- worldwide. In many cases these efforts are fied time period may not anticipate or motivated by emergent environmental allocate responsibility for periodic changes constraints and changing community within the contract period, and would priorities, as in the case of efforts to combat require renegotiation.

The Report of the World Commission on Dams 231 Chapter 8

3.5 All large dams have formalised operat- multi-purpose dams with power compo- ing agreements with time-bound nents. Licences for private owners are time- licence periods; where re-planning or re-licensing processes indicate that bound and are reviewed at periods of 20 to major physical changes to facilities or 40 years, or more frequently, depending on decommissioning may be advanta- the country’s policy and regulations. There geous, a full feasibility study and are recent examples where public agencies environmental and social impact assessment is undertaken. and private owners alike have had to apply for licences for existing dams when new Many dams, particularly irrigation and water regulations were introduced. Such licences supply dams, do not have operating agree- identify activities which the existing own- ments or licences. Where they do exist they ers, including the government, need to are time-bound agreements. They generally comply with. They range from introducing set out the obligations of the public agency and reporting on monitoring programmes to or private entity operating the dam, and dam safety inspections. Harmonising the provide a legal basis for stakeholders to licence expiry dates for all dams in a partic- participate, in an open and transparent ular river basin may have benefits in some manner, in important decisions regarding cases, for example a cascade sequence of physical changes in facilities or reservoir dams where interactive effects and cumula- operations. tive impacts are a consideration.

Licences or operating agree- Where re-licensing processes are in place, There is a trend ments provide requirements for decommissioning may be an option. This towards formal physical, environmental and option usually arises where a dam has licensing of new and social impact monitoring, exceeded its useful life, where safety reasons existing dams operated contingency plans, operating make it less expensive to remove the dam by public and private strategies, the specific require- than to rehabilitate it, or where the costs agencies. ments for publication of moni- (including environmental costs) of further toring and operating results, and operation outweigh the benefits. Experience requirements for the periodic with decommissioning is growing in North needs and performance review of existing America and Europe.8 dams noted earlier. Given the ageing population of dams, safety issues require The effects of decommissioning, particularly more attention in the form of inspections, on the natural environment where ecosys- routine monitoring, evaluations, surveil- tem restoration is a young science, have yet lance systems, and regularly updated emer- to be determined. In some cases environ- gency action plans. Where practical and mental restoration has been well served by feasible, it is also important to update dams decommissioning, in others negative effects to contemporary standards, especially have been observed. Decommissioning of regarding spillway capacity and resistance to larger dams, particularly those with a high earthquakes. build-up of sediments in the reservoir may be problematic, especially if the dam is to be There is a trend towards formal licensing of removed and the sediments released down- new and existing dams operated by public stream. Removal of the dams may have and private agencies. This applies particular- many consequences on land use in upstream ly to single purpose hydropower dams and and downstream areas and on other socio-

232 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

economic values and should be subjected to environmental, social, technical and economic assessment in the same way that new dams are.

Based on the range of issues that may surface with decommissioning, a feasibility study should be carried out to select the overall best solution, considering economic, environmental, social and political factors.

The challenge in licensing is that many countries have insufficient legal frameworks and there may be no consensus on the appropriate level of regulation and licensing. New institutional arrangements may be necessary to formalise licensing operating dams may not have physical provisions to agreements in many countries. At present accommodate all the changes in regulation the institutional responsibility for different and meet all current standards. aspects of the operating agreement may be located in different institutions. Another barrier to overcome is that some existing

The Report of the World Commission on Dams 233 Chapter 8

Strategic Priority 4 Sustaining Rivers and Livelihoods

Key Message

Rivers, watersheds and aquatic ecosystems are the biological engines of the planet. They are the basis for life and the livelihoods of local communities. Dams transform landscapes and create risks of irreversible impacts. Understanding, protecting and restoring ecosystems at river basin level is essential to foster equitable human development and the welfare of all species. Options assessment and decision-making around river development prioritises the avoidance of impacts, followed by the minimisation and mitigation of harm to the health and integrity of the river system. Avoiding impacts through good site selection and project design is a priority. Releasing tailor-made environmental flows can help maintain downstream ecosystems and the communities that depend on them.

Effective implementation of this strategic priority depends on applying these policy principles:

4.1 A basin-wide understanding of the When reviewing alternative locations for ecosystem’s functions, values and require- dams on undeveloped rivers, priority is given ments, and how community livelihoods to locations on tributaries. depend on and influence them, is required 4.4 Project options are selected that avoid before decisions on development options significant impacts on threatened and are made. endangered species. When impacts cannot 4.2 Decisions value ecosystems, social and be avoided viable compensation measures health issues as an integral part of project are put in place that will result in a net gain and river basin development and prioritise for the species within the region. avoidance of impacts in accordance with a 4.5 Large dams provide for releasing environ- precautionary approach. mental flows to help maintain downstream 4.3 A national policy is developed for maintain- ecosystem integrity and community ing selected rivers with high ecosystem livelihoods and are designed, modified and functions and values in their natural state. operated accordingly.

Rationale Rivers and catchments Since the 1970s a growing understanding of Rivers and catchments are particularly impor- nature as the basis for long term human well tant because even relatively small changes in being has replaced the view that it is an array land-use, pollution or flows can have far- of replaceable inputs to the development reaching effects. These often extend thousands process. The 1992 Earth Summit in Rio de of kilometres downstream to the deltas of Janeiro, Brazil established the major rivers and even to the coastal and critical link for all countries marine environments. Mounting global The 1992 Earth Summit between a healthy environment pressure on water resources, and the fragmen- in Brazil established the and economic development, tation by dams of 60% of the world’s rivers, critical link for all refuting the idea that this is only a makes these ecosystems an urgent global countries between a luxury for rich countries. Subse- concern.9 healthy environment and quently 177 countries in the world economic development, have accepted, approved, or Dams disrupt the existing pattern of water refuting the idea that this acceded to the Biodiversity use and reallocate water to new uses. Where is only a luxury for rich Convention and 122 to the water is abundant, or where few people countries. Ramsar Convention on Wetlands. depend on fish, floodplains or deltas for

234 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

their livelihoods, this process can often have ratified the Convention on Biodiversi- proceed smoothly. Where water is scarce, ty and the Ramsar Convention on Wetlands heavily used or supports economically or can use the provisions and guidance of these socially important ecosystems, this redistri- conventions to assist in meeting sustainabil- bution can lead to irreversible impacts, ity objectives. losses, inequities and conflicts. Changes to Weighing negative and positive impacts river flow, leaching of toxic elements from Dams can reduce the risks of flooding. the reservoir bottom, and the creation of However, especially in many developing new habitats for disease vectors in slow countries, reduced flooding may also in- flowing irrigation and drainage canals can crease risks for local people by adversely all impact negatively on human and ecologi- affecting downstream fisheries, grazing and cal health. crop production. In all cases, negative Equitable water use economic and social impacts must be River water is a common resource that weighed equally against positive impacts. should serve the good of all riverine inhabitants and the environment in an equitable In the past, dam construction has caused and sustainable manner. Water reallocation environmental damage due to poor assess- through dams should explicitly take account ments, vested interests, lack of knowledge, of existing uses, and of the species and lack of enforcement of mitigation measures, ecosystems the water supports. Many of the lack of sufficient resources, lack of ongoing broader values that rivers provide are too monitoring or ignorance of ecosystem often ignored in project planning and functions. Research will contin- appraisal. The natural resources associated ue to improve knowledge and with rivers directly support natural habitats understanding of ecosystem The natural resources and the livelihoods and cultural values of complexity. However it is associated with rivers millions of people worldwide. Rivers may difficult to mitigate all ecosys- directly support natural also hold deep spiritual meaning for commu- tem impacts and ecosystem habitats and the nities and societies. responses are rarely fully pre- livelihoods and cultural dictable. A multi-layered values of millions of The State must use effective mechanisms for approach is needed that priori- people worldwide. sharing the available water between users, tises avoidance, especially in bearing in mind the balance between sensitive areas, and has in-built different uses – irrigation, water supply, checks that adapt and respond to observed power generation, and ecosystem – and the ecosystem changes. livelihood and quality of life needs of riverine communities. The policy principles presented here provide a framework for the range of measures The construction and operation of many needed to ensure protection and health of large dams have had significant, and often ecosystems in planning, construction and irreversible, effects on many rivers, riverine operation of dams and their alternatives. No ecosystems and communities. These activi- single principle can be fully effective in ties threaten the sustainability of the isolation from the others nor can a single underlying ecological processes that main- ministry or agency be responsible for them all. tain habitat and biodiversity. Countries that Ecosystem issues are best addressed through a

The Report of the World Commission on Dams 235 Chapter 8

holistic view of the river, with all actors 4.2 Decisions value ecosystems, social incorporating an ecosystem approach into and health issues as an integral part of project and river basin develop- their planning, operations and monitoring. ment and prioritise avoidance of impacts in accordance with a precau- Elaboration of Policy Principles tionary approach.

4.1 A basin-wide understanding of the Large dam projects have frequently incorpo- ecosystem’s functions, values and rated environmental and social considera- requirements, and how community tions too late in project planning to allow livelihoods depend on and influence them, is required before decisions on their full integration into project choice and development options are made. design. Environmental Impact Assessments (EIA) are often seen simply as an obstacle to The river basin is the natural geographical be overcome in getting clearance and unit to assess impacts on ecosystems and approval for the project. Consultants and livelihoods. agencies involved in planning should focus on ecosystem, social and health issues at the Effective avoidance, same time that economic and technical The flow of water links minimisation and mitiga- studies for options assessment begin. Planning riverine ecosystems, tion of negative environ- teams should explicitly incorporate ecosystem, establishing a continuum mental impacts from large health and social findings in the final choice from the top of the dams and their alternatives of project through multi-criteria analysis. catchment to the ocean. require good baseline Upstream water resource information and scientific Strategic impact assessment during the developments cannot be knowledge of the riverine options assessment stage should include separated from their ecosystem, gathered over independent and comprehensive assessment downstream implications. several years. The flow of of ecosystem, social and health impacts and water links riverine ecosys- evaluation of any cumulative or inter-basin tems, establishing a contin- impacts. An independent panel could be uum from the top of the catchment to the used to support impact assessment. ocean. Upstream water resource developments cannot be separated from their Alternatives to large dams are frequently downstream implications. available, and negative impacts of large dam projects vary hugely due to site selection Considering the ecosystem and the design of elements such as its Therefore, project proponents must assess height, intakes, outlets and gates. By delib- the ecosystem consequences of the cumula- erately setting out to avoid projects with the tive impact of dams, dam induced develop- most serious and damaging negative im- ments and other options along the full pacts, project outcomes will prove more length of the river reaching as far as the sustainable and acceptable in future. delta, even where this extends into neighbouring provinces or countries. Where the The precautionary approach resources of riparian communities could be The precautionary approach requires States negatively affected they should be consulted and water development proponents to on the proposal before decisions are made exercise caution when information is (see Strategic Priorities 5 and 7). uncertain, unreliable, or inadequate and

236 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

when the negative impacts of actions on the on rivers by setting aside certain reaches, or environment, human livelihoods, or health entire rivers, from development. For exam- are potentially irreversible. It therefore ple, by 1998 the United States had designat- forms part of a structured approach to the ed 154 rivers under the ‘Wild and Scenic analysis of risks arising from water and other Rivers Act’, covering 17 200 km of a total development proposals. The precautionary 5.6 million kilometres of rivers in the approach is also relevant to risk manage- country. Sweden has set aside four entire ment. Determining what is an acceptable rivers from hydropower development, and level of risk should be undertaken through a Norway has ‘protected’ 35 % of its hydro- collective political process. The process power potential from development. Zimba- should avoid unwarranted recourse to the bwe has passed similar legislation for river precautionary approach when this can specific protection orders. President Kim overly delay decision-making. However, Dae Jung of Korea cancelled a decision-makers faced with scientific uncer- dam on the Tong River in June Determining what is an tainty and public concerns have a duty to find 2000 and declared the area acceptable level of risk answers as long as the risks and irreversibility protected for its natural and should be undertaken are considered unacceptable to society. cultural values. through a collective political process. A precautionary approach therefore entails This approach helps to recon- improving the information base, performing cile development of the risk analysis, establishing precautionary natural resource base with the need to thresholds of unacceptable impacts and risk, maintain genetic diversity as a potential and not taking actions with severe or resource for human development in fields irreversible impacts until adequate informa- like medicine. It also respects the intrinsic tion is available, or until the risk or irrevers- value of rivers. To put this policy into ibility can be reduced, making outcomes practice, the State, appropriate research more predictable. Normally the burden of institutions, communities and NGOs must proof will be on the developer. gather the essential baseline information to inform the policy from a holistic national 4.3 A national policy is developed for maintaining selected rivers with high perspective. Gathering this information ecosystem functions and values in allows the country to select the least envi- their natural state. When reviewing ronmentally damaging options whenever alternative locations for dams on dams emerge as the best option in the undeveloped rivers, priority is given to locations on tributaries. planning process. It enables the country to combine river development with river States should have a policy that excludes protection, achieving a balanced outcome major intervention on selected rivers to that satisfies the objective of sustainable preserve a proportion of their aquatic and development at national and local levels. riverine ecosystems in a natural state. The policy should be an integral part of the Floodplain ecosystems and migratory fish are overall national water policy. frequently associated with the lower stretches of the main-stems of rivers and their A growing number of countries have re- deltas. They are often particularly important sponded to increasing development pressure resources for local people, especially in

The Report of the World Commission on Dams 237 Chapter 8

developing countries. In many cases main- international importance of wetlands or stem dams have negatively affected the developing national wetland policies. Most already fragile livelihoods of downstream States have ratified the UN Convention on riverine communities. Priority under the Biological Diversity and the Ramsar Con- policy should therefore be given to alterna- vention on Wetlands and they should tive sites on the tributaries where impacts urgently apply their guidelines. This will are likely to be less than on the main-stem. help to avoid negative impacts on rare and threatened species. States that have not yet 4.4 Project options are selected that ratified the Conventions are encouraged to avoid significant impacts on threat- do so, and in the meantime to respect their ened and endangered species. When provisions. impacts cannot be avoided viable compensation measures are put in Compensation plans place that will result in a net gain for Where significant impacts on threatened or the species within the region. endangered species are considered unavoida- Too often, planned measures to mitigate ble, after exhausting other water and energy options and other dam project options, impacts on threatened or endangered project authorities should put in place a species have been inadequate. Failure is credible and monitored compensation plan. typically due either to insufficient knowl- This should ensure that the population edge and understanding of the ecosystems status of the species within the region shows concerned, inadequate plans, a net gain that adequately compensates for or lack of institutional and loss of habitat to the project. Such compen- The project authorities regulatory capacity to sation can include protecting other habitats, will finance enforce environmental restoring the species in other locations and compensation as an mitigating measures. captive breeding programmes. The project integral part of project authorities will finance compensation as an costs for the life span of Dam proponents have often integral part of project costs for the life span the project. over-confidently assumed of the project. If monitoring indicates that mitigation measures will compensation is not effective, then addi- work, rather than making tional measures will be required. them work. Responsibility for their imple-

mentation may be spread across many 4.5 Large dams provide for releasing actors. This situation has contributed to a environmental flows to help maintain significant increase in the rate of extinction downstream ecosystem integrity and of species sharing our planet. community livelihoods and are designed, modified and operated Respecting international guidelines to accordingly. reduce impacts Governments have often agreed interna- Dam owners have often viewed releasing tional guidelines on environmental issues, water from a dam for purposes other than yet the record of implementation is poor. power generation or water supply as a waste The existing international conventions of a valuable resource. Some agencies have contain agreed legal provisions for biodiver- declared it their aim to prevent a single drop sity protection and sustainable development of water from reaching the sea. Yet twenty- issues, including measures for conserving the nine countries use environmental flow biodiversity of inland waters, assessing the releases (EFR) from dams to maintain a

238 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

sustainable balance between the purpose of necessary to ensure sustaina- the dam and the needs of downstream bility of the nation’s water ecosystems and resource users. Eleven resources. Therefore the Act countries are considering implementing designates a ‘reserve’ that EFRs.10 This reflects the growing realisation must be identified before any that water and rivers support many complex authorisation of water processes that must be maintained to abstraction for other purpos- achieve sustainable development. Dams es. This basic reserve con- should now be specifically designed to tains the minimum quantity release the necessary flow of good quality and quality of water required water. Targeting particular ecosystem to satisfy basic human needs, outcomes increasingly results in flow releas- protect aquatic ecosystems es that go beyond the historical notion of a and secure the sustainable development and ‘minimum release’, often arbitrarily fixed at use of the water resource in question. The 10% of mean annual flow. A minimum Act opens the way to establishing the release may serve to keep the river wet but it environmental flows required to maintain may not be an ecologically effective meas- ecosystems. ure. Locally driven processes to Where downstream livelihoods depend on establish the objectives of envi- floodplains the release may take the form of ronmental flows will lead to Locally driven a managed flood. Senegal, South Africa and improved and sustainable out- processes to establish Cameroon all operate dams to flood valua- comes for rivers, ecosystems and the objectives of ble downstream floodplains that benefit the riverine communities that environmental flows rural communities. These managed floods depend on them. Ecosystem will lead to improved maintain hundreds of thousands of hectares responses to dam operating and sustainable of grazing land and important fisheries. regimes are variable, so dam outcomes for rivers, owners should undertake regular ecosystems and the Legal measures are often required to enable monitoring and a five yearly riverine communities implementation of environmental flows. evaluation of environmental that depend on them. One example is the new South African performance. This evaluation Water Act. This Act recognises that the should inform modification of ultimate aim of water resource management environmental flows where necessary (see is to achieve sustainability for the benefit of Chapter 9). all users and that the protection of the quality and quantity of water resources is

The Report of the World Commission on Dams 239 Chapter 8

Strategic Priority 5 Recognising Entitlements and Sharing Benefits

Key Message

Joint negotiations with adversely affected people result in mutually agreed and legally enforceable mitigation and development provisions. These provisions recognise entitlements that improve livelihoods and quality of life, and affected people are beneficiaries of the project. Successful mitigation, resettlement and development are fundamental commitments and responsibilities of the State and the developer. They bear the onus to satisfy all affected people that moving from their current context and resources will improve their livelihoods. Accountability of responsible parties to agreed mitigation, resettlement and development provisions is ensured through legal means, such as contracts, and through accessible legal recourse at national and international level.

Effective implementation of this strategic priority depends on applying these policy principles:

5.1 Recognition of rights and assessment of transmission lines and resettlement risks is the basis for identification and developments. inclusion of adversely affected stakeholders in joint negotiations on 5.3 All recognised adversely affected people mitigation, resettlement and development negotiate mutually agreed, formal and related decision-making. legally enforceable mitigation, resettlement and development entitlements. 5.2 Impact assessment includes all people in the reservoir, upstream, downstream and in 5.4 Adversely affected people are recognised catchment areas whose properties, as first among the beneficiaries of the livelihoods and non-material resources are project. Mutually agreed and legally affected. It also includes those affected by protected benefit sharing mechanisms are dam related infrastructure such as canals, negotiated to ensure implementation.

Rationale tion for losses. As a result dam projects have often impoverished adversely affected In the past, dams have displaced people people. from their habitats and livelihoods without giving them any control over alternatives. A negotiated process to identify those Besides those whose land and homes were affected and develop legally enforceable inundated, the adversely affected people mitigation and development measures included natural resource dependent river- depends on a number of enabling condi- ine communities living upstream and tions. Adversely affected people need to downstream of the dam and in show acceptance of the dam project by other affected areas. Because consenting to the process and to the mitiga- Recognition of rights is some groups were not recognised tion and development measures. These an important element as affected, the number of measures should include a share in project in establishing the adversely affected people has benefits and redress and recourse mecha- existing entitlements of been underestimated. Non- nisms. adversely affected recognition, or partial recogni- people at various tion, of the entitlements of The policy principles in this strategic locations. those identified as affected has priority relate closely to those in Strategic resulted in inadequate restitu- Priority 1 and Strategic Priority 4 and should

240 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

be read in conjunction with those strategic The impact assessment studies must identify priorities. and delineate various categories of adversely affected people in terms of the nature and Elaboration of Policy Principles extent of their rights, losses and risks. Socio- economic, demographic and health bench- 5.1 Recognition of rights and assessment mark surveys of all adversely affected of risks is the basis for identification populations must be completed and publicly and inclusion of adversely affected stakeholders in joint negotiations on reviewed prior to drafting mitigation, mitigation, resettlement and develop- resettlement and development plans. ment related decision-making. This process will facilitate assessment of the Recognition of rights is an important element actual magnitude, spread and complexity of in establishing the existing entitlements of impacts and implications for people depend- adversely affected people at various locations. ent on riverine ecosystems. This signals a Existing entitlements are the basis for negoti- departure from the way that social impacts ating new entitlements. The project process were assessed in the past and will empower recognises a range of entitlements including the planners and stakeholders to incorporate the entitlement of affected parties to: the full extent of social impacts and losses in ■ participate in negotiating the outcomes the decision-making process. It will achieve of the options assessment process; a holistic approach to social impacts arising ■ participate in negotiating the implemen- from dams and their infrastructure in river tation of the preferred option and basin contexts. Such an approach is applica- ■ negotiate the nature and components of ble to all options, dams and their alterna- mitigation and development entitlements. tives, and will create a level playing field in the options assessment process. Comprehensive assessment of the nature 5.3 All recognised adversely affected and extent of risks implied by a project people negotiate mutually agreed, allows for accurate assessment of the socio- formal and legally enforceable mitiga- economic conditions and the cultural context tion, resettlement and development of the potentially affected people. The socio- entitlements. economic, cultural, political and health In order to enable all categories of affected impacts must be identified through a number people identified in 5.2 to regain and of assessment methods such as Social Impact improve their livelihoods and welfare, Assessment (SIA), Health Impact Assessment mitigation and resettlement measures should (HIA), impoverishment risk analysis and be considered as a development opportunity cultural heritage impact assessment with focusing on a number of fundamental inputs: active participation of the affected people. ■ compensation for lost assets through 5.2 Impact Assessment includes all people replacement, substitution, cash and in the reservoir, upstream, downstream allowances; and in catchment areas whose proper- ■ ties, livelihoods and non-material livelihood restoration and enhancement resources are affected. It also includes in the form of land-for-land options; those affected by dam related infra- ■ sustainable non-agricultural employment structure such as canals, transmission lines and resettlement developments. and other measures;

The Report of the World Commission on Dams 241 Chapter 8

■ a share in project benefits and other ■ resettlement with land-for-land, sustain- development measures; and able non-agricultural employment and / ■ access to primary services such as school- or other development provisions; ing and health care. ■ resettlement as a community or viable social unit; For compensation to create effective new ■ resettlement close to the original habitat entitlements, customary and legal rights, for effective community recovery; and the future value of land and common ■ protection against land alienation to property resources must all be acknowl- secure negotiated entitlements; and edged. To reach agreement and prevent ■ other livelihood support measures future disputes, a transparent and legally including: agricultural support; access to enforceable mechanism must exist to forests, grazing lands and other common calculate the replacement resources; access to livelihood resources value of all affected assets. To Regaining lost livelihood for all adult members including women; check against under-valuation requires adequate lead and livelihood resources with the poten- of assets, the date for calculat- time and preparation and tial to sustain future generations. ing the value of assets must therefore people must match the date of payment. be fully compensated Implementing a mitigation, resettlement before relocation from and development plan their land, house or In the past, cash compensation Stakeholders must set up a high level multi- livelihood base. has proved ineffective in re- stakeholder committee representing govern- creating lost assets and oppor- ment, the developer and affected communi- tunities in less monetised ties. The committee will be responsible for economies and should be avoided. Where directing implementation of the mitigation, people prefer cash compensation, it must be resettlement and development programme paid with adequate safeguards that enhance and will serve as an appeals forum to hear long term livelihood sustainability. Regain- complaints and resolve disputes. A Mitiga- ing lost livelihood requires adequate lead tion, Resettlement and Development time and preparation and therefore people Action Plan (MRDAP) accepted by the must be fully compensated before relocation affected people should be formalised from their land, house or livelihood base. If through two legally binding contractual compensation payment is delayed, interest agreements. on the compensation amount must be paid ■ a master contract that outlines the to account for inflation obligations of government and the developer to carry out the actions set out Agreeing on mitigation, resettlement and in the MRDAP in time and to the full development Mutually agreed mitigation, resettlement extent. The master contract will also and development provisions should be specify penalties, incentives and reme- prepared jointly with the participation of all dies to facilitate compliance by govern- affected people, government and the devel- ment and the developer. The financial oper. Sustainable mitigation, resettlement and other resources to fully comply with and development provisions and opportuni- the MRDAP have to be secured prior to ties include: signing the master contract.

242 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

■ performance contracts jointly signed by The benefits could be related to government and the developer with project finance, reservoir construction, individual families and the community operation, downstream release and specifying entitlements (compensation, revenue sharing. The benefits under resettlement where necessary and direct each category are elaborated in the benefits from the project), delivery Guideline 20 in Chapter 9. Examples of schedule, and recourse procedures. opportunities include preferential fishing rights on reservoirs, land in the irrigation In order to guarantee implementation, a command area, rights to draw down performance bond should be considered as lands, equity shares, rural electrification part of the master contract in the case of from power generated, ownership of private sector developers. See Policy Princi- tourist facilities, custodian-ship over ple 6.2 under Strategic Priority 6 for a wildlife and other natural resources.11 discussion of performance bonds. The adversely affected people should While the high level multi-stakeholder participate in the identification, selec- committee will carry out preliminary dispute tion, distribution and delivery of bene- resolution and grievance redress functions, fits. The adversely affected people, the master and performance contracts are government and the developer / finan- meant to empower communities and fami- cier should assess and agree on the level lies to seek redress from courts within the of benefits. As a general principle, the country as a final resort. level of benefits should be sufficient to induce demonstrable improvements in the The contracts signed before construction standard of living of the affected people. starts remain in effect through the operation phase. Signing of master and performance All categories of affected people - displaced contracts, and demonstration by govern- and those located upstream, surrounding the ment and the developer of their capacity to area of the reservoir, downstream of the dam fulfil their respective commitments signifies and host communities for resettlement should consent by the affected people to move the be considered eligible in principle. They could project from the design stage to the con- benefit in varying degrees, or they could struction stage. benefit equally, depending on the extent of risk the dam poses to their livelihoods. 5.4 Adversely affected people are recognised as first among the benefi- Benefits could be the in the form of communi- ciaries of the project. Mutually agreed and legally protected benefit ty assets or services; and could be individual sharing mechanisms are negotiated and household focused. Once the stakeholders to ensure implementation. have agreed on the type and level of benefits, they need to decide on delivery mechanisms People adversely affected by a dam project and timing. Commitments on benefits from should be the first to benefit from the the project should form part of the perform- project. Appropriate mechanisms should be ance contracts with affected families and the introduced to ensure equitable distribution community. of development opportunities generated by the dam.

The Report of the World Commission on Dams 243 Chapter 8

Strategic Priority 6 Ensuring Compliance

Key Message

Ensuring public trust and confidence requires that governments, developers, regulators and operators meet all commitments made for the planning, implementation and operation of dams. Compliance with applicable regulations, criteria and guidelines, and project-specific negotiated agreements is secured at all critical stages in project planning and implementation. A set of mutually reinforcing incentives and mechanisms is required for social, environmental and technical measures. These should involve an appropriate mix of regulatory and non-regulatory measures, incorporating incentives and sanctions. Regulatory and compliance frameworks use incentives and sanctions to ensure effectiveness where flexibility is needed to accommodate changing circumstances.

Effective implementation of this strategic priority depends on applying these policy principles:

6.1 A clear, consistent and common set of 6.3 Costs for establishing compliance criteria and guidelines to ensure compli- mechanisms and related institutional ance is adopted by sponsoring, contract- capacity, and their effective application, ing and financing institutions and are built into the project budget. compliance is subject to independent and transparent review. 6.4 Corrupt practices are avoided through enforcement of legislation, voluntary 6.2 A Compliance Plan is prepared for each integrity pacts, debarment and other project prior to commencement, spelling instruments. out how compliance will be achieved with relevant criteria and guidelines and 6.5 Incentives that reward project proponents specifying binding arrangements for for abiding by criteria and guidelines are project-specific technical, social and developed by public and private financial environmental commitments. institutions.

Rationale of affected peoples, under-performance and Governments and other stakeholders need environmental degradation resulting in to be satisfied that once informed decisions legitimate criticism and a serious breakdown are made, all parties will ensure that they in stakeholder trust and confidence in the monitor and comply with process, implementation, and outcomes of obligations throughout the life decision-making. Governments and other of a project. While there are stakeholders need to be good examples of the develop- Many States and international financing satisfied that once ment of innovative compliance institutions have comprehensive policies, informed decisions are guidelines, the WCD Knowl- criteria and guidelines for implementing a made, all parties will edge Base demonstrates failure dam project once the decision to build is ensure that they monitor on the part of developers and taken. In the case of technical requirements, and comply with others to fulfil all voluntary and the contracting process clearly lays out the obligations throughout mandatory obligations for the responsibilities, tasks, monitoring tools, the life of a project. assessment and implementation indicators, rewards and penalties associated of approved projects. As noted with contract implementation. Contracts in the Global Review, the failure to comply typically provide for a penalty or bonus for with obligations has led to impoverishment late or timely completion and the posting of

244 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

a performance bond by the general contrac- developing water and energy All project participants, tor for the site. resources. This will provide a including government framework for assessing compli- agencies, multilateral, Resettlement, environmental mitigation ance internally and externally bilateral and commercial and compensation are rarely, if ever, subject and demonstrating compliance financing institutions, to such rigorous design processes and in a transparent manner to private sector contractual terms. Often, these activities are stakeholders. carried out by government ministries or developers and NGOs agencies and are not covered by contractual There are already many excellent should adopt a clear set obligations. Where they exist, contracts for sets of criteria and guidelines for of criteria and guidelines social and environmental programmes are planning, project assessment, for developing water and oriented towards task completion, without project construction and opera- energy resources. any rewards or penalties related to compli- tions. The WCD Criteria and ance with relevant criteria and guidelines. Guidelines elaborated in Chap- Where resettlement numbers are underesti- ter 9 are not intended to be exhaustive. mated, there may be no clarity on which Rather they provide a set of conditions that agency or firm – aside from government – is must be fulfilled to restore the level of trust responsible. When funds are not available for and confidence in the project process. programme completion, or there is a lack of accountability, failure to resolve outstanding The first step is to ensure that each particu- settlement cases has led to long-standing lar project participant makes a binding problems with displaced groups. commitment to the criteria and guidelines that apply to them. This should include Trust and confidence in the capacity and government agencies, bilateral and multilat- commitment to meet obligations must be eral financing institutions, private sector restored if new projects are to create more developers and NGOs. In the case of positive development outcomes and avoid the government agencies, this may involve level of conflict that has occurred in the past. incorporating particular criteria into the This requires the formation of new relation- regulatory frameworks, operational procedures ships and new and more effective means of and staff guidelines. For private sector devel- ensuring compliance. opers this may involve obtaining certification, adopting internal practices and procedures, Elaboration of Policy Principles establishing codes of conduct or entering into integrity pacts, to ensure compliance with best 6.1. A clear, consistent and common set of social and environmental practice. criteria and guidelines to ensure compliance is adopted by sponsoring, contracting and financing institutions Ideally, participants should not only accept and compliance is subject to independ- the WCD recommendations, but also ent and transparent review. harmonise their criteria and guidelines with those of other bodies. Not every detail needs All project participants, including govern- to be the same, but all project participants ment agencies, multilateral, bilateral and should use common parameters. For exam- commercial financing institutions, private ple, different international financing agen- sector developers and NGOs should adopt a cies or national and provincial authorities clear set of criteria and guidelines for can harmonise their criteria and guidelines.

The Report of the World Commission on Dams 245 Chapter 8

By harmonising the criteria they employ for include the development of a stewardship social and environmental management, council on dams or the development of an multi-lateral and bilateral financiers, includ- International Organisation for Standardiza- ing Export Credit Agencies, will prevent tion (ISO) standard for dam management . developers from turning to financiers with weak guidelines to fund unacceptable In recent years the Forestry Stewardship projects. Project participants will reduce Council (FSC) has made rapid progress on needless duplication of effort by employing a certification in the forest products industry. standard set of guidelines and tools such as The FSC has the capacity to influence strategic impact assessment and environmen- behaviour through the certification of an tal impact assessment to bring projects to internationally traded product. decision points in a more timely and less costly manner. Advancing the systematic management of dams-related impacts through existing Ensuring compliance with criteria and recognised international standards, such as guidelines the ISO, has the distinct advantage of Project participants, in particular govern- tapping into an existing and long-standing ment and developers will have to show that international structure and network. The they have lived up to their commitments in Commission encourages industry, and order to restore trust and confidence in the particularly dam operators, to adopt ISO process. An external review process involv- 14001 within a broader compliance frame- ing an Independent Review Panel could be work that includes performance bonds, the best way to achieve this (see Guideline integrity pacts, and other tools.12 ISO-based 22 in Chapter 9). Advancing the use of such standards such as the ISO 9000 and 14000 panels will require: series could be supplemented by a sector ■ establishing an accredited list of experts specific technical specification document for through a multi-stakeholder advisory dams, incorporating the Commission’s final group; recommendations and guidelines. ■ giving such panels the power to invoke the ‘stop lights’ that the Compliance This would require agreement and effort on Plan will establish; and the part of industry, multilateral banks, ■ ensuring information from the independ- NGOs and affected peoples groups, together ent panel is available to the public. with willingness on the part of the ISO to develop the specification (see recommenda- Another mechanism that can be used either tion in Chapter 10). The development of an with, or separate from, an Independent international certification system based on Review Panel is independent certification. the Commission’s report cannot replace the To obtain certification, participants must integration of its guidelines into national show they conform with international regulatory frameworks by governments. It standards for practices and procedures should be seen as a complementary ap- through regular monitoring and review by proach, not as an alternative. It would an accredited external body. encourage the private sector to promote and adapt standard voluntary codes of conduct This has a number of potential practical increasingly required in today’s competitive manifestations in the case of dams. These markets.

246 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

6.2. A Compliance Plan is prepared for ments (including judicial processes), greater each project prior to commencement, reliance can be placed upon direct regula- spelling out how compliance will be achieved with relevant criteria and tion. Where regulatory systems and support- guidelines and specifying binding ing institutions are still evolving, the role of arrangements for project-specific financiers, insurers and contractors, and the technical, social and environmental commitments. availability of a broad range of non-regulatory measures will be of particular importance in assisting States to proceed with accepta- Different States are at different stages in ble developments. developing regulatory systems and institutional capacity and their systems will evolve Binding arrangements must be in place for to reflect their particular circumstances. The social and environmental measures. The nature of existing systems will influence the agreed terms of resettlement and environ- range of compliance measures available. mental management conditions need to be How these measures interrelate to achieve a incorporated into legally binding and satisfactory compliance package will need to publicly available documents, and into any be determined on a case by case basis. relevant statutory approvals issued by the State. An overarching Compliance Plan is the best Two mechanisms, used with way to ensure that compliance activities and An overarching significant success in other fields, measures are effectively pursued and imple- Compliance Plan is the mented, and should be developed for each can be applied to dam related best way to ensure that project. This plan will set out how the projects to ensure fulfilment of compliance activities developer will ensure compliance with commitments. They are: and measures are project related obligations. It may refer to effectively pursued and regulatory and non-regulatory processes and Performance bonds, backed by implemented, and obligations. Where the plan relies upon the financial assurances, to provide should be developed State’s institutional capacity to ensure financial security that obligations for each project. compliance, the developer may have to will be met (such as guaranteeing enhance this institutional capacity before the fulfilment of performance the project proceeds. The measures taken contracts to fulfil social and environmental need to create trust and confidence that the commitments), and Compliance Plan can and will be met. Trust funds to hold and manage funds set In considering the range of tools and mech- aside for a particular purpose (such as anisms available to ensure compliance, resettlement or environmental measures). greater use should be made of existing conditions directly linked to achieving Both of these measures can provide States ongoing compliance. These conditions need and stakeholders with a means of achieving to be built in at key stages of the project a new level of trust and confidence in cycle to provide maximum incentive to compliance, particularly in relation to: comply. In States with well-developed ■ the need to ensure that the full costs of regulatory systems and institutional arrange- resettlement and environmental per-

The Report of the World Commission on Dams 247 Chapter 8

formance, including monitoring and for holding and distributing royalties to fund auditing, are met; ongoing initiatives. ■ the need to secure sufficient financial As with the granting of operating licences, security up front to protect the State and all contingencies cannot be anticipated. A the community from the risk of default; balance must be struck on a case by case and basis between providing the necessary level ■ the importance of providing a financial of certainty to stakeholders that commit- incentive to the contractor to complete ments will be met, and sufficient flexibility environmental and rehabilitation works. to accommodate open and transparent adaptive management. Social and environmental performance bonds, supported by adequate financial 6.3. Costs for establishing compliance assurances to cover identified resettlement mechanisms and related institutional costs and environmental measures should be capacity, and their effective application, are built into the project budget. posted before project work starts. Such performance bonds are already used for the Arrangements need to be put into place to technical components of dam ensure that the costs of implementing the projects. Other types of projects Social and Compliance Plan are built into the project carrying high environmental environmental budget. Where projects are developed by the risk have also used them performance bonds, private sector the State and affected people successfully. supported by adequate should be satisfied that these costs are fully financial assurances to financed and that appropriate arrangements In cases where performance cover identified are in place to link compliance to key stages bonds are not appropriate, such resettlement costs and in the project cycle. as those where the State is the environmental developer, trust funds should be When additional institutional capacity is measures should be used to achieve the same required, the costs should be explicitly posted before project objective. Trust funds are included under funding for compliance work starts. widely employed to ensure that activities. Compliance activities represent a funds are used for a designated transaction cost associated with the project purpose. They can be employed to hold and should be treated as such. Competing funds for dam related resettlement and options may impose differing compliance environmental measures and for monitoring burdens on society. This should be reflected and auditing costs through the life of the in analysing the acceptability of these project. The trust fund deed needs to be options. Past experience shows that there is publicly available and the trustees need to an inherent risk of failure in both the be independent from the project sponsors. satisfactory completion of the task and the effort to ensure and enforce compliance. Trust funds can be effectively used either alone or in conjunction with performance Because of the high cost of compliance and bonds to secure financing for ongoing the risk of failure, a project’s ability to monitoring and auditing obligations deliver on the Compliance Plan must be throughout the life of the project. These explicitly addressed in the multi-criteria mechanisms also provide an effective means analysis to assess options.

248 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

6.4 Corrupt practices are avoided prevent corruption, including extortion through enforcement of legislation, by their officials; and voluntary integrity pacts, debarment and other instruments. ■ the State to reduce the high costs and distortion of public procurement proce- All States need to adopt and implement dures caused by corruption. common and consistent anti-corruption legislation. Chapter 6 outlines progress 6.5 Incentives that reward project propo- made on this front, particularly through the nents for abiding by criteria and guidelines are developed by public ratification of the OECD Convention on and private financial institutions. Combating Bribery of Foreign Public Officials in International Business Transac- Compliance with norms, regulations and tions. practices often becomes widespread when the cost of compliance falls below the Implementation, monitoring and enforce- cost of non-compliance. Enforcing ment of such agreements will depend on the regulations and agreements is one Compliance activities resources dedicated to such efforts and way to support this process. represent a transaction existing business practice. Integrity Pacts cost associated with the can be employed as a project-specific The level of risk and the transac- project and should be mechanism to complement national legislation costs of constructing dams treated as such. tion. These pacts have the advantage of not have risen steadily due to the requiring explicit legislation and can be conflict over dams. This gives used without waiting for lengthy legislative project developers a financial incentive to processes where laws do not exist. comply with the Commission’s recommendations in order to reduce conflict, leading An Integrity Pact is a voluntary undertaking to an improvement in market access, that sets the contractual rights and obliga- financing terms and profitability. tions of all the parties to a procurement contract. This eliminates uncertainties A project following the WCD Criteria and about the quality, applicability and enforce- Guidelines is likely to be: ment of laws in a specific country. Integrity ■ an economically, socially and environ- Pacts can create greater trust and confidence mentally sound project; and in the decision-making process, a more ■ a project that has achieved public hospitable investment climate and public acceptance through an inclusive partici- support for government procurement and patory process of needs and options licensing programs. assessment.

The Integrity Pact has two principal goals, This will clearly entail additional up-front namely to enable: costs but this investment will bear fruit later ■ companies to abstain from bribery by in terms of reduced risk and increased providing assurances that competitors benefits. will also abstain from bribery and that government procurement agencies will The importance of public recognition follow transparent procedures and should not be underestimated. Compliance

The Report of the World Commission on Dams 249 Chapter 8

incentives could include awards, benefits for the socially and environmental- prizes, and other forms of public ly responsible investor. A few examples of recognition in the dams commu- promising mechanisms to encourage inves- nity, introducing healthy compe- tors to comply with criteria and guidelines tition to do better. Sanctions are listed below: should include public exposure ■ and debarment, either temporary Socially Responsible Investing The current growth in socially responsible or permanent, from participation investing (SRI) in both United States and in tenders and contracts. European Markets far outstrips the growth of Private sector financial service ‘traditional’ investment funds and provides a providers and international window of opportunity for financing financial institutions (IFIs) must projects that emerge from proper options work together to ensure that assessment. projects and companies that ■ Lower Cap on Insurance Liability comply with the WCD Criteria Premiums in the insurance industry are and Guidelines have access to partly based on the extent of a project’s one or more of the following incentives: liability. A lower cap on liability for dam ■ enhanced access to private equity projects certified as complying with the capital; WCD Criteria and Guidelines could be ■ better terms on debt finance (loans) and arranged particularly where legally binding insurance products; arrangements that reduce the risk of non- ■ lower rates on bond financing; compliance are in place. ■ improved accounting for environmental ■ Bond Rating Systems assets and liabilities; Bonds are increasingly used at the sovereign ■ preferential and expanded access to debt and corporate level for infrastructure financing from IFIs and their private projects. A rating system similar to the sector windows; Standard & Poor credit-worthiness rating ■ risk guarantees from IFIs; and that reflects compliance could favour WCD ■ interim credit mechanisms to assist compliant bonds. pipeline projects to implement WCD ■ Accounting for Environmental Costs recommendations. and Liabilities Projects that adopt the WCD Criteria and In the case of IFIs many of Guidelines will present a more transparent Compliance incentives the policies, procedures and could include awards, picture of the environmental costs and instruments are already being prizes, and other forms liabilities of projects and companies. This developed. In the case of of public recognition in will lower their market risk and reduce private sector instruments the dams community, volatility brought about by stakeholder there is a continued need for introducing healthy activism in this area. innovation in order to competition to do better. convert adherence to criteria and guidelines into tangible

250 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

Strategic Priority 7 Sharing Rivers for Peace, Development and Security

Key Message

Storage and diversion of water on transboundary rivers13 has been a source of considerable tension between countries and within countries. As specific interventions for diverting water, dams require constructive co-operation. Consequently, the use and management of resources increasingly becomes the subject of agreement between States to promote mutual self-interest for regional co-operation and peaceful collaboration. This leads to a shift in focus from the narrow approach of allocating a finite resource to the sharing of rivers and their associated benefits in which States are innovative in defining the scope of issues for discussion. External financing agencies support the principles of good faith negotiations between riparian States.

Effective implementation of this strategic priority depends on applying these policy principles:

7.1 National water policies make specific through various means of dispute provision for basin agreements in shared river resolution including, in the last instance, basins. Agreements are negotiated on the the International Court of Justice. basis of good faith among riparian States14 . They are based on principles of equitable and 7.4 For the development of projects on rivers reasonable utilisation, no significant harm, shared between political units within prior information and the Commission’s countries, the necessary legislative strategic priorities. provision is made at national and sub- national levels to embody the Commis- 7.2 Riparian States go beyond looking at water as sion’s strategic priorities of ‘gaining public a finite commodity to be divided and acceptance’, ‘recognising entitlements’ embrace an approach that equitably allocates and ‘sustaining rivers and livelihoods’. not the water, but the benefits that can be derived from it. Where appropriate, negotia- 7.5 Where a government agency plans or tions include benefits outside the river basin facilitates the construction of a dam on a and other sectors of mutual interest. shared river in contravention of the principle of good faith negotiations 7.3 Dams on shared rivers are not built in cases between riparians, external financing where riparian States raise an objection that is bodies withdraw their support for projects upheld by an independent panel. Intractable and programmes promoted by that disputes between countries are resolved agency.

Rationale Such conflicts are often caused by proposals to Conflict over transboundary rivers usually store or divert water by constructing dams. results from a power imbalance amongst Experience suggests that disputes over water riparians where one State or province is can be resolved and co-operation developed, sufficiently influential to exert its authority even where disagreements in other spheres of over others. Generally upstream States are international relations remain unresolved. considered to be in a more influential position as they can control the water There are 261 international river basins source, but regional power imbalances may worldwide. Most do not have agreements also make it possible for downstream ripari- covering water allocation principles. Nego- ans to exert influence over upstream States. tiation of such agreements between riparian Similar conflicts may also occur within States has proceeded on a case-by-case basis States where rivers cross internal political without any overarching globally binding borders. legal instrument.

The Report of the World Commission on Dams 251 Chapter 8

The 1966 Helsinki Rules on the In the absence of effective international Experience suggests Uses of Waters of International agreements, other measures need to be that disputes over Rivers adopted the principle that invoked. The ability of States to implement water can be resolved each State had ‘a reasonable and dam projects on shared rivers is often related and co-operation equitable share in the beneficial to financial and technical support from developed, even uses of waters in an international external agencies and the effectiveness of where disagreements basin’, but these rules have no public opinion in influencing public policy. in other spheres of status in international law. international relations While the principles set forth in In this regard, countries fall into three broad remain unresolved. the Helsinki Rules represent categories: what many experts contend are ■ those with the financial and technical long accepted principles, these Rules have resources to be totally independent; not achieved the level of a binding interna- ■ those that require financial or technical tional treaty. Until it enters into force, this support for a significant proportion of is also true of the UN Convention on the the project itself; and Law of the Non-Navigational Uses of ■ those that may be capable of undertaking International Watercourses. Through a the project independently, but rely on resolution of the United Nations General external support for other projects and Assembly in 1997, this Convention attract- programmes in the same sector. ed support from 103 countries, but three countries voted against it and 27 abstained. In addition to the application of legal The Convention took 27 years to develop principles, external financing agencies have and has yet to be ratified by enough coun- influenced and can continue to influence tries to bring it into force. Among the countries in the second and third categories. opposing and abstaining countries are those In the first category, active networking across with major dam building programmes them- borders can inform public opinion and encour- selves or with an interest in restricting devel- age moves towards a policy of co-operation. opment projects within other riparian States. Elaboration of Policy Principles International efforts to develop a universal framework for negotiations appear to have 7.1 National water policies make specific had limited effect and some countries refuse provision for basin agreements in to respect what can generally be considered shared river basins. Agreements are negotiated on the basis of good faith as a growing body of international opinion. among riparian States. They are based The Commission views the principles of the on principles of equitable and reason- UN Convention as an emerging body of able utilisation, no significant harm, prior information and the Commis- customary law and considers that States will sion’s strategic priorities. reduce the possibility of conflict if they are prepared to endorse and adhere to them. The approach adopted by a national or sub- This contrasts with a situation where some national authority towards shared rivers may countries have followed a unilateral ap- have a direct and significant influence on proach to the use of water resources and other water resource management strategies. reject the need for an integrated basin-wide The principles embodied in the 1997 UN framework for water resources management. Convention on the Law of the Non-

252 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

Navigational Uses of International Water- this intention is reflected in the South courses warrant support. States should make African National Water Act. Such provi- every effort to ratify the Convention and sions would provide a common basis for bring it into force. Where there are obsta- riparian States to move towards cles to endorsing the Convention, the more integrated management following key principles it embodies can still agreements for shared watercours- The Commission’s provide a framework for further dialogue es. On the basis of these agree- message is grounded between riparian States: ments, the affected States can on the need to obtain ■ ‘equitable and reasonable utilisation’ that adopt a progressive approach to the consent and co- promotes the optimal sustainable use of institutional development, operation of riparian the river, taking into account the inter- starting with exchange of infor- States in the ests of other riparians; mation, joint scientific teams to management of shared ■ ‘no significant harm’ to other riparians and analyse data, and joint arrange- water resources. compensation or mitigation for any harm ments for monitoring the imple- caused; and mentation of agreements. ■ ‘prior information’ referring to the need to Success with a progressive approach like this inform other riparian States on planned will provide the confidence to embark on measures that may have a significant co-operation in other areas. Internationally effect on them. there is significant interest in transboundary The meaning of these terms is still evolving. issues and external financing agencies have Particularly, the application of the principle expressed interest in supporting riparian of ‘no significant harm’ will often conflict at States that agree on a common approach. a basic level with many applications of the Such financing should be provided within a principle of ‘equitable and reasonable broad framework of co-operation, rather utilisation’. These interactions have not than as a specific project-related compo- fully been resolved legally or customarily, nent. suggesting that in their application, these principles should be read alongside the 7.2 Riparian States go beyond looking at water as a finite commodity to be Commission’s strategic priorities when divided and embrace an approach that planning future water resources and hydro- equitably allocates not the water, but power developments. the benefits that can be derived from it. Where appropriate, negotiations include benefits outside the river basin The Commission’s message is grounded on and other aspects of mutual interest. the need to obtain the consent and co- operation of riparian States in the manage- Innovative solutions are needed to solve ment of shared water resources including apparently intractable problems. Often, both consumptive and non-consumptive negotiations over shared rivers have devel- uses. In addition to having ratified interna- oped into disputes over allocating what may tional agreements, individual States should appear as an insufficient resource. A more specifically address shared river basins in equitable and sustainable resolution may be their water policy or legislation, providing possible by shifting from a primary focus on clarity on their intention to co-operate in the allocation of the water resource, to a water resources management. For example, focus on the benefits that derive from the

The Report of the World Commission on Dams 253 Chapter 8

use of the water, encompassing considera- intervention. Conducted in a manner tion of wider development objectives and consistent with openness and information the options available to meet them. This sharing, a competent, independent entity shift provides an opportunity to look more acceptable to all riparian States should constructively at alternative programmes for conduct strategic and project-related impact meeting development objectives. assessment studies following the practice outlined in Chapter 9. It is possible to expand the horizon of negotiations further to include other issues The level and intensity of impact assess- that optimise the comparative advantages of ments will depend on the planning stage, two or more States. Such synergies may but in all cases should include environmen- result from differences in location, climate tal, social, health and cultural heritage or resource endowment. To some extent, the assessments. Safeguards are needed to ensure Helsinki Rules began this shift. independence in directing and financing the In the wider negotiation arena, assessment team. Mechanisms such as A more equitable and the principles of sharing establishing a joint trust fund for imple- sustainable resolution benefits can include an array of menting the assessments should be consid- may be possible by other resources, including co- ered. The impact assessments should be seen shifting from a primary operation in other sectors, or as part of the joint institutional strengthen- focus on the allocation of financial payments. ing activities of riparian States to provide a the water resource, to a common, interactive approach and a sound focus on the benefits that An approach centred on wider basis for political dialogue. derive from the use of development objectives creates the water. a link between discussions at Where disputes cannot be resolved, an transboundary level and independent panel should be established strategic planning processes that goes beyond the remit of the impact within countries that can be used to define assessment. The creation and operation of needs more clearly and map out a wider such a panel is defined in the 1997 UN range of alternatives. Early engagement can Convention on the Law of the Non-Naviga- avoid disputes becoming polarised around a tional Uses of International Watercourses. specific project proposal and entrenching Good faith negotiations may of themselves negotiating positions. lead to mutually agreeable outcomes. In the event that disputes remain, and if parties do 7.3 Dams on shared rivers are not built in not have recourse to dispute resolution cases where riparian States raise an through international, regional or bilateral objection that is upheld by an independent panel. Intractable disputes agreements, the affected parties could refer between countries are resolved through the matter to the International Court of various means of dispute resolution Justice (ICJ), either by mutual agreement, or including, in the last instance, the directly if both parties have previously International Court of Justice. submitted to the compulsory jurisdiction of Openness and information sharing is a key the ICJ by declaration under Article 36 of first step in any transboundary water sharing the Statute of the Court. situation. From this can follow an independent and objective assessment of the conse- These policy principles apply to all water- quences and impacts of any proposed courses including tributaries. Their applica-

254 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

tion to tributary rivers is a location specific by national and sub-national authorities, or issue dependent on the significance and a mix of the two. Dam projects may also be relationship of the tributary to the main subject to clearance from national and sub- river and the interests of riparian States. In national environmental cases where countries have already reached agencies. Both licences and States should embody basin level agreements for the main stem environmental clearance can the principles of shared river, such agreements should be extended be used to ensure that accept- waters in national water to significant tributary rivers taking into ed environmental and social legislation and develop a consideration the Commission’s strategic provisions are met. Where locally appropriate policy priorities. Where no agreements exist, federal funds are sought for framework for riparian States should adopt an integrated project assistance, they should addressing such issues. approach looking at the management of the be made conditional on main-stem and its tributaries within the compliance with the Commis- context of the entire river basin. sion’s strategic priorities. As part of their overall responsibility, States should be more 7.4 For the development of projects on pro-active earlier in the planning process to rivers shared between political units facilitate resolution of disputes between within countries, the necessary legislative provision is made at national and riparian provinces. sub-national levels to embody the Commission’s strategic priorities of 7.5 Where a government agency plans or ‘gaining public acceptance’, ‘recognis- facilitates the construction of a dam on ing entitlements’ and ‘sustaining rivers a shared river in contravention of the and livelihoods’. principle of good faith negotiations between riparians, external financing The Commission’s seven strategic priorities bodies withdraw their support for projects and programmes promoted are as relevant to rivers shared across sub- by that agency. national boundaries as they are to those wholly within one province or those shared The international community needs to take between States. Where their legal system a strong and concerted stand in the case of provides for national-level control of water shared rivers. While the decision to build a issues, States should embody the principles dam is often considered a sovereign deci- of shared waters in national water legislation sion, the decision of external agencies to and develop a locally appropriate policy support a dam depends on whether the framework for addressing such issues. In proposed project complies with that agency’s other cases, States at a federal level should policies and guidelines. It is therefore of explore mechanisms to encourage good concern that bilateral, multilateral and practice and provide incentives for compli- export credit agencies have not yet harmo- ance with the strategic priorities. nised their policies towards shared watercourses. These often-inconsistent policies Although many sub-national governments make it more difficult to improve the way have a mandate to manage water, a national transboundary issues are handled. The or federal government may exercise sanc- complexity of the situation is increased by tions through legislative and regulatory the disparate and fluid nature of financial mechanisms and provide financial incen- support. This inconsistency often results in tives. In many cases dam development and situations where, although an external operation depend on project licences issued agency may not be directly financing a dam

The Report of the World Commission on Dams 255 Chapter 8

on a shared watercourse, its independent expert assessment of social and support for other projects in environmental impacts. the same sector allows national resources to be In the absence of agreement among riparian allocated for this purpose. States, external agencies should make their involvement conditional on the findings of an It is essential that external independent commission as envisaged under agencies harmonise their the 1997 UN Convention on the Law of the policies towards shared Non-Navigational Uses of International waters and deal with the Watercourses or other appropriate mecha- sector as a whole rather than nisms agreeable to all parties. In cases where with specific projects. Such policies should States proceed with projects in the absence of incorporate aspects of notification to such a commission, or reject its findings, the riparian States, the desirability of ‘consent’ external financing agency should withdraw its or ‘no objection’ from riparian States, and support from the sector concerned.

Endnotes landmark case for the Ainu of Japan, on 28 March 1997 a local court in Sapporo, 1 For example in many Latin American Hokkaido, recognised the Ainu as an countries (Argentina, Bolivia, Brazil, Chile, indigenous and minority people. The Asian Colombia and Ecuador) and in Australia, Development Bank has also adopted a policy Canada, India, New Zealand, and the on ‘indigenous peoples’ to guide its opera- Philippines national laws recognise indige- tions. For a more nuanced discussion of the nous people’s rights. Specific legislation on applicability of the concept of indigenous indigenous people’s rights includes the peoples to Asia, see Kingsbury, 1998. Philippines Indigenous Peoples Rights Act 4 Daes1996b; World Bank, 1991. (1997), Australian Aboriginal Land Rights 5 Nepal: Modifiying the intake, installing an (Northern Territory) Act 1976 and Chile’s extra desander, dredging the forebay and Indigenous Law 1993. Recent Canadian refurbishing the generators/turbines and practice promotes the need for free, prior informed consent of indigenous and tribal power house control systems at the Trushuli- peoples to projects in certain contexts. Devighat hydropower station in 1995 Goldzimer, 2000, soc013, WCD submission; (originally built in the 1970’s) improved average annual power generation by 46%. Hart, pers. comm. 2000; IDS, 2000. NEA 1997. 2 See endnote 1 3 Colchester, 1993, 1995; Tenant, 1994; Gray, Lao PDR: Nam Ngum is a 150 MW hydro- 1995; Kingsbury, 1995; Fisher, 1993; May- power plant near Vientiane, the capital of bury-Lewis, 1996; Daes, 1996a, p72; Pritch- Lao PDR. Surplus power is exported to ard, 1998a, p44; Pritchard, 1998b, p61. Some Thailand and the power sales agreement Asian and African countries already accept between the two countries is revised every 4 that the term ‘indigenous peoples’ applies to years. Based on updated hydrology and other the ‘tribal peoples’ and ‘cultural minorities’ variables, studies were undertaken to within their borders, including Cambodia, optimise the monthly and day-to-day Botswana, Nepal and Philippines. In a operation of reservoirs and turbines. Assum-

256 Dams and Development: A New Framework for Decision-Making Strategic Priorities – A New Policy Framework

ing the same average tariff, revenue from power shareholder with a 50.1% share and Hydro sales to Thailand increased by about 10%. Quebec owns the remaining 49.9% of the Lahmeyer International, 1990. shares. Minashtuk is the first project devel- 6 Johnston, 2000. oped by Hydro-Innu. Besides guaranteeing a share in profits, it allows the Montagnais to 7 Acreman et al, 2000; WCD Thematic II.1 design a project according to their priorities Ecosystems. and in the long-term reinvest the profits in a 8 465 cases in the United States, 3 in France, manner that supports the economic develop- 1 in Norway, a few in Canada. ment of their community. Milewski et al 9 Revenga et al, 2000. 1999, soc196, WCD Submission. 10 Many countries are in the early stages of 12 ISO 14001 is an internationally recognised developing EFRs, for example adopting and standard that identifies the basic processes of implementing relevant legislation, making an environmental management system enquiries into available environmental flow (EMS). It allows an organisation to effective- methodologies and developing policy, but are ly identify, monitor and control its environ- not necessarily at the stage of formally mental impacts. It does not specify perform- undertaking EFRs or determining EFRs for ance requirements, beyond a commitment to freshwater systems. At least 29 countries comply with applicable regulations and other have used EFRs including: Australia, Austria, commitments. The standard enables third- Canada, Denmark, Finland, France, Germa- party certification for organisations that ny, Hungary, Indonesia, Ireland, Italy, Japan, conform to its specifications. While there are Korea, Lesotho, Moldavia, Namibia, New many models for EMS design, the ISO 14001 Zealand, Norway, Portugal, Puerto Rico, standard is emerging as the dominant South Africa, Spain, Swaziland, Sweden, standard, with over 17,000 certificates Switzerland, Taiwan, The Czech Republic, granted as of June 2000. Concerns do, The Netherlands, The Ukraine, The United however, exist whether ISO standards can Kingdom, The United States. The following assure changed performance effectively. countries are exploring the use of EFRs: Corbett and Kirsch, 200. Angola, Botswana, Brazil, Chile, Israel, 13 ‘Rivers’ is used here as a general term. The Mexico, Mozambique, Nepal, Thailand, strategic priority and policy principles relate Vietnam, Zimbabwe. Tharme, 2000. equally to all types of waters which are or 11 Benefit sharing through Equity Shares- might be impacted by dams. Minashtuk Project in Canada. The Minash- 14 The term ‘riparian State’ is used to mean any tuk project in Canada illustrates benefit State through which a transboundary river sharing through a limited partnership flows or forms part of its boundary, or which company. Here the Band Council of the includes part of the catchment area of a Montagnais of Lac Saint-Jean is the majority transboundary river.

The Report of the World Commission on Dams 257 Criteria and Guidelines – Applying the Strategic Priorities

Chapter 9: Criteria and Guidelines – Applying the Strategic Priorities

any technical guidelines for Mthe construction of dams and their associated infrastructure have been developed by professional technical networks to ensure high

engineering and quality standards. What is lacking is a comprehensive and integrated framework for decision-

making on the provision of water and energy services. The Commission’s contribution is to provide a framework

that emphasises a structured process incorporating the full range of social, environmental, technical, economic

and financial criteria and standards.

The Report of the World Commission on Dams 259 Chapter 9

The framework is built upon the seven instruments and should be incorporated by strategic priorities described in Chapter 8 governments, professional organisations, and derives its strength from recognising the financing agencies, civil society and others rights and assessing the risks of all stake- as they continue to improve their own holders in the process. relevant guidelines and policies over time (see Figure 9.1). Social, environmental, governance and compliance aspects have been undervalued This chapter shows how implementing a in the past. It is here that applying the decision-making process based on the Commission’s proposals will make a differ- Commission approach will safeguard rights, ence. The framework identifies key decision reduce the risk of conflicts emerging, and points and incorporates associated criteria lower overall costs. The framework provides that translate the Commission’s policy the opportunity for agencies and communi- principles into a programme for implemen- ties to screen out unfavourable alternatives tation. Within this framework the Commis- at an early stage. It is intended to open sion proposes a set of guidelines firmly channels of dialogue between stakeholders, anchored in examples of good practice from increase mutual understanding and help the Knowledge Base to describe how its decision-makers, practitioners and affected policy principles can be realised. These people assess whether needs have been guidelines add to existing decision-support adequately addressed. Proposals for dam projects that emerge from the Figure 9.1 WCD Criteria and Guidelines strengthen other decision support screening process will have greater instruments public acceptance.

Turning the strategic priorities

Civil society and their underlying policy gy • • Codes of practice principles into reality requires a nment • Public information onment Social audit GoverWater and ener new focus for planning and • policies Social and envir • management in the water and policies ds Inter • Regulations and • • • energy sectors. This chapter ds standar International conventions International law Basin agr ds national agr concentrates on what needs to change in the way water and ds eements

ds WCD energy management plans are , health and safety eements national standar Criteria ater echnical standar developed and projects are de- T W standar ISO standar Inter and • • signed and implemented. Bringing • Guidelines Multilateral and bilateral about this change will require: or • ganisations • • Operational policies ■ Impact assessments planners to identify stakehold- Pr ocess guidelines ers through a process that recognises rights and assesses

PrivateCodes sector of practice Professional • Due diligence policies risks; organisations • Certification • • Standards ■ • Sector guidelines States to invest more at an earlier stage to screen out inappropriate projects and facilitate integration across

260 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

sectors within the context of the river 4. Project implementation: confirming basin; compliance before commissioning. ■ consultants and agencies to ensure 5. Project operation: adapting to changing outcomes from feasibility studies are contexts. socially and environmentally acceptable; ■ the promotion of open and meaningful The contractual steps of signing agreements participation at all stages of planning and issuing licences are located within this and implementation, leading to negotiat- overall framework. Although in these latter ed outcomes; stages the Commission has focused on issues related to dam projects, the principles and ■ developers to accept accountability general guidance presented here are also through contractual commitments for relevant to non-dam options for water and effectively mitigating social and environ- energy services. mental impacts; ■ improving compliance through inde- The five decision points are supported by a pendent review; and set of key criteria that describe the processes ■ dam owners to apply lessons learned from required for compliance. The criteria are past experiences through regular moni- presented in the form of checklists for each toring and adapting to changing needs decision point that provide a clear and open and contexts. mechanism for determining whether the Commission’s recommendations have been The changes will involve reform of existing followed and the process can proceed to the planning processes and an emphasis on the next stage of planning or implementation. key stages where decision-makers and stakeholder groups can verify compliance. The criteria cover the full planning and Among the multitude of decisions to be project cycles and include aspects related to taken, five critical decision points have existing dams. There are also been identified as having a particularly many dams currently in the strong influence on the final outcome. The planning, design, or construction Five critical decision first two relate to water and energy plan- stage. These ‘dams in the pipe- points have been ning, leading to decisions on a preferred line’ should also be assessed to identified as having a development plan. identify improvements that can particularly strong 1. Needs assessment: validating the needs be made. Practical steps are influence on the final for water and energy services. proposed to determine the extent to which such current projects outcome. They are 2. Selecting alternatives: identifying the comply with the Commission’s supported by a set of preferred development plan from among recommendations, and to identify key criteria that describe the full range of options. how any needed adjustments can the processes required be made. for compliance. Where a dam emerges from this process as a preferred development alternative, three Recognising that guidelines are further critical decision points occur. available from other sources, the Commis- 3. Project preparation: verifying that sion focused principally on what needs to be agreements are in place before tender of done differently. Introduction of a new the construction contract. decision-making framework through appli-

The Report of the World Commission on Dams 261 Chapter 9

cation of the criteria and guidelines will add 2. Selecting alternatives: identifying the significant value to existing processes and preferred development plan from among the full form a basis for good practice in water and range of options. The preferred development energy resource development. Taken togeth- plan is selected through a participatory er, the criteria and guidelines will improve multi-criteria assessment that gives the same development outcomes and minimise significance to social and environmental problems encountered in the past. aspects as to technical, economic and financial aspects and covers the full range of Five Key Decision Points: policy, programme, and project options. Within this process, investigations and The WCD Criteria studies are commissioned on individual As noted, the Commission has identified options to inform decision-making as five key stages and associated decision required; for example, demand-side points for the energy and water sector. The management studies or feasibility studies. most fundamental of these is selection of the preferred development plan. This deter- Where a dam emerges as a preferred option, the mines what options will be pursued to meet following key decision points occur for needs and whether or not a dam is to be project preparation, implementation and built. This decision is only taken after needs operation. and the available options to meet those needs have been fully assessed. Each of the 3. Project preparation: verifying agreements are five stages requires a commitment to agreed in place before tender of the construction procedures culminating in a decision point contract. The preparation stage covers that governs the course of future action and detailed planning and design. Licences allocation of resources (see Figure 9.2). At issued for development of a project incorpo- each decision point it is essential to test rate any conditions that emerge from the compliance with preceding processes before options assessment process. Tendering the giving authority to proceed to the next construction contract is conditional upon stage. These points are not exhaustive, and reaching negotiated agreements for benefit- within each stage many other decisions are sharing mechanisms and for mitigation, taken and agreements reached. The five key compensation, development and compli- stages and associated decision points are ance measures, in addition to technical generic and need to be interpreted within requirements. the overall planning contexts of individual countries. 4. Project implementation: confirming compliance before commissioning. The 1. Needs assessment: validating the needs for implementation stage covers procurement water and energy services. Confirmation is and construction. Issuing the licence to required that plans for water and energy operate is contingent on implementation of development reflect local and national specific benefit sharing and mitigation needs adequately. An appropriate decentral- measures at various stages through the ised consultation process is used to validate implementation period. Compliance with the needs assessment and modify it where all relevant time-bound commitments is necessary. required before commissioning the project.

262 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

5. Project operation: adapting to changing Figure 9.2 Five key decision points in planning and project contexts. Any decisions to modify facilities, development operating rules, and licence conditions to meet changing contexts are based on a participatory review of project performance Needs Assessment and impacts. Validate the needs for water and energy services The five key stages and decision points provide a framework within which decision- Criteria 1 makers and stakeholder groups can be assured of compliance with agreed procedures and commitments. The benefits of this Selecting Alternatives approach include lowering risks to liveli- Identify the preferred hoods and cost escalation, reducing the development plan among Investigative studies the full range of options number of disputes, and encouraging local Criteria 2A ownership. In the short term, additional Criteria 2 financial resources for needs and options assessment will be required to achieve compliance with the Commission’s policy principles, and efforts will be required to strengthen institutional capacity. In the NON-DAM OPTIONS DAM OPTION longer term, the potential exists for major cost savings and increased benefits.

Policy, programme, Project Preparation The remainder of this chapter describes each projects of the five key decision points and provides a Assign Verify agreements are responsiblity for in place before tender of related list of criteria for checking compliance. implementation the construction contract Stakeholder involvement is central to these Criteria 3 processes. The composition of a forum of stakeholder groups associated with each stage is different and evolves throughout the process (see Guideline 1: Stakeholder Analysis). Project Implementation Confirm compliance before commissioning

Criteria 4

Project Operation

Adapt to changing context

Criteria 5

The Report of the World Commission on Dams 263 Chapter 9

Stage 1: Needs assessment: validating the needs for water and energy services priorities and the formulation of develop- Intended outcome ment objectives. A clear statement of water and energy services needs at local, regional, and national levels An open and decentralised planning process that reflects decentralised assessments and provides opportunities for public scrutiny. In broader national development goals. An assessment based on participatory methods situations where this has not been imple- appropriate to the local context resulting in a mented, a programme of national and sub- clear set of development objectives that guide national public hearings, targeted communi- the subsequent assessment of options. ty consultations, and field surveys can assess the validity of the needs assessment. The subject of the consultations may relate to an Determining needs and setting priorities overall development strategy, master plan, between and within sectors are continuous sector plan, or basin plan, and the breadth processes specific to individual countries. of consultation can be determined accord- The Commission’s policy principles describe ingly. The WCD Knowledge Base demon- characteristics that should be reflected in strates the need for such a review, since such processes and define a shift in emphasis plans are often narrowly focused, reflect to more open and inclusive procedures. social and environmental impacts inade- Country-level responses are required to quately, are weak in identifying affected ensure that priority setting embodies a fair groups, and fail to deal adequately with the representation of basic water and energy distribution of costs and benefits. needs and provides the appropriate balance between local and national demands. In countries where a large proportion of the population does not have access to basic A country’s policy framework for water, services, a key parameter in the validation energy, social issues and the environment process should be the extent to which basic guides the whole planning process. The human needs will be met. To ensure that open and participatory approach to needs these needs are given prominence, the process and options assessment envisaged by the of validation should empower those who are Commission requires a review of these least able to influence planning systems. policies to identify and address elements that may hinder its implementation. Responsibility for this validation process rests with the State. Independent facilitation for The primary influence defining a country’s consultations and surveys and the presence of portfolio of development activities, is a set civil society groups will enhance confidence of development objectives that may be that the needs of disadvantaged groups are embodied in a five-year plan or in a regula- being considered. Briefing materials, records of tory planning framework. Ensuring that the meetings, and results of the overall process outcome of the needs assessment for water should be available in appropriate languages. If and energy services reflects stakeholder the development objectives are not confirmed priorities requires an entry point early in the by the validation exercise, they should be planning process. The Commission proposes reviewed and updated using processes consist- a validation process to confirm the setting of ent with the Commission’s policy principles.

264 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

Stage 1 Criteria Checklist

NEEDS ASSESSMENT Selecting alternatives Project Preparation Project Implementation Project Operation

Needs assessments may have been conducted through a range of processes including national, regional, sector-specific, or basin-wide plans. The verification process to be applied will need to be tailored to suit the particular circumstances.

Gaining Public Acceptance Comprehensive Options Assessment ■ A consultation plan was developed using a ■ Legal, policy and institutional frameworks were stakeholder analysis to define the groups in- reviewed and any bias against resource conserva- volved. The plan defines mechanisms for verify- tion, efficiency and decentralised options, and ing needs at the local, sub-national and national any provisions that hindered an open and level (Guideline 1). participatory assessment of needs and options ■ Verification of the needs for water and energy were addressed. services was achieved through a process of public consultation and the results of public consulta- Addressing Existing Dams tion were disseminated to stakeholders. ■ Outstanding social and environmental impacts ■ Development objectives reflect a river-basin- from past projects were evaluated and incorporat- wide understanding of relevant social, economic, ed into the needs assessment (see Chapter 8, and environmental values, requirements, func- policy principle 3.3). tions, and impacts that identifies synergies and potential areas of conflict. Sustaining Rivers and Livelihoods ■ An appropriate process was established to address ■ Ecosystem baseline studies and maintenance any disparities between the needs expressed needs were assessed at a strategic level (Guide- through the public consultations and the stated lines 14, 15) development objectives.

The Report of the World Commission on Dams 265 Chapter 9

Stage 2: Selecting alternatives: identifying the preferred development plan

Intended outcome that can deliver benefits within the short- term. A mix of alternatives that reflects the needs and meets the development objectives has been selected through a multi-criteria assess- Studies are required to assess the extent to ment of the full range of policy, programme, which policy and programme options can and project alternatives and included in a meet the development objectives. The preferred development plan. policy principles cover a number of such areas, including: ■ optimising existing investments by A major concern about past projects is that increasing operational efficiency and too limited a range of options was consid- improving productivity; ered before deciding to construct a dam. To address this, the inventory of options must ■ demand-side management assessment; respond to the agreed development objec- ■ decentralised supply options and commu- tives (Stage 1) and explicitly identify the nity-level initiatives; and beneficiary groups. The inventory of options ■ policy and institutional reforms. needs to be sufficiently diverse in terms of policy, programme, and project alternatives, Some options will need reconnaissance, pre- project scale, and geographic coverage. feasibility, and feasibility studies appropriate to the stage in the process and incorporating Strategic impact assessments provide an social and environmental surveys and initial level of screening to remove alterna- impact assessments. The decision to allocate tives that have unacceptable social and financial resources to such studies should be environmental consequences. They need to taken within the overall context of the reflect the importance of avoiding adverse options assessment process. For example, it impacts and the precautionary approach. may have been agreed that detailed investi- The assignment of relative weights to gations of supply-side approaches should designate the importance of various parame- await the outcome of demand-side studies ters should be a participatory process and that could influence the scale of any project form the basis of a multi-criteria analysis to intervention. screen and rank alternatives. For example, the gestation period in delivering benefits, Criteria relevant to project-related studies the scale of adverse impacts and costs are all are described as a subset of the process of key considerations. selecting alternatives (see Stage 2A). The findings of the studies are fed back into the The information available on each option screening process for consideration with all will not be at the same level of detail. other remaining options. This approach Decisions need to be taken during the deviates from existing practice in many screening process on whether to commission countries by cutting the direct link between further investigations and studies on indi- the feasibility study and project approval. It vidual alternatives, while at the same time encourages a broader consideration of all not jeopardising or delaying alternatives feasible options in setting priorities.

266 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

Stage 2 Criteria Checklist

Needs Assessment SELECTING ALTERNATIVES Project Preparation Project Implementation Project Operation

Gaining Public Acceptance ■ Approval to proceed with any project-level ■ Stakeholders participated in creating the invento- investigations was informed by a comprehensive ry of options, assessing options, and in negotiating assessment of options (see Criteria Checklist 2A). those outcomes that may affect them (Guide- ■ Rejection of any options was explained in an lines 1, 2). open and timely manner. ■ An agreed dispute resolution mechanism for Addressing Existing Dams negotiated processes was established with the ■ Provisions were made for resolving outstanding participation and agreement of stakeholders social and environmental impacts (see Chapter 8, (Guideline 2). policy principle 3.3) ■ Indigenous and tribal peoples gave their free, prior and informed consent to the inclusion in the Sustaining Rivers and Livelihoods development plan of any planned option that ■ An established policy exists to maintain selected would potentially affect them (Guideline 3). rivers with high ecosystem functions and values in their natural state. Comprehensive Options Assessment ■ Consideration of options took into account: ■ Strategic impact assessments and life cycle analy- avoiding dams on the main-stem of rivers wherev- sis were integrated and undertaken as an initial er possible; avoiding or minimising negative step in the process (Guidelines 4, 7, 8, 14, 17). impacts on endangered species, ecosystems, ■ A multi-criteria assessment was used to screen and livelihoods, human health and cultural resources; select preferred options from the full range of and respecting the provisions and guidance of identified alternatives (Guideline 6). relevant international treaties. ■ The screening of options: Recognising Entitlements and Sharing ■ covered all policy, programme, and project alternatives; Benefits ■ gave social and environmental aspects the ■ For any project option, stakeholders negotiated same significance as technical, economic and the guiding principles and criteria for: benefit- financial factors; sharing, mitigation, resettlement, development ■ gave demand-side options the same signifi- and compensation measures (Guidelines 2, 18, cance as supply options; 20). ■ prioritised consideration of improving perform- Ensuring Compliance ance of existing systems; ■ Sufficient institutional capacity exists, or will be ■ considered river-basin-wide aspects and enhanced, to monitor and enforce commitments cumulative impacts; for social and environmental components. ■ took account of potential changes in climate; and ■ reflected the precautionary approach. Sharing Rivers for Peace, Development and ■ Distributional and risk analyses were conducted at Security an appropriate level (Guidelines 9, 11) and ■ Any objections from riparian states were resolved environmental and social impacts were valued through good faith negotiations or independent where appropriate (Guideline 10). dispute resolution procedures (Guideline 26).

The Report of the World Commission on Dams 267 Chapter 9

Stage 2A: Investigative studies

The authority to proceed with a preparatory Preliminary negotiations with project- study for a dam project should not be a signal affected people, their community represent- that the project will be implemented. Rather, atives, and other stakeholders are central to such a decision should be integrated into the the preparatory studies in considering overall options assessment process. This will mitigation measures for any unavoidable provide a break in the traditional planning adverse impacts and investigating benefit- cycle to encourage more open decision- sharing plans. By the time a study reaches making. Viewing project options within the feasibility stage, the scope of such measures overall framework of options assessment also should be clearly defined in order to reduce facilitates the rejection of projects that fail to the likelihood of protracted negotiations meet social and environmental objectives in and a breakdown of discussions later in the favour of better alternatives. The WCD process. For the proposed project to be part Knowledge Base has demonstrated that more of a preferred development plan, the accept- rigorous estimates of project costs are also ance of the project affected people and the required as a part of such studies, and the risk prior informed consent of indigenous of cost overruns must be fully considered in peoples should be obtained. the assessment process. Ultimately the results of the study, including Meaningful participation in preparatory any outstanding issues, will be fed back into studies is central to the success of the the screening and ranking exercise (see investigation and the ultimate outcome. Criteria Checklist 2) for comparison with Careful analysis to recognise the rights and remaining alternatives prior to any decision assess the risks of stakeholder groups is to proceed further with detailed project essential. A forum of stakeholder groups development. The following plans, with needs to be identified based on project indicative budgets, need to be developed as boundaries. Agreement on the participatory a minimum requirement to act as a founda- elements of the studies should be formalised tion for any further project planning: in a consultation plan. ■ an outline environmental management The strategic impact assessment undertaken programme, including provision for an early in the options assessment will have environmental flow to maintain down- outlined the key unknowns and the areas to be stream ecosystems; investigated across all sectors. The issues will ■ an outline social mitigation, resettle- be defined in more detail in the scoping stage ment, and development plan; and of project-related impact assessments. On this ■ an outline monitoring plan, including basis, terms of reference and work plans for the outcome-based indicators. diverse range of sector specialists can be integrated. Project-related impact assessments A compliance plan will be required to have to go beyond environmental and social cover these aspects and other regulatory impact assessments to include health and requirements throughout subsequent cultural impacts. To be effective, they require stages of project planning, development, an improved level of baseline studies. and operation.

268 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

Stage 2A Criteria Checklist

Needs Assessment SELECTING ALTERNATIVES Project Preparation Project Implementation Project Operation

Project-related pre-feasibility and feasibility studies need to meet the following criteria. Policy and programme related studies may also be required, and are covered in Criteria Checklist 2.

Gaining Public Acceptance ■ Impacts on fish have been assessed and measures to avoid or minimise impacts were considered, ■ Stakeholders participated in baseline, impact and including an effective fish pass where feasible investigative studies and the negotiation of (Guideline 16). outcomes that potentially affect them (Guide- lines 1, 2, 14, 17). Recognising Entitlements and Sharing ■ The studies and impact assessments were open Benefits and independent, and were preceded by a partici- ■ Stakeholders negotiated agreements for com- patory scoping phase (Guideline 5). pensation, mitigation, resettlement, develop- Comprehensive Options Assessment ment and monitoring measures affecting them, including draft contracts where necessary ■ The investigations were analysed on a river- (Guideline 19). basin-wide understanding of social, economic, ■ Effective benefit-sharing strategies were identi- and environmental values, requirements, func- fied and agreed with people adversely affected tions, and impacts including cumulative impacts, by the project (Guideline 20). and the precautionary approach was applied. (see Guideline 5). Ensuring Compliance ■ The recommendations of studies undertaken on ■ Institutional capacity to monitor and enforce resource conservation measures, demand-side commitments for social and environmental management, local supply-side options and components of the project was analysed and improvement of existing systems were reflected measures to strengthen capacity identified. in the demand forecast for the sector. ■ An independent panel reviewed the assessment ■ Within-project alternatives were assessed using a of impacts and the planning of social and multi-criteria approach (Guideline 6). environmental mitigation plans (Guideline 22). Addressing Existing Dams Sharing Rivers for Peace Development and ■ Studies examined possible synergies from interac- Security tive operation of related water resource infra- ■ Riparian states were notified of options affecting structure in the basin. them and agreed procedures for impact assess- Sustaining Rivers and Livelihoods ments. Objections were addressed through good faith negotiations and agreed dispute resolution ■ An environmental flow requirement to maintain procedures (Guideline 26). downstream species, ecosystems and livelihoods was defined (Guideline 15).

The Report of the World Commission on Dams 269 Chapter 9

Stage 3: Project preparation: verifying commitments are in place before tender of the construction contract

Intended outcome The responsibilities of the developer in relation to mitigation, development of Clearance to tender the construction contract affected communities, and benefit-sharing is given by the relevant authority and includes conditions for the award of the contract and will be clearly reflected in the licence and operations. Mitigation and monitoring the compliance plan. The operation phase measures are formalised into contracts between responsible parties, and compliance should be contingent on compliance with arrangements are in place. specific commitments identified in the licence. Conditions for impoundment and commissioning should be explicitly stated. The Commission considers that all large dams should have time-bound licences. Where a If the tender cost estimate is substantially large dam emerges as a preferred option, a higher than the feasibility study estimate, licence for project development should be the choice of project should be reviewed to issued to the developer by the appropriate see if it still meets the selection criteria. A regulatory agency. Project preparation contin- similar check is required if needs have ues with detailed planning and design stages, changed substantially since the outset, or if including drafting of tender documents and the project scope has changed materially. If plans for benefit-sharing and mitigation. the actual price of the lowest responsive Adaptation of the criteria described here may tender exceeds agreed cost limits, procure- be necessary where the design and construc- ment should be interrupted and options tion are part of a single package. reviewed.

Negotiations with all project-affected Social and environmental mitigation people, their community representatives, measures should be defined in the tender in and other stakeholders will continue in similar detail to construction elements, good faith based on the outline agreements namely the ‘bill of quantities’. The tender reached during the feasibility stage. They should clearly identify responsibilities of the would cover all environmental and social contractor, the developer, and the govern- plans; development programmes, including ment in relation to: benefit-sharing plans; and construction- ■ the environmental management plan; related issues. A clear agreement with the affected people on the sequence and stages ■ measures to mitigate adverse social of resettlement will be required before impacts, including development opportu- construction on any project preparatory nities for affected communities; works begins, such as access roads or river ■ access to and management of new diversion works. In cases where these resources in the reservoir; negotiations stall, an independent dispute ■ the construction method and schedule, resolution process is required. The negotiat- and the construction camp; ed agreements will result in signed contracts between the developer and affected commu- ■ impact monitoring and reporting during nities and individuals, with clear targets for the operations stage; and assessing compliance. ■ compliance instruments.

270 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

Stage 3 Criteria Checklist

Needs Assessment Selecting Alternatives PROJECT PREPARATION Project Implementation Project Operation

Gaining Public Acceptance Recognising Entitlements and Sharing Benefits ■ Stakeholders participated in the project design and the negotiation of outcomes that affect them ■ Mitigation, resettlement, monitoring, and devel- (Guidelines 1, 2). opment plans were agreed with affected groups, and relevant contracts signed (Guideline 19). ■ Indigenous and tribal peoples gave their free, prior, and informed consent to the project as ■ Detailed benefit sharing mechanisms, and the designed (Guideline 3). means to deliver them, have been agreed and set in place with affected groups (Guideline 20). Comprehensive Options Assessment Ensuring Compliance ■ The stakeholder forum participated in assessing alternatives for the detailed layout of the dam, ■ Independent panels reviewed and endorsed associated infrastructure, and its operation. mitigation plans (Guideline 22).

Addressing Existing Dams ■ Provisional sums for mitigation are included in the tender, and their financing has been confirmed. ■ Cumulative and interactive impacts of existing infrastructure were addressed in the design of the ■ A Compliance Plan was prepared, presented to dam and agreements reached with stakeholders the stakeholder forum and formalised. Individual and operators to modify operating rules of exist- compliance measures include mechanisms for ing dams where needed. dispute resolution (Guideline 21).

Sustaining Rivers and Livelihoods ■ The developer has allocated funds for an effective ■ Acceptable rules were developed for reservoir monitoring and evaluation system covering filling, commissioning and operation. project performance, safety and impacts. Institu- tional capacity exists to monitor and enforce ■ The final design includes provisions for agreements effectively. emergency drawdown and decommissioning and is sufficiently flexible to accommodate changing ■ A transparent process for short-listing contractors future needs and values, including ecosystem and selecting tenders is in place and contractors needs and ecosystem restoration (Guideline 12). with a record of under-performance or corruption on past projects were identified and debarred ■ An environmental management plan incorporat- where appropriate. ing environmental flows and other mitigation and enhancement measures was agreed with ■ Relevant performance bonds have been secured, stakeholders and defines monitoring and evalua- trust funds established and integrity pacts signed tion programmes. (Guidelines 23, 24, 25).

■ The developer provided sufficient evidence to ■ The licence for project development defines the demonstrate that proposed mitigation and responsibility and mechanisms for financing development measures will be effective in meet- decommissioning costs. ing their objectives. Sharing Rivers for Peace, Development and Security ■ Resolution was achieved where affected riparian states had outstanding objections (Guideline 26).

The Report of the World Commission on Dams 271 Chapter 9

Stage 4: Project implementation: confirming compliance before commissioning

Intended outcome achieved before the project is commissioned and enters full operation. This includes the Clearance to commission the project is given broad commitments of the project developer by the relevant authority after all commitments are met. Relevant elements of performance as laid down in the project licence, compli- bond sureties are released. The operating ance plan, and related agreements, as well as licence is confirmed, including specific the commitments of the contractor acting as requirements for monitoring, periodic review and adaptive management. agent of the developer. Once the developer has met specified staged commitments, the associated financial sureties of any perform- Issuing the licence to operate will be contin- ance bond or outstanding contractual gent on compliance with mitigation measures payments may be released. in addition to technical requirements. The licence will contain a number of conditions Arrangements for public safety must be in for the operation stage, including compliance place in order to warn the downstream with operating rules, public notifications, dam population that sudden releases of water safety, monitoring and periodic review. All may occur and may be dangerous. Compen- contingencies cannot be anticipated, and a sation should be paid for any loss of liveli- balance should be struck on a case-by-case hood, such as the loss of fishing opportuni- basis between the necessary level of certainty ties. If test operations cause downstream and sufficient flexibility to accommodate inconvenience, compensation should also open, transparent, adaptive management. be paid – for example, if a test takes place during the dry season and damages recession Phasing of resettlement is required where agriculture plots. the reservoir is being impounded as the dam is constructed. Special attention is necessary A range of agreements on initial and long- to ensure that compensation and develop- term operation should be incorporated in ment measures are in place well in advance. the operating licence and provisions for implementation verified. Licence periods Reservoir filling, commissioning of produc- should not normally exceed 30 years. They tive capacity and the initial years of opera- should include: tion are critical phases that require special ■ agreement on environmental flow releases attention, intensive monitoring and contin- to the downstream river; ued dialogue with stakeholder groups. ■ releases to the downstream river for other Agreements on operating conditions are functions (navigation, water supply, required for three stages of commissioning: downstream irrigation); ■ reservoir filling; ■ operating rules during normal and excep- ■ test operation; and tional floods; ■ initial operation. ■ procedures for opening spillway gates; ■ monitoring and publication of relevant Full compliance with technical, environ- operation data; and mental and social measures must be ■ periodic review of operating rules.

272 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

Stage 4 Criteria Checklist

Needs Assessment Selecting Alternatives Project Preparation PROJECT IMPLEMENTATION Project Operation

Gaining Public Acceptance Recognising Entitlements and Sharing ■ Stakeholders participated in monitoring mitiga- Benefits tion measures and in negotiating outcomes that ■ The mitigation, resettlement and development affect them (Guidelines 1, 2). action plan has been implemented and disputes ■ Consultation mechanisms were agreed in resolved (Guideline 19). advance with stakeholders for any technical, social, environmental, or other problems that Ensuring Compliance may be encountered during reservoir filling ■ An independent panel reviewed and endorsed and commissioning. implementation of social, environmental, health ■ Contingency plans for emergency drawdown of and cultural heritage mitigation measures (Guide- the reservoir were agreed with stakeholders line 22). before commissioning and were widely ■ Preparations have been made to implement disseminated. licence conditions for operations, implement continuing mitigation measures, undertake Comprehensive Options Assessment monitoring and regular evaluation, and dissemi- ■ Affected stakeholders have reviewed any changes nate information. proposed to the tender design that substantially ■ Monitoring of social, environmental and techni- affect impacts, mitigation measures, benefit cal aspects includes an intensive phase to cover sharing, operational practices, or the monitoring the rapid changes that occur in the impoundment programme. and commissioning period. Addressing Existing Dams ■ The developer has complied with pre-commissioning commitments as defined in the Compli- ■ Institutional co-ordination mechanisms that ance Plan (Guideline 21). recognise interactive effects and cumulative impacts are in place to adjust operation of Sharing Rivers for Peace, Development and existing dams. Security Sustaining Rivers and Livelihoods ■ Mechanisms were initiated for sharing monitoring information with riparian provinces or States ■ Required environmental mitigation measures (Guideline 26). were implemented.

The Report of the World Commission on Dams 273 Chapter 9

Stage 5: Project operation: adapting to changing contexts

Intended outcome The evaluations would be: ■ comprehensive across all environmental, Conditions for operating under the licence are fulfilled and the licence conditions modified as social, economic, and institutional necessary to adapt to changing contexts. impacts; Monitoring programmes feed back into project ■ operation. A process is initiated to decide on integrated to cover the interactions reparations, if necessary. between impacts; ■ long-term to consider impacts over several decades; and The objectives of dam management need to ■ be transformed from purely technical to cumulative to reflect impacts of other development-oriented goals that include social structural and non-structural measures in and environmental considerations. This has the basin. major repercussions for operation, monitoring, and evaluation of both existing dams and In addition to periodic evaluations, parties dams implemented in the future. Adaptive other than the dam operator, or exogenous management is needed to continuously assess factors may prompt re-planning studies. For and adjust operational decisions within the example, a strategic or sectoral impact changing context of environmental, social, assessment may indicate the need for a re- physical and market conditions. This requires planning exercise for a group of projects. a close relationship between the local commu- Licence conditions should make provision nity, other stakeholders and dam owners and for such studies and for any required changes operators in order to minimise problems and to operating conditions through good faith quickly resolve any that do arise. negotiations that recognise the rights of all parties. Any re-planning studies should be Monitoring programmes should: done on the basin level. ■ include a full range of technical, environ- Licences should specify transparent processes mental, social, and economic parameters for stakeholder participation in operations and decided openly with the stakeholders; procedural requirements for monitoring and ■ have an intensive phase in the first five evaluation, safety inspection, contingency years, or after a major change in operation; planning and information disclosure. The ■ feed back into project operations; and regulatory authority or responsible govern- ■ be available to all stakeholders, perhaps in ment line agency must ensure compliance the form of an annual report. with provisions of the licence. Non-commercial aspects of the licence should be made A comprehensive project evaluation is public. Re-licensing procedures should exam- required three to five years after commission- ine present-day alternatives and be informed ing and at regular intervals thereafter (every by an integrated review of project performance five to ten years is suggested). The evaluation and impacts. Impact assessments should be is undertaken by the dam operator but is a undertaken for all major changes including stakeholder-driven process and may draw on decommissioning where dams are no longer the Commission’s case study methodology. required or are too expensive to maintain.

274 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

Stage 5 Criteria Checklist

Needs Assessment Selecting Alternatives Project Preparation Project Implementation PROJECT OPERATION

These criteria are relevant to both existing dams (Chapter 8, Strategic Priority No.3) and the operational stages of future dams.

Gaining Public Acceptance Recognising Entitlements and sharing benefits ■ Stakeholders are identified for consideration of operational issues and any proposed changes that ■ Detailed benefit-sharing mechanisms are modi- impact on them or the environment (Guideline 1). fied as necessary with the agreement of affected groups (Guideline 20). Comprehensive Options Assessment Ensuring Compliance ■ Periodic evaluations of all aspects of project operation and performance are undertaken with the ■ Adverse social and environmental impacts and involvement of the stakeholder forum every 5 to 10 reparations issues are referred to the appropriate years and agreements renegotiated as necessary. recourse body (Guideline 19). ■ Modernisation programmes and alternative ■ Annual reports of project monitoring pro- operational regimes are considered as part of grammes, including social and environmental periodic reviews, replanning, or relicensing aspects, are issued promptly and corrective exercises through a participatory multi-criteria measures are initiated to address issues raised in approach (Guideline 13). the reports. ■ Monitoring and evaluation programmes should ■ The requirements of remaining performance explicitly consider the influence of climate bonds or trust funds outlined in the Compliance change (namely increasing and decreasing rainfall Plan are periodically reviewed, and financial and flows) on benefits and dam safety. guarantees are released on satisfactory compli- ■ A full feasibility study, including analysis of ance with agreed milestones (Guideline 23). alternatives and impact assessment, is undertaken ■ Dam safety and inspection programmes are for any proposal for any major physical change, implemented. including decommissioning. Sharing Rivers for Peace, Development Sustaining Rivers and Livelihoods and Security ■ Operations take account of environmental flow ■ Mechanisms exist to share monitoring informa- requirements (quantity and quality) and ecosystem tion and resolve issues as they arise. and social impacts are monitored (Guideline 15).

The Report of the World Commission on Dams 275 Chapter 9

A Special Case: Dams in ■ provide support to vulnerable and the Pipeline disadvantaged stakeholder groups to participate in an informed manner; The strategic priorities and policy principles ■ undertake a distribution analysis to assess outlined in Chapter 8 are as relevant to who shares the costs and benefits of the projects already at an advanced stage of project; planning and development as they are to ■ develop agreed mitigation and resettle- the selection of a project in the earlier ment measures to promote development options assessment stage. opportunities and benefit-sharing for Currently a large number of dam displaced and affected people; The Commission’s projects are at various stages of Knowledge Base has planning and development. ■ avoid, through modified design, any severe demonstrated that it is While acknowledging that and irreversible ecosystem impacts; never too late to delays in implementation can ■ provide for an environmental flow improve outcomes. cause unacceptable delays in requirement and mitigate or compensate delivering intended benefits, the any unavoidable ecosystem impacts; and WCD Knowledge Base has ■ design and implement recourse and demonstrated that it is never too late to compliance mechanisms. improve outcomes. On this basis, the Commission proposes an open and partici- Governments may also use the review of patory review of ongoing and planned ‘dams in the pipeline’ as an opportunity to projects to ascertain the extent to which compare the existing policy framework for project formulation can be adapted to planning and implementation of water and accommodate the principles outlined in this energy options with the criteria and guide- report. lines proposed by the Commission. This can serve to launch a process of internal review The essence of the process is that stakehold- and modification of existing policies and er groups should have an opportunity to legislation, and reinforcement of appropriate define the scope of the review and to capacity that will facilitate implementation propose changes in keeping with the Com- of the Commission’s recommendations in mission’s recommendations. The extent of future. any additional study or changes in project configuration would depend on the stage of At specific stages of planning and project planning, design, or construction and be development, regulators, developers and, determined from a synthesis of the stake- where appropriate, financing agencies holder consultations and, where appropriate, should ensure that the following project- an inter-ministerial review. General actions related points are addressed: to guide the review for all projects would Projects at feasibility stage include: ■ undertake a stakeholder analysis based ■ The stakeholder forum confirmed that on recognising rights and assessing risks the set of options considered was appro- to identify a stakeholder forum that is priate, or identified other alternatives to consulted on all issues affecting stake- consider as part of the project impact holders; assessment.

276 Dams and Development: A New Framework for Decision-Making Criteria and Guidelines – Applying the Strategic Priorities

■ Any bias in selection of alternatives is removed or justified in a transparent fashion (for example subsidies to particular sub-sectors or groups). ■ Demonstrated public acceptance exists for the recommended options. ■ The assumptions underlying the economic, financial, and risk analysis are justified and subject to sensitivity analysis. ■ Mechanisms for benefit-sharing are identified. ■ An environmental flow requirement is determined.

Projects at detailed design stage ■ The stakeholder forum is consulted on decisions related to project layout, operation and mitigation and develop- ■ An adequate social, environmental and ment measures and relevant agreements technical monitoring plan is financed by are negotiated with affected groups. the developer. ■ The operating rules and ■ Environmental flow requirements are Governments may also commissioning plan are determined and incorporated into the use the review of ‘dams agreed with a stakeholder design and operation rules. in the pipeline’ as an forum. ■ A Compliance Plan is prepared, and opportunity to compare ■ recourse mechanisms are identified. A comprehensive post- the existing policy project review is agreed for ■ Compliance mechanisms are provided framework for planning two to three years after for in tender documents. and implementation of commissioning, and every water and energy ■ Benefit-sharing contracts are negotiated five to ten years thereafter. options with the criteria for displaced and project affected people. and guidelines ■ A process for stakeholder involvement This process of review implies proposed by the during operation is established. added investigations or commit- Commission. ments, the re-negotiation of Projects under construction contracts and the incorporation ■ The record of compliance is reviewed of a Compliance Plan. As in the case of and a compliance plan is developed for initial planning, the additional financial remaining mitigation measures. costs incurred will be recouped in lower ■ Existing commitments for resettlement overall costs to the operator, to government, and benefit-sharing are converted into and to society in general as a consequence of formal contracts. avoiding negative outcomes and conflicts.

The Report of the World Commission on Dams 277 Chapter 9

A Set of Guidelines for international guidance and current good Good Practice practice. Further information is available on many of these aspects in the WCD Knowl- The guidelines outlined here describe in edge Base. general terms how to assess options and plan and implement dam projects to meet the The guidelines are presented under the same Commission’s criteria. The 26 guidelines sub headings as the Commission’s seven add to the wider range of technical, finan- strategic priorities. There are clear linkages cial, economic, social and environmental between individual guidelines and cross guidelines. They are advisory tools to references to them are given in the criteria support decision-making and need to be checklists for the key decision points of the considered within the framework of existing planning and project cycles.

Strategic Priority 1: Gaining Public Strategic Priority 4: Sustaining Acceptance Rivers and Livelihoods 1 Stakeholder Analysis 14 Baseline Ecosystem Surveys 2 Negotiated Decision-Making Processes 15 Environmental Flow Assessment 3 Free, Prior and Informed Consent 16 Maintaining Productive Fisheries

Strategic Priority 2: Comprehensive Strategic Priority 5: Recognising Options Assessment Entitlements and Sharing Benefits

4 Strategic Impact Assessment for Envi- 17 Baseline Social Conditions ronmental, Social, Health and Cultural 18 Impoverishment Risk Analysis Heritage Issues 19 Implementation of the Mitigation, 5 Project-Level Impact Assessment for Resettlement and Development Action Environmental, Social, Health and Plan Cultural Heritage Issues 20 Project Benefit-Sharing Mechanisms 6 Multi-Criteria Analysis Strategic Priority 6: Ensuring 7 Life Cycle Assessment Compliance 8 Greenhouse Gas Emissions 21 Compliance Plans 9 Distributional Analysis of Projects 22 Independent Review Panels for Social 10 Valuation of Social and Environmental and Environmental Matters Impacts 23 Performance Bonds 11 Improving Economic Risk Assessment 24 Trust Funds 25 Integrity Pacts Strategic Priority 3: Addressing Existing Dams Strategic Priority 7: Sharing Rivers 12 Ensuring Operating Rules Reflect Social for Peace, Development, and and Environmental Concerns Security 13 Improving Reservoir Operations 26 Procedures for Shared Rivers

278 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

Gaining Public Acceptance

1. Stakeholder Analysis ■ Identify those at risk through vulnerability or risk analysis and consider them as core The absolute value and the relative signifi- stakeholders, including those who face cance of ‘stakes’ vary, especially in what risk to their livelihoods, human rights, they represent for the interested party. and property and resource rights. Special Stakeholders have unequal power and this attention should be given to indigenous can affect their ability to participate in and and tribal peoples, women and other influence decisions. A stakeholder analysis vulnerable groups as they may face based on recognising rights and assessing greater risks from development interven- risks should be used to identify key stake- tions (Guideline 3). In the case of a dam, holders for planned activities. The analysis the analysis should include those up- will also seek to understand and address stream, downstream and in the proposed potential factors that may hinder their reservoir area. Relevant civil society involvement. The analytic approach can groups or scientists are included in a involve stakeholder workshops, community- stakeholder forum to ensure that environ- level surveys, key informant surveys, and mental risks, for which there may be no literature review. champion, are adequately reflected and discussed. The stakeholder analysis leads to the ■ constitution of a temporary stakeholder Identify constraints to establishing a level playing field for stakeholder involvement. forum as a basis for participation and, where The use of capacity building, institution- relevant, negotiation processes throughout al strengthening, quota systems (for the planning and project cycles. A stake- example, to ensure proper representation holder forum is a dynamic construct and will of vulnerable groups such as women), or need to be applied to meet changing needs support mechanisms, such as NGOs or through the planning and project cycles independent facilitators to correct any beginning with needs assessment/verifica- imbalance of influence should be ex- tion and options assessment. The composi- plored. Financial support may be neces- tion of a stakeholder forum, the level of sary to ensure adequate participation. representation of various interests, and the means of facilitating the process changes The government planning body sponsoring from stage to stage. the planned interventions is responsible for initiating the stakeholder analysis leading to The stakeholder analysis will: constitution of a forum and will participate ■ Recognise existing rights and those who hold in it. The final structure and composition of them. Those groups whose livelihoods, a stakeholder forum should be decided in a human rights and property and resource consultative process. The assistance of rights may be affected by an intervention independent facilitators may be helpful in are major rights holders and thus core achieving this outcome. stakeholders in a stakeholder forum within which negotiated outcomes A stakeholder forum is therefore formed of should be achieved. individuals representing various groups and

The Report of the World Commission on Dams 279 Chapter 9

interests. Such structures may exist already Effective participation in a stakeholder and their capacity can be strengthened or forum must be facilitated through timely modified. Where such structures do not exist, access to information and legal and other a stakeholder forum is established as the necessary support. This is particularly the representative body of the stakeholders. The case with indigenous and tribal peoples, extent to which a stakeholder forum should be women, and other vulnerable groups. formalised to enable representation for the groups identified through a rights-and-risks Sufficient time must also be allowed for the approach will depend on country contexts. wider body of stakeholders to examine The status of a forum, and the selection of its information and to consult amongst them- representatives, should however ensure selves before decisions are made. Where effective participation for all interested and dispute resolution mechanisms are required affected parties and accommodate changes for negotiated processes, see Guideline 2. over time.

2. Negotiated Decision- stakeholders should be as free as possible Making Processes from external manipulation. Communi- ties may legitimately decide to discontin- A negotiation process is one in which stake- ue their involvement in the process if holders – identified through the Stakeholder their human rights are not respected or Analysis (see Guideline 1) – have an equal in the event of intimidation. opportunity to influence decisions. Negotia- tions should result in demonstrable public ■ Adequate time is allowed for stakeholders acceptance of binding and implementable to assess, consult and participate. agreements and in the necessary institution- ■ Special Provisions for Prior Informed al arrangements for monitoring compliance Consent. In negotiations involving and redressing grievances. All stakeholder indigenous and tribal peoples, mecha- forum members should share a genuine nisms to resolve disputes should follow desire to find an equitable solution and procedures recommended by the Com- agree to be bound by the consensus reached. mission (Chapter 8, Strategic Priority 1.4 Attributes of a fair negotiation and Guideline 3). process ■ Addressing Power Imbalances. Authorities ■ The Representation of Stakeholders in the should make available adequate financial stakeholder forum is assured through a resources to enable stakeholder groups free process of selection, ensuring the who are politically or financially weak, effective and legitimate representation of or who lack technical expertise or all interests. organised representation to participate ■ The Integrity of Community Processes effectively in the process. These resourc- should be guaranteed through assurances es may include financial support to that they will not be divided or coerced, representatives for logistics, for income recognising that differences and internal foregone, for capacity building and for conflicts may arise. The process and the requesting specific technical advice.

280 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

■ Transparency is ensured by jointly defin- ■ guarantee access to all relevant informa- ing criteria for public access to information to the stakeholder forum in an tion, translation of key documents and appropriate language; and by holding discussions in a language ■ at the outset, agree on the timeframe for local people can understand. the key milestones within the decision- ■ Negotiations are assisted by a facilitator or making process. mediator, where stakeholders request it, selected with the agreement of the Compliance with the process outlined above stakeholders. will be a fundamental consideration in determining whether the negotiations For this to be a legitimate process, the stake- process was conducted in good faith. holders should: ■ agree on the appropriate structures and When a negotiated consensus cannot be processes for decision-making, the required achieved through good faith negotiations mechanisms for dispute resolution (including within the agreed-upon timeframe, the any third party involvement), and the cir- established independent dispute resolution cumstances in which they will be initiated; mechanisms are initiated. These may ■ agree that the interests at stake and include amicable dispute resolution, media- legitimate community needs are clearly tion, conciliation and/or arbitration. It is identified, in particular on the basis of important that these are agreed upon by the relevant rights and risks; stakeholder forum at the outset. Where a ■ ensure that the available alternatives, settlement does not emerge, the State will their relevant consequences and uncer- act as the final arbitrator, subject to judicial tainties are given full consideration; review.

3. Free, Prior and Indigenous and tribal peoples are not Informed Consent homogeneous entities. PIC should be broadly representative and inclusive. The Free, prior and informed consent (PIC) of manner of expressing consent will be guided indigenous and tribal peoples is conceived as by customary laws and practices of the more than a one-time contractual event – it indigenous and tribal peoples and by nation- involves a continuous, iterative process of al laws. Effective participation requires an communication and negotiation spanning appropriate choice of community represent- the entire planning and project cycles (see atives and a process of discussion and Chapter 8, policy principle 1.4). Progress to negotiation within the community that runs each stage in the cycle – options assessment parallel to the discussion and negotiation including priority setting and selection of between the community and external actors. preferred options, and preparation, imple- At the beginning of the process, the indige- mentation and operation of the selected nous and tribal peoples will tell the stake- option – should be guided by the agreement holder forum how they will express their of the potentially affected indigenous and consent to decisions including endorsement tribal peoples.1 of key decisions (Guideline 1).

The Report of the World Commission on Dams 281 Chapter 9

An independent dispute resolution mecha- guidelines or frameworks, as these must be nism to arrive at a mutually acceptable negotiated as the process proceeds. The agreement should be established with the outline presented in Guideline 2: Negotiated participation and agreement of the stake- Decision Making Processes is intended to offer holder forum, including the indigenous and some basic direction to those involved with tribal peoples, at the beginning of any such processes and independent dispute process. It is inappropriate to set rigid resolution bodies.

Comprehensive Options Assessment

4. Strategic Impact ronmental assessments (EA), basin-wide Assessment for EAs, regional EAs, and cumulative EAs.2 Environmental, Social, SA should be concerned with the uses and Health and Cultural impacts of existing water and energy Heritage Issues projects, as well as alternatives for meeting future needs. In practice, SA may have Strategic impact assessment (SA) is a different levels of detail, depending on relatively recent tool that can be used to where it is applied. At one level, the SA provide a new direction to planning process- would scan and identify priority issues to be es. It provides an entry point that defines addressed subsequently in more detailed who is involved and maps out the broad planning exercises. For example, the SA issues to be considered. The Commission would identify whether evaluations of proposes that the SA process starts by recog- existing projects have been undertaken, or nising the rights to be accommodated, assess- whether outstanding social issues on specific ing the nature and magnitude of risks to the projects had been addressed. At this level, environment and affected stakeholder the SA would also assess whether a suffi- groups, and determining the opportunities cient range and type of options are being offered to these groups by different develop- considered in the different planning process- ment options (Guideline 1). It should also es to meet future needs. In cases where the identify where conflicts between various SA is more elaborate and detailed, the rights exist and require mediation. exercise may be extended to host a generic SA takes the concept of project level impact options assessment process using stakeholder assessment and moves it up into the initial groups and multi-criteria formulations to phases of planning and options assessment. screen and rank options (Guideline 6). It is a broad assessment covering entire The general goals of SA include: sectors, policies and programmes, and ensures that environmental, social, health ■ recognising the rights of stakeholders and cultural implications of all options are and assessing the risks; considered at an early stage in planning. It is ■ incorporating environmental and social a generic term that includes a range of criteria in the selection of demand and planning tools for example, sectoral envi- supply options and projects before major

282 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

funds to investigate individual projects existing dams and other assets from are committed; economic, technical, social and environ- ■ screening out inappropriate or unaccept- mental perspectives. able projects at an early stage; SA must be revisited at appropriate intervals ■ reducing up-front planning and prepara- with periodic ‘state-of-the-sector’ reporting. tion costs for private investors and Important variables determining the frequency minimising the risk that projects en- and intensity of this on-going process include counter serious opposition due to envi- developments in the economy, in technology, ronmental and social considerations; and in demography and in public opinion. Review ■ providing an opportunity to look at the of SA reports at the highest political level (for option of improving the performance of example Parliament) is desirable.

5. Project-Level Impact rently and interactively with regular Assessment for exchange of information between all study groups. Environmental, Social, ■ IAs should be carried out independently Health and Cultural of the interests of the project developer Heritage Issues and financing mechanisms should reflect Project-level impact assessment (IA) is this independence. already standard practice in many countries, ■ IA should include an Environmental and the term is used here to include envi- Impact Assessment, a Social Impact ronmental, social, health and cultural Assessment, a Health Impact Assessment impacts. Deficiencies in past implementa- (see Box 9.1), and Cultural Heritage tion have been identified and improved Impact Assessment (see Box 9.2) as processes are needed.3 explicit components and comply with international professional standards. The The following changes are proposed to the assessments should be sufficiently de- way project level IAs are implemented: tailed to provide a pre-project baseline ■ Projects should be subject to a two-stage against which post-project monitoring IA: the first is a scoping phase, including results can be compared. full public participation, that identifies ■ An independent panel of experts (re- key issues of concern and defines the porting formally to the highest environ- terms of reference for the second, assess- mental protection authority) should be ment, phase (Guideline 1). appointed to assist the government and ■ The timing of the IA should allow the the developer in reaching sustainable results to feed into the final design of the social and environmental outcomes project. There should be a total integra- (Guideline 22). The developer must tion of technical, environmental and respond to all issues raised by the panel social studies during the design stage. and explain how they will be addressed. Although executed by different study The panel’s findings and the developer’s groups, these studies should run concur- response are to be made public within a

The Report of the World Commission on Dams 283 Chapter 9

reasonable period (for example, six ■ The IA process continues through and weeks). beyond project construction and ade- ■ Developers should open a local liaison quate institutional and financial arrange- office to ensure adequate access to ments for social and environmental audit information for local affected communi- and monitoring should be included in ties in appropriate languages. the planned measures. Contracts with monitoring agencies (for example research ■ The IA process should culminate in a institutes or NGOs) should be agreed prior series of written agreements with those to tendering for construction. departments or organisations that are required to implement mitigation, ■ A redress procedure should be put in development and compensation plans, or place that provides mechanisms for respond to the impacts. The scope of addressing grievances during the resettle- these agreements must be fully defined ment plan and following construction. prior to tendering for construction. ■ IAs should be public documents, posted on relevant websites, and disseminated in appropriate languages. ■ IA should be guided by the precautionary Box 9.1: Health impact assessment approach. A health impact is a change in health risk reasonably attributable to a project, programme or policy. A health risk is the likelihood of a The precautionary approach requires States health hazard or opportunity affecting a particular community at a and water development proponents to particular time. Health Impact Assessment (HIA) is part of the overall risk assessment process (economic, social and environmental exercise caution when information is risks) to assess the viability of a project. uncertain, unreliable, or inadequate and The Health Impact Assessment has the following components. when the negative impacts of actions on the ■ Assessing the health condition of people in reservoir, infra- environment, human livelihoods, or health structure, downstream, resettlement, irrigation, and other are potentially irreversible. A precautionary impact areas. The assessment process should engage local approach entails improving the information people and resources. The parameters derived from the information collected constitute the baseline health situation base, performing risk analysis, establishing of the population. precautionary thresholds of unacceptable ■ Predicting changes in health determinants that can be impacts and risk, and not taking actions reasonably attributed to the project and that could affect with severe or irreversible impacts until people during each stage of the project. The changes, taken adequate information is available or until together, produce health outcomes or changes in health states. These are expressed in a minimum of three ranks: no the risk or irreversibility can be reduced, change, increased health risk, and health enhancement. making outcomes more predictable. Nor- Factors determining health outcomes in past projects involv- mally the burden of proof will be on the ing comparable social, economic and environmental conditions can be used to enhance predictability. developer. ■ Assessing the cost of preventing and mitigating the potential health impacts in the overall cost assessment of the project. Decision-makers are faced with the dilemma ■ Developing measures to prevent, minimise and mitigate of how to reconcile competing or conflicting health impacts with the participation of the potentially rights and needs. The precautionary ap- affected people and incorporating these into contractual proach forms part of a structured approach implementation arrangements with adequate financial provision. to the analysis of risk, as well as being Source: WHO, 1999, WCD Working Paper relevant to risk management. Determining what is an acceptable level of risk should be undertaken through a collective political

284 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

process. The process should avoid unwar- Box 9.2: Cultural heritage impact assessment ranted recourse to the precautionary ap- Cultural heritage resources are the cultural heritage of a people, a nation proach when this can overly delay decision- or humanity as a whole, and can be on the surface, underwater or making. However, decision-makers faced underground. They comprise: ■ with scientific uncertainty and public Cultural practices and resources of current populations - religions, languages, ideas, social, political and economic organisations, and concerns have a duty to find answers as long their material expressions in the forms of sacred elements of natural as the risks and irreversibility are considered sites, or artefacts and buildings; unacceptable by society. ■ Landscapes resulting from cultural practices over historical and prehistoric times; and ■ Archaeological resources, including artefacts, plant and animal remains associated with human activities, burial sites and architectural elements. 6. Multi-Criteria Analysis Cultural Heritage Impact Assessment (CHIA) requires adequate time for Multi-criteria analysis (MCA) processes use a successful completion and should be looked at in two stages. Firstly, where regions and river valleys are known to be rich in cultural resources, mix of qualitative and quantitative criteria to landscapes, or archaeological resources, consideration of these elements assess and compare options that may be should be included in Strategic Impact Assessments (Guideline 4) and policies, programmes, or projects. Stakeholder- used as a criterion in selecting options and avoiding impacts. Secondly, a project level mitigation plan is developed where a dam option proceeds driven MCA processes are flexible and open, to full feasibility phase. based on the concept of a stakeholder forum The following procedural aspects need to be considered: (see Guideline 1). Their primary purpose is to ■ financial resources should be specifically allocated to CHIA; provide a structured process to screen and rank ■ the assessment team should include archaeologists and, if necessary alternatives and help understand and resolve architects and anthropologists; differences between groups of stakeholders ■ where cultural assets have significant spiritual or religious significance all involved in development decisions.4 activities should be planned with the consent of relevant communities; ■ assessments should culminate in a mitigation plan to address the cultural heritage issues identified through minimising impacts, or The multi-disciplinary planning team with a through curation, preservation, relocation, collection or recording; and time-bounded mandate supports all phases ■ a separate report should be produced as a component of the overall IA of the MCA process. The results at each process. stage should be made available to the Source: Brandt and Fekri, 2000, WCD Working Paper stakeholder forum and for wider public review. A public hearing Figure 9.3 Preference matrix for ranking options should be held on the outcome at High 100 each stage if the significance of Attractive Projects the process warrants it. If the to Proceed to Further Study or MCA process cannot resolve all Development conflicts, the use of MCA may still assist in identifying policy scenarios and the way different 50 groups and interests perceive them. There are many ways of presenting Projects Dropped the results of the MCA. One of from Further them is the preference matrix, Consideration which demonstrates the equal Environmental and Social Preference Environmental 0 50 100 emphasis given to social, environ- Score) (Composite Environmental/Social Low High mental, technical and economic Technical Preference (Composite Technical/Economic Score) parameters (see Figure 9.3).

The Report of the World Commission on Dams 285 Chapter 9

Attributes and steps of MCA processes to Step 5: select the best mix of options are: Options are screened by the planning team Step 1: according to the agreed criteria, results are presented to representatives of the stakeholder The sponsoring agency prepares terms of group for approval and subsequently an- reference for the overall process and a stake- nounced for wider public review or comment. holder analysis, and establishes an information centre. Representative stakeholder groups are Step 6: contacted, and the general public is informed through print and electronic media. Sequential steps of coarse and fine ranking of options (where the number of options is Step 2: large) are prepared by the planning team A stakeholder forum is formed and repre- and submitted to the representatives of the sentatives of stakeholder groups identified stakeholder forum at each stage. The list of subject to public review and comment. A options at each stage is made public and an multi-disciplinary planning team is formed adequate period for comment is provided to support the process and assembles an between each stage. Public hearings may be initial inventory of options. held at each stage if appropriate. Step 3: Step 7: Public comment is invited on the options The final selection of options that would inventory including proposals for additional form the basis for detailed planning is options to be considered. The stakeholder forum confirms the comprehensiveness and presented to agencies, communities, or adequacy of the options inventory. Where groups responsible for the detailed planning. necessary, additional steps are taken to expand the inventory. These steps lead to preparation of a limited set of diverse development plans comprising Step 4: a range of options emerging from the The stakeholder forum decides on the criteria screening process. The multi-criteria exer- for screening the options and criteria for cise may be repeated to evaluate these coarse and fine ranking of options are estab- alternative plans and select a preferred lished with input from the planning team. development plan.

7. Life-Cycle Assessment deciding which options to include in subsequent stages of planning. Alternatively, Life-cycle assessment (LCA) is an options the information generated by LCA is used assessment procedure used in the energy to develop regulatory policies, for example sector to compare ‘cradle-to-grave’ perform- policies addressing barriers limiting the ance, environmental impacts, and market market penetration of options otherwise barriers and incentives for different demand considered to be in the interest of society.5 and supply options. LCA is located at the front end of the planning cycle. Its results LCA assessments can be simple and generic may be fed into multi-criteria screening and or exhaustively detailed, data-rich, and ranking processes, which are a basis for elaborate. LCA procedures that quantify the

286 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

potential impacts of different options on indicators (for example net efficiencies, land, air and water resources, including the consumption of resources, or the greenhouse gas (GHG) emissions, can be impact per unit of output of the option – transferred and adapted to different such as land use, water use, GHG emis- countries. The analytical framework used to sions, and other gaseous, liquid or solid assess the direct, indirect, and hidden pollution streams); and incentives and market barriers for different ■ identification of the range and magni- options through the full chain of tude of the direct, indirect, and hidden development is also transferable. subsidies, external factors and incentives across each stage of the life cycle of each LCA would typically include: option. ■ categorisation of the different stages in the life cycle of each option where the The most advanced use of LCA is in the impacts and effects are relevant (for power sector, where it is particularly used example from resource extraction to consider the GHG emissions of various through transport, manufacturing, options. These factors are becoming the building, operation and refurbishment to prime driving force behind energy and decommissioning); power sector policies in many countries ■ identification of the material flows and including Europe, Australia and Canada, resource impacts at each stage and and reflect the Kyoto Protocol (Guideline comparison of each option using a set of 8).

8. Greenhouse Gas February 2000 decided that net emissions Emissions from reservoirs above baseline emissions are the appropriate estimates. To calculate net Recent research indicates that reservoirs can emissions, the planner must: emit greenhouse gases. Precise assessments ■ assess the carbon (CO , CH ) and are especially important to assist in selecting 2 4 nitrogen cycles (N O) in the pre-im- climate-friendly options and if hydropower 2 projects seek to benefit from any form of poundment watershed context – this carbon credit. The emissions from the involves establishing a carbon budget, natural pre-impoundment state should be including description of flow rates, included in the comparison with other concentrations, residence time and other options. Good field studies with modelling relevant measures; predictions of emissions should be an ■ assess future changes to carbon inputs in explicit component of relevant feasibility the watershed from various activities, studies.6 including deforestation; ■ assess the characteristics of proposed Procedures to calculate emissions for con- reservoir(s) and inundated area(s) that ventional and renewable options are well will change the carbon cycle, including established and available but are continually size, temperature, bathymetry, primary evolving. An expert workshop convened by productivity and other relevant measures the Commission and held in Montreal in after dam completion; and

The Report of the World Commission on Dams 287 Chapter 9

■ assess the cumulative emissions from understanding of the scale of GHG emis- multiple dams on a watershed basis in sions to temperate and semi-arid regions of cases where a dam and its operations are the world and to catchments with large linked to other dams. urban populations. Such data will prove More baseline measurements are required on helpful in taking informed decisions on reservoirs on existing projects to extend energy options and climate change.

9. Distributional Analysis direct costs and benefits of the project, of Projects including those external social and environmental impacts that are to be valued Distributional analysis provides stakeholders (Guideline 10). This builds on the finan- and decision-makers with information on cial and economic cost-benefit analyses. who will gain and lose from a project and is an essential tool in promoting more equita- Selecting options: Integrated distributional ble distribution of benefits and costs.7 These analysis at a preliminary level should be gains and losses may be expressed in eco- initiated during the early stages of screening nomic or financial terms, or they may be and selecting options as part of the strategic more simply expressed as changes in physical impact assessment. It can be carried forward quantities. In some cases only the direction of at an increasing level of detail for projects a specific impact may be discernible. that emerge for further consideration from this process. At the preliminary level of Integrated distributional analysis requires analysis, a matrix is prepared to identify the assessment of the full range of project groups that will either receive benefits or impacts including financial, social, environ- bear the costs of the project and indicate the mental and economic aspects assessed either approximate scale of such costs or benefits. in a qualitative fashion, quantified in non- A qualitative equity assessment should also monetary terms, or valued in financial or be undertaken and inform the screening economic terms. A number of methods process about the comparative impacts of focusing on specific aspects of distribution alternatives on vulnerable groups in society. can be used within the overall approach at different stages of the planning cycle. Feasibility stage: A more detailed and integrated distributional analysis should be ■ Equity (or poverty) assessment comprises undertaken during the feasibility study and an assessment of the impacts (in economic include both an economic distribution or non-economic terms) and risks of a analysis and equity assessment. The use of a project on specific sub-populations or macroeconomic or regional analysis is groups of concern. recommended for projects with a significant ■ Macroeconomic or regional analysis irrigation component or inter-basin transfer includes an analysis of the wider economic where there are broader objectives in terms impacts using either a simple economic or of redistributing income between regions or fiscal impact analysis or a formal regional making a sustained contribution to the or macroeconomic model. macroeconomy. The distributional analysis ■ Economic distributional analysis includes should be undertaken in full consultation an explicit analysis of distribution of the with project stakeholders.

288 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

10. Valuation of Social and Many of these valuation methods are Environmental Impacts equally applicable in the developing world and capacity to apply them increased rapidly The methodologies and applications to in the 1990s. They have been adapted to value environmental and social impacts of the rural, developing context, particularly in dams can be used to ensure that impacts are combining participatory approaches with internalised in the economic analysis where valuation methods and integrating econom- appropriate and possible (see Table 9.1). ic valuation into multi-criteria analysis. Where it is undesirable or not possible to Many of the external impacts of large dams express such impacts in economic terms, affect household livelihoods and thus should they should be considered separately as be assessed using relatively straightforward parameters in the multi-criteria analysis (see market or revealed-preference methods. In Guideline 6). particular, a series of relatively straightforward methods such as productivity and Expertise and experience with these meth- substitute-goods methods may be applied to ods in industrialised countries are wide- estimate how changes in water quantity, spread, and many examples exist of their quality, and flow regime affect household application to the impacts of dams.8 Typi- productivity and consumption. These cally, valuation of the impacts of new dams methods also apply to the impact of changes or the decommissioning of old dams in such in water flows on downstream communities countries deals with recreation, tourism, and their natural resources, as well as impacts fisheries and, increasingly, people’s prefer- on major ecosystem functions and services ence for healthy ecosystems. where these provide an economic good. For example, sediment flows and deposition along A range of methods is available, including the coast, which if interrupted, may lead to a those based on observed market behaviour, need for erosion control measures. the stated preferences of individuals, or modelling of choices made by respondents Studies of this nature should involve at least three steps: in market surveys. Their purpose is to value previously hidden costs and benefits and ■ a scoping exercise to identify and select make them explicit in decision-making. impacts to be valued; Whatever the context, the derivation of ■ valuation studies; and monetary values for the unmitigated environmental and social impacts of projects is Table 9.1: Valuation methods necessary when it assists the transparent, Observed Behaviour Hypothetical Behaviour participatory, and explicit examination of Direct Market Prices Stated Preferences project and policy alternatives. Whether Competetive market prices Contingent valuation these include the valuation of cultural, Shadow pricing (dichotomous) choice, biodiversity or other intangible values in willingness-to-pay, bidding games monetary terms will depend on the local Indirect Revealed Preferences Choice Modelling Productivity methods Contingent referendum context and on stakeholders’ views. As Avertive (defensive) expenditure Contingent ranking noted, such aspects are often better ad- Travel cost Contingent behaviour Hedonic pricing Contingent rating dressed as an individually weighted compo- Substitute goods Pairwise comparisons nent in a multi-criteria analysis. Source: Freeman, 1993; Pearce, pers. comm. 2000; Barbier, pers. comm. 2000.

The Report of the World Commission on Dams 289 Chapter 9

■ public meetings to report back to the (see Guideline 5). Finally, the information stakeholder forum on the results of the generated through valuation studies should studies. have an explicit role in informing not only applicable cost-benefit and distributional The scoping exercise may be incorporated into analyses, but also the negotiations between the initial stage of project impact assessment stakeholders and decision-makers.

11. Improving Economic For example, in determining economic risk, Risk Assessment the probabilities of different values for inflows and power generation can be com- All infrastructure projects and commercial bined with probabilities of cost overruns in a undertakings involve risk, uncertainty and cost-benefit analysis to result in a probabili- irreversibility. Project risk assessments ty distribution of net returns. This provides generally take into account technical, a robust assessment of the risk of different economic and financial aspects.9 The outcomes (see Box 9.3). It introduces a Commission recognises the nature of social more effective approach than the simple and environmental risks and that these can sensitivity analysis used to assess the effect be addressed through other mechanisms of potential changes in important variables (Guidelines 4, 5, 18). where the cost-benefit analysis may have been re-run for a number of individual The following are recommended as a general scenarios. For example, the sensitivity approach for technical, financial and analysis is used to see whether the project is economic risk assessment: still profitable when the planned project ■ the assessment of risks should be includ- costs increase by 20%. ed in all steps of the planning cycle; In implementing these general recommen- ■ identification and selection of risks for dations on economic risk analysis across the assessment should be undertaken as part planning cycle, a number of specific sugges- of the larger stakeholder and multi- tions should be considered. criteria processes; ■ past performance of large dams should be At all stages: used to identify likely ranges for the ■ improved prediction of project costs by variables and values to be included in using a frequency distribution of the cost risk and sensitivity analysis; and overruns for similar projects. ■ sensitivity analysis should be complement- ed by a full probabilistic risk analysis. At options assessment stage: ■ a simple sensitivity analysis using agreed Good practice involves the use of probabil- value ranges for key variables; and istic risk analysis, a quantitative technique ■ a qualitative comparison of options that employs the probability distributions of under consideration in terms of the individual variables to produce a consolidat- uncertainty associated with the cost and ed single probability distribution for the benefit streams of each project. criteria of interest.

290 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

At the feasibility stage Box 9.3: Ghazi-Barotha, Pakistan ■ a full probabilistic risk analysis of economic profitability; The World Bank appraisal of the Ghazi-Barotha hydroelectric project in Pakistan used a probabilistic risk assessment of the economic rate of ■ modelling of changes and variability in return (EIRR) of the power expansion programme, with the risks hydrological estimates that may result summarised under four scenarios: demand uncertainties, cost profiles, schedule delay, and amount of additional capacity provided by private from climate change and their effect on projects. Each scenario has three alternative states. Probabilities were delivery of services and benefit flows; assigned to each scenario so that a weighted average EIRR could be and obtained. This yielded 54 total possible outcomes. For each one the expected value of the EIRR, calculated as probability times its own ■ investigation of the likely benefits of risk EIRR, is then summed over all outcomes to give the expected EIRR. A reduction measures and the costs this probability distribution of EIRR was then calculated for the overall power sector programme and for the project alone. entails. The results indicate that the risk-weighted EIRR on the overall investment programme is 18.5%. This is lower than the Base Case estimate, but considerably higher than the opportunity cost of capital at 12%. The probability of the EIRR falling below the opportunity cost of capital is estimated at 8%. The risk-weighted EIRR proved quite robust to changes in the basic probabilities. Source: World Bank, 1995.

Addressing Existing Dams

12. Ensuring Operating The compensation would be limited to those impacts caused by the breach, over Rules Reflect Social and and above the natural flood event. Environmental Concerns Releases to the downstream river to satisfy drinking water and environmental require- Agreements on operating conditions should ments should be maintained during river reflect commitments to social and environ- diversion. If, for technical reasons, flow is mental objectives in addition to the com- interrupted, the operator must guarantee mercial interests. At all times the safety and that alternative supplies of drinking water well being of the people affected must be will be made available to the downstream guaranteed throughout the project cycle. All population. operating agreements should be available to stakeholder groups. Reservoir filling River diversion during construction During the reservoir filling period, there Emergency warning and evacuation plans should be releases of good quality water to are needed in the event of overtopping of the downstream river to satisfy drinking temporary diversion works. Licence condi- water, irrigation, and environmental re- tions should assign responsibility for com- quirements. If the water quality is expected pensation to the downstream population for to be poor, then – as with the agreement on any damages that occur during such events. river diversion – alternative supplies of

The Report of the World Commission on Dams 291 Chapter 9

drinking water must be made available for ■ minimum technical releases to the the downstream population. downstream river (for navigation, water supply, downstream irrigation and so on); Test operation ■ maximum ramp rates for downstream Test operation of the spillway (if gated), of releases (to avoid problems with naviga- other outlet works, and of the turbines can tion and damage to the river banks); lead to major sudden releases to the down- ■ water allocations during normal operation; stream river, endangering human and ■ operation during normal and exceptional animal life. The operator will be responsible floods; for warning the downstream population that sudden releases may occur and may be ■ warning of people potentially affected dangerous. Local fishers will be compensated and rules for evacuation of people and for days when fishing is impossible. If test animals; operation takes place during the dry season, ■ opening of spillway gates; people experiencing damage to recession ■ periodic safety inspection by independ- agriculture must be compensated. ent parties; ■ drawdown procedure if dam safety is in Operation doubt; A range of agreements on the operating ■ phase should be covered in the licence: monitoring of relevant operation data and dissemination of data to stakeholders; and ■ environmental flow releases to the ■ downstream river; periodic review of operating rules.

13. Improving Reservoir Detailed technical guidelines are available on Operations ways to change reservoir operations either by adapting existing rule curves or introducing A range of project specific non-structural more modern computerised decision support and structural methods to adapt, modify, systems, including real-time data inputs, improve, or expand operations of dams and simulations and forecasting. In adapting associated facilities may be considered at reservoir operations owners/operators should: different periods in time. Structural meas- ■ work with stakeholders to collect views ures may include modernising equipment on current reservoir operations and views and control systems and improving civil on the need, concerns, and limitations of structures such as spillways, intakes and potential future changes in water release canals. Non-structural measures generally patterns, including downstream impacts; involve a change in reservoir operation ■ confirm any change in the priority of practices to optimise benefits, cater to water uses (such as environmental flows) changing water use priorities, enhance and evaluate the scope to use flow conjunctive operation, or improve sediment forecasting to optimise reservoir opera- management. Dam safety improvement and tion (Guideline 15); updating contingency plans for operation of ■ use simulation models where feasible, to reservoirs in extreme flood events are other assess the scope for optimising the supply aspects of adaptive management.10 of water and energy (for example timing,

292 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

quantity) into the system (for example Sediment management is one area where irrigation canal system and conjunctive use increased attention is needed. A sediment of groundwater, power grid or water management plan would consist of: distribution system) to improve the overall ■ monitoring sediment in the reservoir, value of the services in the system; including quantitative and qualitative ■ assess the ability to operate the reservoir analysis of sediment to verify properties to optimise delivery of services using and pollution levels; computerised models; ■ minimising sediment deposition in ■ assess the scope to further optimise reservoirs where possible by sluicing or interactive operation of the reservoir density current venting; with other reservoirs, diversions or facilities using basin-level decision ■ removing accumulated deposits where support systems; possible by drawdown flushing (drawing ■ provide clear responsibilities and proce- the water level down during high-flow dures for emergency warning and improved seasons), and excavation of sediments; preparedness of downstream countries, and operator training and downstream evacua- ■ catchment management programmes to tion in extreme flood events; and reduce sediment inflow to the reservoir ■ ensure monitoring systems are in place and where possible as part of a basin-wide feed into operational decision making. plan.

Sustaining Rivers and Livelihoods

14. Baseline Ecosystem river and its associated ecosystems. Baseline Surveys surveys should gather relevant information on: The effectiveness of mitigation, enhance- ■ the life cycle of important fish species ment, compensation and monitoring meas- (especially migratory species); ures require better baseline knowledge and understanding of ecosystems. Baseline ■ the distribution of habitat for threatened assessments inform both the national policy or endangered species; on maintaining rivers and requirements for ■ important areas for biodiversity; and environmental flows and other compensa- ■ key natural resources for riverine com- tion and mitigation measures. They are not munities. restricted simply to an ‘impact statement’, but instead gather the necessary baseline The studies should explicitly identify where information prior to alternatives being modifications to flow or water quality will assessed.11 have significant impacts on biodiversity, habitats, or riverine communities and The baseline surveys aim to establish the provide the scientific basis for testing flow link between the hydrological regime of the and quality scenarios against ecosystem

The Report of the World Commission on Dams 293 Chapter 9

responses (Guideline 15). Such studies baseline surveys, assisted where necessary by would normally be undertaken over several international networks. Enhanced local and seasonal cycles. regional capacity will help identify, understand and manage environmental impacts, Appropriate research agencies staffed with hence improving environmental outcomes specialised scientists should undertake for current and future dams.

15. Environmental Flow flow, water available for off-channel use, and the social and economic implications. Assessments Sophisticated habitat-modelling techniques Dams should provide for an environmental provide additional detailed information on flow release to meet specific downstream the flows required for specific valued river ecosystem and livelihood objectives identi- species or features, where the targeted rivers fied through scientific and participatory are of high conservation importance or have processes. In some cases managed floods may a high likelihood of conflicts over water. be necessary to maintain downstream floodplains and deltas. Several approaches EFRs are an integral part of the impact are available for assessing environmental assessment process (see Box 9.4.). Continual flow requirements (EFR), ranging from interaction with the design team is essential ‘instream flows’, which refer to within-bank to ensure that the least damaging and most flows, to ‘managed flood releases’ designed flexible options are retained and that the to overtop and supply floodplains and deltas. dam design reflects the structural and ‘Environmental flow’ includes all of these and operational needs of the flow release. stresses the need to meet clear downstream The Knowledge Base provides guidance on social and ecosystem objectives rather than the following steps to informed decision- simply releasing a quantity of water.12 making leading from baseline surveys to environmental flows13 : Environmental flow assessments (EFA) can be done at several levels of detail, from a Step 1: Situation assessment simple statement of water depth to provide Identify the extent of the targeted river wetted habitat for a particular fish species to system likely to be affected by a dam – a comprehensive description of a flow upstream, downstream, and in the reservoir regime with intra-annual and inter-annual basin – and alert decision-makers to the variability of low flows and floods in order likely ecological and social issues that will to maintain complex river ecosystems. need to be addressed. This draws on data in Confidence in the suitability of an EFA to the baseline surveys, where these exist. meet its objective is linked to the level of Step 2: Specialist surveys and investment in appropriate specialist inputs. identification of ecosystem components Holistic methodologies contribute to a A range of specialists (ecologist, geomor- detailed understanding of the merits and phologist, sociologist, and resource econo- drawbacks of a series of competing water mist) undertakes field surveys to provide a resource options in terms of required river comprehensive description of the affected

294 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

river. The studies link flow-related informa- ■ a definition of the priorities of the tion with ecological and social values. competing users, and a description of the resulting flow regime and its effect on Step 3: Developing predictive river condition; and capacity on biophysical responses to dam-related flow changes ■ river rehabilitation downstream of an The team develops data sets, models and existing dam, in which case the best that various analytical tools that can be used in can be achieved within the design scenario creation to assist decision-making limitations of the dam is described. (Step 5). These may include, for instance, In addition, the ‘no development’ scenario specifying conditions needed for a certain should always be included. fish species to spawn, or how water quality differs between the rising and falling arms of Step 6: Selection and a flood hydrograph, or how downstream implementation of one scenario fisheries and pastures will be affected. This requires: Step 4: Predicting social impacts of ■ reflection of the chosen scenario in the the biophysical responses dam design and the Environmental The present river use, exploitation of river- Management Plan; and related natural resources, and health profiles ■ monitoring of implementation to ensure of the affected people and their livestock are that objectives are met quantified, and possible flow-related health risks are identified. Box 9.4: Design and cost of environmental flows - Pollan Dam, Step 5: Creating scenarios Ireland

Scenarios are created that include social, The EIA results for the Pollan Dam showed that migratory salmon were biophysical, and economic parameters and present upstream of the dam site, and that the dam would act as a present a series of future options for deci- barrier to salmon movements, affecting the fishery. The environmental water releases were designed to meet the seasonal needs of the sion-makers. Scenarios may be defined by: migratory fish. Design modifications had major implications for ■ the volume of water required as yield from structures such as the concrete dam, spillway, and downstream channel. The capital cost of all environmental protection measures is the dam – the rest is allocated to the river; estimated to have increased the total cost of the project by 30% (from ■ protection of a valued species, $6 million to $8 million). The flows have been effective in maintaining the salmon population and the recreational fishery. community, or river feature, in which Source: Smith, 1996; Bridle, pers. comm. 2000 case a flow regime to achieve this would be described;

16. Maintaining Productive Fish passes should be tested and shown to be efficient mitigation tools. Fish pass design Fisheries has focused on the needs of leaping salmo- The impact of dams on fish and fisheries is nids that usually dominate fish communities of major concern in many parts of the world. in fast-flowing rivers in the industrialised Several issues need attention in order to northern countries. Yet many fish species in maintain productive fisheries (see also slow-flowing tropical rivers are unable to use Guideline 15).14 this kind of fish pass as they do not leap.

The Report of the World Commission on Dams 295 Chapter 9

Box 9.5: Benefits of improving fish passes project proponent, the agencies responsible for developing or managing the reservoir In 1976 a pool-and-weir type fish pass was incorporated into a tidal fishery, and the fishers, with priority given barrage on the Burnett River in SE Queensland, Australia. Assessment of the fish pass in 1984 and 1994 showed it to be ineffective, with only to affected people. Fisheries management 2 000 fish of 18 species ascending over a 32 month period. The fish objectives for dams include: pass was modified to a vertical-slot design with low water velocity and turbulence. Over 17 months 52 000 fish of 34 species used the rede- ■ preventing the loss of endangered and/or signed fish pass. Non-leaping fish are now able to use the fish pass, commercially important fish biodiversity; benefiting the entire fishing community on the river. ■ maintaining fish stock abundance; Source: Flanders, 1999; env219, WCD Submission ■ ensuring the long-term sustainability of the catch, employment and income; and Where fish passes exist, their effectiveness ■ producing fish for local consumption and should be measured and their design im- exportable fish products. proved where efficiency is low. For new dams, proposed fish pass designs should be Reservoir fisheries management concerns tested hydraulically and their appropriate- focus on protecting spawning grounds in ness for the target species explicitly assessed. affluent inflow areas, stocking to increase (See Box 9.5.) production (for example, of a small pelagic fishery) and advice on management of the Where the reservoir fishery will be assessed water level to reduce impacts that harm fish as a project benefit, the proposal should stocks. Downstream river fisheries manage- explicitly include regional experience of ment focuses on aeration of anoxic dis- similar reservoir fisheries, rigorous assess- charge water from the dam, provision of ments of potential reservoir productivity, effective fish passes, reduction of turbulence and proposals for the institutional mecha- in the stilling pool, and mitigation of fish nisms to manage the new fishery. Relevant losses on the downstream floodplain contracts should be established between the through flow releases.

Recognising Entitlements and Sharing Benefits

17. Baseline Social ■ a low-intensity appraisal during options Conditions assessment, linked to Strategic Impact Assessment; and Constructing a social baseline is central to ■ a more comprehensive baseline during the planning and implementation process. It Project-level Impact Assessment once an provides key milestones against which intervention emerges from the options project performance and positive and assessment process. negative impacts on people can be assessed through periodic monitoring and evalua- In light of the significance of the impacts tion. It is also a key input to strategic social that can occur between the time a decision impact assessment (Guideline 4).15 is made to develop a project and its actual Social baseline assessments should be done implementation, the second baseline study at two stages in the planning process: may need to be updated at the tender stage

296 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

of a project. Subsequent monitoring activi- methods such as key informants, oral ties can follow standard practice. testimonies and preference assessment, direct observations and literature review. For Baseline studies should be undertaken for all the assessment of social processes, some of impact areas, and in particular the areas and the participatory appraisal methods (such as communities likely to be positively and the Venn diagram of institutional processes) negatively affected by the project. Groups to can be useful. Aerial photos, satellite be considered include communities to be imagery, and geographic information systems displaced, prospective host communities, can be combined with participatory commu- downstream and upstream communities, and nity resource mapping exercises. command area inhabitants (irrigation schemes, transmission line corridors, and The State is responsible for ensuring that other infrastructure areas). The social social baseline information is collected. The baseline study should be participatory and task should be carried out by independent involve discussion and feedback through institutions selected in consultation with stakeholders (Guideline 1). It should also be the stakeholder forum. Large projects should accessible to the general public. be considered as an opportunity for building local capacity (in relevant government Some of the common techniques used to assess baseline social conditions combine agencies, academic and research institu- gender-sensitive household surveys, commu- tions, and civil society organisations) to nity-level participatory appraisals, and other undertake social assessment and monitoring.

18. Impoverishment Risk access to common property resources; and loss Analysis of socio-cultural resilience through a community’s inability to secure its interests.16 The impoverishment risks and reconstruction analysis model for resettling affected The internal logic of the model suggests and displaced populations adds substantially that: to the tools used for explaining, diagnosing, ■ preventing or overcoming the pattern of predicting, and planning for development. impoverishment requires risk reversal; This guideline should be read in conjunc- ■ tion with Guideline 4: Strategic Impact explicit identification of risks in advance Assessment, Guideline 5: Project-related is crucial for planning counter-risk Impact Assessment and Guideline 17: Base- measures; and line Social Conditions. At the core of the ■ the transparent recognition of risks in model are three fundamental concepts: risks, advance will allow planners and affected impoverishment and reconstruction. Impov- people to search for alternatives to avoid erishment risks are analysed by separating displacement or to respond with mitiga- out the components of the displacement tion and development measures or process. They are landlessness; joblessness; strategies and coping approaches. homelessness; social, economic, and political marginalisation; food insecurity; in- The strategy to implement the impoverish- creased morbidity and mortality; loss of ment risk model includes the following:

The Report of the World Commission on Dams 297 Chapter 9

■ the baseline study covering such aspects successfully overcome impoverishment as numbers of affected people, availabili- risks. ty and access to resources, sources of livelihood and social, cultural, demo- A four stage, two-generation model that graphic, economic and political condi- would enable affected communities to reach 17 tions and processes (Guideline 17) – full development includes : these studies must incorporate variables ■ developing benefit sharing, mitigation, to construct key elements of the risk and development plans with the partici- model, in addition to collecting data on pation of the affected people; other aspects; ■ enabling resettled people to cope and ■ the baseline study providing information adapt following displacement, with to understand how social, economic and continued support from the government cultural networks, physical environment and civil society groups; and resources support the well-being of ■ supporting economic development and individuals, households and communi- community building within resettlement ties; and areas; and ■ mitigation, development and benefit- ■ hand over of resettlement sites and sharing measures to improve the liveli- incorporation within broader social and hoods and well-being of affected people, political institutions at a stage when and to provide the social and physical resettlement and development plans are environment that would enable individ- fully realised and capable of sustaining uals, households and communities to the gains for future generations.

19. Implementation of the A master contract ensures that the MRDAP Mitigation, provisions and responsibilities are clearly understood and assigned, while a perform- Resettlement and ance contract formalises provisions and Development Action commitments with affected families and Plan communities. These two legally binding contracts are found at the government and A mitigation, resettlement, and develop- developer level and the affected persons and ment action plan (MRDAP) is negotiated community level. Where government between all affected peoples, the govern- ministries or departments act as the devel- ment and the developer. It generally has two oper, there may not be a need for a master elements – a master contract and a perform- contract, but their obligations in this regard 18 ance contract. The affected people were should be clearly stated in the MRDAP. In identified through an Impoverishment Risk such cases, they will enter into agreements Analysis (Guideline 18). One component of directly with affected people through the MRDAP may be a Project Benefit- performance contracts. Sharing Mechanism (Guideline 20). The overall obligations and responsibilities of At the government and developer the government and the developer will be level included in the Compliance Plan (Guide- The MRDAP should have legal status. line 21). Countries with resettlement and rehabilita-

298 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

tion acts or policies should make suitable ■ provides for continuous monitoring of amendments to ensure that contracts apply implementation by an independent field to all affected communities including monitoring team, selected with the downstream communities. consent of the affected people and ■ The master contract is concluded be- reporting to the multi-stakeholder tween the developer (public corpora- committee; tions, private, or joint) and the govern- ■ confirms the composition and role of a ment. In the contract, the developer panel of experts for the implementation agrees to carry out all the actions set out phase (Guideline 22), appointed by and in the MRDAP in a timely manner. It reporting to the multi-stakeholder specifies government responsibilities for committee to assess whether the MR- providing support to acquire land, staff, DAP is being implemented correctly, schools and so on. rehabilitation objectives are being ■ A private sector developer should sign a achieved and project benefits are being performance bond supported by financial provided to affected people; and security (Guideline 23). ■ establishes a mechanism for dispute ■ Where the government undertakes to resolution – the multi-stakeholder provide other services (including land committee is responsible for hearing acquisition, road building and health disputes and grievances related to care), the responsible line ministry execution of the performance contract enters into agreements with other where the mitigation and development appropriate ministries to provide them. office has been unable to resolve issues. If the committee cannot resolve the The master contract: dispute or grievance, the matter will be referred to the appropriate judicial body. ■ specifies penalties, incentives, remedies, and other measures to facilitate compli- At the community and affected ance by the government and the devel- persons level oper; Based on the provisions of the master con- ■ provides for the establishment of a tract, performance contracts are agreed with mitigation and development office for the community and affected persons detailing: implementation purposes. This is usually ■ compensation, resettlement, and devel- staffed by government officials drawn opment entitlements; from various ministries supported by staff from the developer; ■ schedule and method of delivery; ■ confirms the role of a multi-stakeholder ■ institutional arrangements to deliver the committee as a subgroup designated by the commitments; stakeholder forum (including senior ■ obligations and responsibilities of the government officials, the developer, NGOs parties in the contract, namely affected and affected peoples groups) to deal with peoples, community, government and grievances and supervise the work of the developer; and mitigation and development office; ■ recourse procedures. ■ empowers the mitigation and development office to monitor the implementa- The master and performance contracts have tion of the MRDAP; to be agreed at the project feasibility stage

The Report of the World Commission on Dams 299 Chapter 9

and signed prior to tendering the construc- project implementation. The multi-stake- tion contract. The signing of the perform- holder committee addresses all disputes ance contracts by the affected persons and related to performance contracts. communities signals their consent for

20. Project Benefit-Sharing pumped irrigation from the reservoir, and Mechanisms benefits from managed flows and floods.

Adversely affected people are entitled to Community Services-Related: Provision of share in project benefits. Beneficiaries and better and higher levels of service including benefits need to be identified and will form health, education, roads and public trans- part of the Mitigation, Resettlement and port, and drainage; income support for Development Action Plan (see Guideline vulnerable or needy households; agricultural 19). The nature of agreed benefits can take support services including preferential many different forms. planting materials and other inputs; community forests and grazing areas; market and Type of project benefits meeting spaces. Project Revenues-Related: A percentage share of project revenues/royalties, the construc- Household-Related: Skills training and tion budget and other profits. A joint interim family support; interest-free loans enterprise with affected people having a for economic activities, housing improve- share of equity. ments, provision of start-up livestock, access to public works or work for wages, free or Project Benefit-Related: Provision of irrigated subsidised labour-saving devices or productive land or an opportunity to purchase irrigated machinery, access to preferential electricity land, access to irrigation water, provision of rates, tax rates, water and service charges. electricity supply, domestic water supply from the project as appropriate. Right to Identification, assessment, and reservoir fisheries, cultivation in the draw- delivery of benefits down area of the reservoir, and contract to Definition of Beneficiaries: Beneficiaries manage recreational/water transport facilities. include all people in the reservoir, upstream, downstream, and in catchment areas whose Project Construction and Operation-Related: properties, livelihoods, and non-material Employment in construction, plant opera- resources are affected; and also those affect- tion, and service sector of the project. ed by dam-related infrastructure such as Financial and training support for self- canals, transmission lines, resettlement, and employment contract to provide goods and other factors. services. Identification of Beneficiaries: Baseline surveys Resource-Related: Preferential access to, or must establish the nature and extent of loss custodianship of, catchment resources for to livelihoods and enumerate all categories defined exploitation and management of adversely affected and displaced individuals, purposes, catchment development such as families, and communities. This will be done planting fruit trees or reforestation, access to with the participation of the affected people

300 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

and reflect a rights-and-risks approach (Guide- Benefit Delivery and Redress Mechanisms: The line 17). mitigation and development office is responsible for the delivery of benefits to the affected Eligibility and Level of Benefits: All adversely people (Guideline 19). The multi-stakeholder affected people are entitled to benefits. The committee will hear all representations level of benefits must be assessed, agreed upon relating to identification of beneficiaries, by the parties involved (affected people, apportionment of benefits, performance government, and developer/financier) and contracts, and delivery of benefits. included in the performance contract.

Ensuring Compliance

21. Compliance Plans ■ The use of voluntary measures. These will include tools such as comprehensive The preparation of an overarching Compli- compliance criteria and guidelines, ISO ance Plan by the developer will address certification, integrity pacts (see Guide- technical, financial, social and environmen- line 25) and the independent review of tal obligations and commitments and internal processes and commitments. provide the means for the developer to ■ The level of in-country institutional capaci- describe clearly how compliance will be ty. Where it is insufficient to meet the ensured for a particular project. The stake- requirements of the plan, provision must holder forum will be able to monitor com- be made for training and other technical pliance against the plan, which will be a assistance, as required, to ensure suffi- publicly available document.19 cient capacity is put in place. States are at different stages in the develop- ■ The use of performance bonds, supported by ment of regulatory systems and institutional financial guarantees and trust funds. The capacity. The range of tools selected to use of one or both of these measures will ensure compliance for any particular project be needed to ensure sufficient funds have would vary from case to case. While varia- been set aside to secure performance. tions in systems and capacity will result in They will need to be developed and project-specific Compliance Plans, the level applied in a manner that best suits the of compliance should be consistent. particular circumstances (see Guideline 23: Performance Bonds and Guideline 24: In using Compliance Plans in connection Trust Funds). with the construction of dams, a number of ■ The cost of compliance. The cost of issues will need to be addressed on a case-by- compliance will need to be built into the case basis, including the following: plan, the project budget and the evalua- ■ The laws applicable to the construction of tion process. dams. These will vary from country to ■ Performance indicators and benchmarks country, and the Compliance Plan will need to be established against which need to be consistent with local laws. compliance can be assessed.

The Report of the World Commission on Dams 301 Chapter 9

22. Independent Review one social scientist) to ensure continuity Panels for Social and through the different project phases. Plan- ning and appraisal may call for different Environmental Matters skills and composition to those needed for Independent review panels (IRP) should be monitoring implementation of an environ- established for all dam projects. They differ mental management plan or the resettle- from tribunals, commissions, judicial reviews ment and development programme. or other recourse mechanisms as their principal task is reviewing assessment of In establishing an independent panel, States impacts and the planning, design and and financing agencies should consider the implementation of social and environmental following: mitigation plans. In some countries their ■ Project level IRPs should be established recommendations can be binding on all by the State (as developer or regulator or parties. In others they are only advisory. The the Ministry of Environment), in agree- scope of the IRP powers is laid out in its ment with the stakeholder forum, as terms of reference. They report to the soon as the options assessment has regulator, developer, consultants, affected decided on a dam as a possible option, peoples and financing agency to help ensure and prior to project-level impact assess- the best possible social and environmental ment beginning. outcomes. The IRP is not a dispute resolu- ■ IRPs are funded by the State, the devel- tion mechanism, but may assist in bringing oper or a financing agency according to issues to the attention of the relevant body local circumstances. The IRP forms an for resolution. integral part of project costs. ■ The primary reporting responsibility of IRPs offer independent assessments of the IRPs should be to the national govern- issues that should be dealt with in project ment involved and more specifically the level impact assessments and project imple- responsible project agency and regulator. mentation, while also providing a mecha- The IRP should include members able to nism to transfer best practice from one effectively address the major issues that are project to another, both nationally and addressed by state-of-the-art ecosystem, internationally. IRPs further provide a demographic, social and health assess- quality control function to assure the ments. They have the prerogative to add developer, regulator, financing agency and additional members to deal with issues for affected groups that the necessary standards which the IRP has insufficient expertise. are being met and that laws or guidelines are complied with, as laid out in the Compli- Panels should include at least one host ance Plan. They usually perform functions country national and at least one member in the social and environmental domain supported by any affected people. The IRP is similar to independent engineering inspec- independent of all parties and its terms of tors for technical issues. reference should allow the panel to look into any issues deemed important without The composition and tasks of IRPs can be the need to justify such examination. adapted to different stages of the project cycle, although it will be useful to maintain ■ The developer ensures systematic infor- core members (normally one ecologist and mation distribution to the IRP, which

302 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

has access to all project-related docu- ■ The developer has the obligation to mentation. show how they are responding, or intend ■ All reports following panel missions must to respond, to the issues raised by the become public documents once the IRP. developer or appropriate agency has had ■ Frequency of IRP visits to the project a reasonable time to comment (usually area should be flexible. In some phases one month). Should the developer or one per year may be sufficient, though other agencies request assistance on a once construction starts, six monthly confidential basis then that is an issue for intervals would be more appropriate. them and the IRP to negotiate.

23. Performance Bonds mitigation works, monitoring, auditing and decommissioning, or to aspects of Performance bonds supported by financial each of these activities. Bonds should be guarantees provide a secure way of ensuring carefully targeted to activities identified compliance with commitments and obliga- in an approved management plan and, tions (see Box 9.6). They are used by mining preferably, should apply to the developer and environment protection agencies and in who is ultimately responsible for the the construction industry in many different entire project. The developer may in countries. The bond is called upon, either in turn enter into performance bonds with whole or in part, to meet unfulfilled obliga- contractors. tions and commitments or is released when ■ commitments are met, either in whole or in The form of security, including insurance part, depending upon the circumstances. cover, to be provided. A package of measures can be used that collectively Performance bonds have been used widely results in providing sufficient financial in the construction industry to ensure that assurance. The use of bank guarantees is work is completed within the specified time a cost-efficient method of providing period and to specified standards (including financial assurance, but there are many during the construction of dams). They are others, including insurance cover. also used in relation to activities that carry a ■ Who will hold the bond and hence determine high risk to the environment; for example whether to release or use the security. The ensuring that mine sites will be rehabilitated. relevant government agency (the environmental protection agency or mines In applying the use of performance bonds to department) has been used in most the social and environmental mitigation countries where performance bonds have measures related to the construction of been used to date. However, a well- dams, a number of issues will need to be structured trust fund can also be used, addressed on a case-by-case basis, including particularly where the government is also the following: the developer (Guideline 24). ■ The activities the bond will apply to. The ■ The appropriate level of financial assurance. bond could apply to a wide range of Considerations include the higher cost activities, such as physical resettlement to government to do the works, a contin- and provision of benefits, environmental gency sum for high-risk activities,

The Report of the World Commission on Dams 303 Chapter 9

Box 9.6 Financial assurances and the Environment Protection making provision for staged assurance, Agency, Victoria, Australia and providing a discount for quality management, good past practice, and/or The use of performance bonds supported by financial assurances has now been successfully applied in a number of different areas. For a lower risk activity. example, following a chemical waste management company becoming ■ The stages of the development when the bankrupt and leaving the government with a potential liability to dispose of abandoned waste, the Victorian Environment Protection security will be released. Partial release Agency (EPA) required financial assurances from 38 companies provides a form of financial incentive to involved in the waste industry. The level of the financial assurance was the developer to discharge its responsi- arrived at based upon the extent of the environmental risk – which has since been reduced where improved environmental management bilities. systems have been put into place. While the EPA has not called on any ■ Regular review of the level of security to financial assurances to date, the programme has been successful in raising the performance of the industry overall and has protected the reflect the actual costs. A fall-back provi- government from financial risk. The programme is being extended to sion is needed that allows the State to cover landfills and major petroleum product storage sites. extract the difference from the developer Source: Robinson, pers. comm. 2000 where the level of security proves to be insufficient.

24. Trust Funds the project. This could include providing an effective means for the collection and Trust funds have been used over a long distribution of royalties from dam-related period of time, and in a wide variety of activities to fund ongoing initiatives. situations, to ensure that funds set aside for a particular purpose are used for that pur- The use of trust funds would be most appro- pose (see Box 9.7). In recent years they have priate where the proponent is the State. In been applied to the establishment and such cases, the concern is no longer one of ongoing management of government- ensuring that the risk is not passed from the protected areas, through initiatives funded developer to the government, rather it is by the Global Environment Facility ensuring that the risk is not passed from the amongst others. There is scope to extend State to the affected communities and to these funds to other areas where there is a the environment. need to set aside monies to be applied for a particular purpose, such as benefit sharing In using trust funds in connection with the and mitigation measures associated with the construction of dams, a number of issues will construction of dams. They could also be need to be addressed on a case-by-case basis, used for decentralising responsibility to including the following: affected communities for planning and ■ The laws applicable to the establishment of implementing their own mitigation, devel- trust funds. These will vary from country opment and resettlement programmes. to country, and the trust deed will need to be consistent with local laws. Trust funds could be effectively used, either alone or in conjunction with bonds, to ■ The content of the trust deed. This will secure the financing of ongoing obligations need to include an open and transparent in relation to monitoring and auditing – process for appointing trustees and activities that must continue for the life of administering the fund and for setting

304 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

out the activities the funds will be Box 9.7 Suriname Central Nature Reserve applied to, such as resettlement, environ- In 1998, the Government of Suriname announced the creation of the mental mitigation, monitoring and Central Suriname Nature Reserve, a protected area that covers 10% of auditing. The deed must be publicly the country. The ongoing maintenance of this protected area is available. secured through the Suriname Conservation Foundation Trust Fund announced in April 2000, which will ultimately administer a $15-million ■ The trustees of the fund. The trustees will endowment. This sum will be invested in the fund through contribu- need to be sufficiently independent from tions from numerous donors. The fund will be used for long-term management support, ecological surveys, conservation awareness and the developer and have the confidence education, and ecotourism as a conservation enterprise. The fund’s of stakeholders. trustees are drawn from government, sponsoring institutions, the private sector, and indigenous peoples. ■ The role of affected people. Their role in Source: Famalore, pers. comm. 2000 managing trust funds in relation to mitigation, resettlement and development needs to be defined.

25. Integrity Pacts ■ The form and content of the pact. The form and content of the pact must Integrity pacts relate to the procurement comply with accepted international process, namely the supply of goods and models and past applications. services. (See Box 9.8 and Chapter 8 ■ The level of in-country institutional capaci- Strategic Priority 6: Ensuring Compliance) ty. Where this is insufficient to meet the They are voluntary undertakings aimed at requirements of the integrity pact, reducing corruption and founded on con- provision must be made for training and tractual rights and obligations. They can be other technical assistance, as required, to used as one component of a Compliance ensure sufficient capacity is put into Plan. Integrity pacts are of particular use in place. This should be covered in the situations where regulatory systems and Compliance Plan (Guideline 21). institutional capacity are weak, but they have universal application.20

Box 9.8 Mendoza Province, Argentina Integrity pacts in various forms have now been tried and tested in many countries. The Provincial Governor of Mendoza Province, Argentina decided in 1997 to amend procurement rules to include an Integrity Agreement between the Government of the Province and companies interested in In applying the use of integrity pacts to the bidding for government contracts. Government commitments under construction of dams, a number of issues will this agreement include providing full transparency in relationships with suppliers, ensuring that employees will not accept or demand any need to be addressed on a case-by-case basis, bribes, informing the State Prosecutor of any violations, requiring a bid including the following: bond, excluding violators from future contracts, and having the State Prosecutor oversee implementation of the policy. Source: Wiehen, 1999

The Report of the World Commission on Dams 305 Chapter 9

Sharing Rivers for Peace, Development, and Security

26. Procedures for Shared should receive adequate technical informa- Rivers tion about the proposed project and the results of any impact assessments, and A basin-wide perspective is promoted for should respond in writing within six open discussion of the issues, negotiation on months of the notification with their sharing the benefits, and the mitigation of findings and response to the proposed any adverse impacts. The procedures for project; and equitable and reasonable utilisation, no ■ as required to cover any additional data significant harm, prior notification, impact and information that is available and assessment, and dispute resolution will build necessary for an accurate evaluation by on provisions of the UN Convention on the any potentially affected riparian States. Law of the Non-Navigational Uses of International Watercourses and other In the event that properly notified riparian international agreements. Such provisions States do not respond in a reasonable and are also relevant to rivers within a country timely manner, the notifying State would shared between a number of sub-national proceed with planning and development, entities. subject to its observance of the relevant Prior notification international law principles and the Commis- sion’s strategic priorities and policy principles. States considering options that may have a significant impact on other riparian States In the event that a State fails to notify should notify those States at various stages another riparian State which could poten- and establish an effective channel of com- tially suffer significant harm by the proposed munication between all potentially affected action, the potentially affected State should parties. Notification should occur: be able to request and receive information, ■ at an early stage of planning, as part of make their views known, including propos- the strategic impact assessment, and ing modifications, and be part of a negotiat- should allow potentially affected riparian ed settlement before any action to construct States at least three months to identify the dam is taken. If this opportunity is relevant issues for inclusion in subse- denied, remedies should be available through quent preparatory studies and impact the International Court of Justice (ICJ), or assessments; other appropriate mechanisms. Any external ■ during the scoping stage of impact assess- financing for the dam should be conditional ments, to allow agreement on mechanisms upon resolution of the issue as described in for sharing technical data and information, Chapter 8 (see policy principle 7.5). and for participation in project-related impact assessments – potentially affected Basin-wide impact assessment riparian States should respond within three The Commission’s Guidelines on Strategic months of the notification; Impact Assessment and Project-Level Impact ■ prior to selecting an option on a shared Assessment, taken together with prevailing river as part of a preferred development regulatory requirements, provide the frame- plan – potentially affected riparian states work for a basin-wide assessment of impacts.

306 Dams and Development: A New Framework for Decision-Making Guidelines for Good Practice

Impact assessments designed to suit context- Dispute resolution specific situations need to: In the event a dispute cannot be resolved ■ include a participatory basin-wide within six months, either through good faith scoping phase; negotiations or independent dispute resolu- ■ take into consideration the submissions tion, it should be referred to a fact-finding of riparian states and affected communi- commission as detailed in Article 33 of the ties; and UN Convention on the Law of the Non- ■ be subjected to review by an independ- Navigational Uses of International Water- ent panel agreed upon by all potentially courses. Failing resolution through this affected riparian states. body, the dispute should be heard by the ICJ either through case specific agreement or All states should give the independent panel through compulsory jurisdiction in Article access to all necessary information. 36 of its statute.

Endnotes 10 WCD Thematic Review IV.5 Operations. 11 Brown and King, 1999; Brown et al, 1999, 1 WCD Thematic Review I.2 Indigenous People. Contributing paper for WCD Thematic 2 WCD Thematic Review V.2 Environmental Review II.1 Ecosystems. and Social Assessment; WHO, 1999, WCD 12 WCD Thematic Review II.1 Ecosystems. Working Paper on Human Health; Brandt and Hassan, 2000, WCD Working Paper on 13 Brown and King, op cit; Brown et al, 1999, Cultural Heritage Management. Contributing paper for WCD Thematic 3 WCD Thematic Review V.2 Environmental Review II.1 Ecosystems. and Social Assessment; WHO, op cit; Brandt 14 Bernacsek, 2000, Contributing paper for and Hassan, op cit. WCD Thematic Review II.1 Ecosystems. 4 WCD Thematic Review V.1 Planning. 15 WCD Thematic Review V.2 Environmental 5 Ibid. and Social Assessment. 6 WCD Thematic Review II.2 Global Change. 16 WCD Thematic Review I.3 Displacement; 7 WCD Thematic Review III.1 Economic Cernea, 2000. Analysis, Chapter 9. 17 Scudder, 1997c. 8 WCD Thematic Review III.1 Economic 18 WCD Thematic Review I.3 Displacement. Analysis, Chapter 4. 19 WCD Thematic Review V.4 Regulation. 9 WCD Thematic Review III.1 Economic Analysis, Chapters 2, 6, 7, 8. 20 Ibid; Wiehen, 1999.

The Report of the World Commission on Dams 307 Beyond the Commission – An Agenda for Change

Chapter 10 Beyond the Commission – An Agenda for Change

ur report has distilled more Othan two years of intense study, dialogue, and reflection by the World Commission on Dams (WCD), the WCD Secretariat, the

WCD Forum, and hundreds of individual experts on every facet of the dams debate. It contains all the

significant findings that result from this work and expresses everything that we, the Commission, feel is

important to communicate to governments, the private sector, civil society actors, and affected peoples –

in short, to the entire spectrum of participants in the dams debate.

The Report of the World Commission on Dams 309 Chapter 10

The Commission alone is ties downstream, by taxpayers and by the The debate about dams ultimately responsible for the natural environment. began well before the conclusions and recommenda- Commission, and it will ■ Lack of equity in the distribution of tions presented here. continue well beyond it. benefits has called into question the We hope that one of the value of many dams in meeting water Dams and Development: A New lasting results of the and energy development needs when Framework for Decision-Making WCD process will have compared with the alternatives. provides a solid basis for been to change the tenor ■ By bringing to the table all those whose assessing options for energy of that debate from one rights are involved, and who bear the and water development, and of lack of trust and risks associated with different options for for planning and implement- destructive confrontation water and energy resources development, ing projects that can achieve to co-operation, shared the conditions for a positive resolution of the desired benefits without goals and more equitable conflicts and competing interests are exacting an unacceptable cost development outcomes. created. for anyone affected, or for our ■ Negotiating outcomes will greatly environment. If all parties improve the development effectiveness now adopt, adapt, and implement our of water and energy projects by eliminat- recommendations in good faith, much of the ing unfavourable projects at an early energy currently focused on the controversy stage, and by offering as a choice only surrounding large dams can be channelled those options that key stakeholders agree into improving development outcomes on a represent the best ones to meet the needs co-operative and sustainable basis. But this in question. will not happen unless the wider dams constituencies – those who entrusted us The directions are clear. It is one thing, with the mandate two years ago and who however, to see this. It is another to actively created the Commission as a platform for break through traditional boundaries of dialogue – become the heirs of our work, go thinking, step into a different frame of mind forth with it, and multiply its impact. and look at familiar issues from another The evidence we present is compelling. We perspective. This is what the Commission feel confident that the WCD Knowledge has had to do, and we have shown over two Base provides overwhelming support for the years that it works. We have seen similar main messages in the report. We believe constructive processes at work among many there can no longer be any justifiable doubt of the constituencies that have participated about the following: so actively in our work as they have come to understand what motivates other interest ■ Dams have made an important and groups. significant contribution to human development, and the benefits derived It is time now to bring the debate home. from them have been considerable. The controversy over dams has appropriate- ■ In too many cases an unacceptable, and ly been raised to the international stage. A often unnecessary and high price has dissipation of that controversy, however, been paid to secure those benefits, should allow decisions about fundamental especially in social and environmental water and energy development choices to be terms, by people displaced, by communi- made at the most appropriate level. This

310 Dams and Development: A New Framework for Decision-Making Beyond the Commission – An Agenda for Change

level is one where the voices of powerful We urge all groups to study this report and international players and interests do not discuss how to adopt or adapt its recommen- drown the many voices of those with a dations, bearing in mind that it results from direct stake in the decisions to be taken. For consultations that, in terms of inclusiveness this to work, all the actors have to make and breadth of scope, are beyond the reach that commitment to step out of their of any individual interest group. In this way familiar frame of reference. We recommend the report will serve as a common platform that all parties begin by using the report as for all parties to develop forms of implementa- the starting point for discussions, debates, tion appropriate to their context and status. internal reviews, and reassessments of existing procedures, and for an assessment of This section provides an illustrative list of how they can address a changed reality. short- and medium-term actions specific to individual stakeholder groups. The list is Strategic Entry Points for addressed to all stakeholders, including Follow-up governments, the private sector, bilateral and multilater- Nobody can, of course, simply pick up the al funding agencies, profes- We urge all groups to report and implement it in full. It is not a sional associations, and civil study this report and blueprint. This section proposes a number of society. By applying it they can discuss how to adopt or entry points to help organisations identify immediately begin incorporat- adapt its immediate actions they might take in ing the content and spirit of recommendations, response to the Commission’s report. the report in their own profes- bearing in mind that it Engaging through these entry points would sional practices and help results from consultations initiate permanent changes to advance the maintain the momentum for that, in terms of principles, criteria, and guidelines in the change generated by the work inclusiveness and breadth report. of the Commission. All groups of scope, are beyond the should consider the following: reach of any individual These entry points don’t aim to be compre- interest group. hensive. Instead, they illustrate the sorts of ■ Active dissemination of actions different constituencies can take the report that would, collectively, bring about a Water and energy develop- permanent shift in the debate over our ment interacts with many sectors and water and energy future. disciplines and involves a variety of actors. Assist, where you can, in translating the The Commission calls on all interested report into different languages, teaching parties to recognise that its recommenda- materials, checklists, and other tools that tions are in the enlightened self-interest of will help shift gears from prescription to all concerned, and form a solid base for good implementation. Help ensure that the report faith negotiations around water and energy reaches as many of those concerned about sector planning based on accepted norms the dams debate as possible. NGO and and on the evidence in our Knowledge Base. professional networks especially can help We are aware that many organisations ensure that people affected by dams or those involved with dams, water and energy will planning and managing dams worldwide have to review existing criteria and guide- have access to the Commission’s report and lines and adapt them in light of our report. recommendations in appropriate languages.

The Report of the World Commission on Dams 311 Chapter 10

We call on local media and publications of ■ Public pledges professional bodies to write and comment on Following such reviews, organisations and the report and on its proposals for the future. governments are urged to issue a public statement of support for the direction that ■ Review the report the report takes. Beyond that general The Commission recommends that all inter- endorsement, make such pledges as specific ested groups put in place an appropriate as possible – for example, by endorsing the consultative process to review the report and seven strategic priorities set out in propose a fitting response that will lead to Chapter 8. Governments and other organi- effective implementation and incorporation of sations are further invited to report on what its recommendations in their respective actions they have taken as a result of such policies. This may take place at the national reviews, and how their policies and actions level by government have changed. These reports can also be sent initiative, at the regional to the WCD website ([email protected]) to Following such reviews, level, or at the level of facilitate sharing of lessons and information. organisations and individual organisations. In governments are urged to some cases, such as the ■ Evaluation, monitoring, learning issue a public statement of harmonisation of the social The Commission has noted with dismay the support for the direction that and environmental provi- absence of formal processes for evaluating the report takes. sions of the Export Credit the long-term performance and outcomes of Guarantee Agencies or large dams projects worldwide, despite the professional associations, extensive interna- billions of dollars spent on them. This tional consultation will be essential. remains a huge gap in the process of learning from past good and bad practices, and Here are some questions that need answers. has severely constrained the capacity of the sector to learn rapidly from experience, both ■ What are the reactions to the Commis- nationally and internationally, and to sion’s key proposals? promote adaptive management. ■ How will these affect ongoing activities, immediately and in the long term? The Commission therefore urges all parties ■ Which recommendations can be adopted? – national governments, aid agencies, Why not all of the recommendations? financiers, professional associations, and ■ What practices can be changed in private-sector consultants – to invest more response to the report? resources in evaluating past performance through open, participatory processes that ■ What national or international networks build on the Commission’s case study or organisations might help implement methodology and adapt it to national the recommendations? circumstances. Evaluation should normally be undertaken every five to ten years as The Commission welcomes a response from some impacts only emerge over time. all parties concerning the proposals arising from the review processes. These will be ■ Review dams currently under posted on the WCD website (send to development [email protected]) and on any websites that The Commission recommends that public follow uptake and implementation of the and private developers, financiers, and Commission’s report. consultants involved in dams projects in all

312 Dams and Development: A New Framework for Decision-Making Beyond the Commission – An Agenda for Change

stages of the planning and development with many partners. They complement what process give high priority to reviewing these lies in the body of the report. proposals against the framework laid out in Chapter 9, and to adjusting procedures and National governments adapting projects where necessary. ■ Establish an independent, multi-stake- ■ Promote capacity building holder committee to address the unre- Adopting the recommendations of the solved legacy of past dams. Commission has implications for institu- ■ Require a review of existing procedures tional capacity and financing to manage the and regulations concerning large dam transition in water and energy management projects. that the Commission is recommending. ■ Develop a specific policy statement Lack of capacity should not be an argument governing stakeholder participation in for not adopting the Commission’s report. options assessment and planning, setting Capacity must be built if good outcomes are out the range of considerations that will to be achieved, including strengthening be incorporated. civil society and particularly empowering women to make their voices heard. Bilateral ■ Review legal, policy and and multilateral donors, national and institutional frameworks to Capacity must be built if international NGOs, are urged to support assess and remove any bias good outcomes are to be this transition in developing countries and against resource conserva- achieved, including wherever possible to offer support to inter- tion, efficiency and decen- strengthening civil society national networks assisting in this process. tralised options, and any and particularly Investing in the capacity and process for hindrance to open partici- empowering women to options assessment and decision-making patory processes. make their voices heard. should be seen as an investment in a long- ■ Introduce and support a term strategy of lowering the costs of future UN General Assembly projects. resolution that welcomes the publication of the Commission’s report, invites Taking the Initiative – governments to accept and implement Institutional Responses its recommendations and transmits the report to the Rio+10 process as a positive This section contains recommendations example of multi-stakeholder co-opera- addressed at specific stakeholders in the tion that can result in a substantial dams debate. They are not intended to advance towards sustainable develop- provide a comprehensive list of what we ment. expect, but instead to illustrate some of the salient actions that we believe each group Line ministries should consider as it moves from debate over the report’s thrust into actions to ■ Issue criteria and guidelines for promot- implement its provisions. They are entry ing third party review and dispute points for follow-up. These recommenda- resolution around large dam projects. tions result not only from the Commission’s ■ Adopt the practice of time-bound review of experience with past dams, but licences for all dams, whether public or also from two years of analysis and dialogue privately owned.

The Report of the World Commission on Dams 313 Chapter 10

Civil society groups Drainage, International Hydropower Associ- The WCD is a model that ation, International Energy Agency, Inter- can be, if not replicated, National NGOs and national Association for Impact Assess- then at least adapted to international NGO ment.) other similar controversial networks ■ Promote a culture of evaluation and self- issues that are also at the ■ Participate in cross-cutting reflection to ensure continuous learning interface of environment partnerships with other from all aspects of large dam projects and development. These actors on key issues. through adopting appropriate procedures. are polarised issues such ■ Gather, analyse and dissemi- ■ Extend national committees to include a as the future of nate information widely to consultative group of NGOs, environ- biotechnology in promote transparency and mental scientists and affected peoples’ development, or the role openness. of large mines and groups. ■ Monitor compliance with extractive industry in ■ Set up joint work programmes with these agreements and assist any development. groups at the national and regional levels aggrieved party to seek to learn from past experience. Frances Seymour, World resolution of outstanding Resources Institute disagreements or to seek ■ Develop processes for certifying compli- recourse. ance with WCD guidelines. ■ Actively assist in identifying ■ Extend national and international the relevant stakeholders for water and databases, such as the ICOLD World energy projects using the rights-and-risks Register of Dams, to include social and approach. environmental parameters. ■ Contribute to the establishment of International Organisation for appropriate forums for stakeholders to Standardisation (ISO) enable them to identify, articulate and ■ Explore the codification of the Commis- represent their legitimate rights. sion’s guidelines in a sector-specific Affected peoples’ organisations guidance document or standard that incorporates social impact management ■ Develop proposals for follow-up dialogue as well as public reporting and negotiated on the Commission’s report with govern- decision-making. A high degree of ment authorities and project developers. developing country and civil society ■ Identify unresolved social and environ- participation in the standards-setting mental impacts and convince the rele- process is needed to ensure legitimacy. vant authorities to take effective steps to address them. The private sector ■ Develop support networks and partner- Suppliers, contractors, developers, ships to strengthen technical and legal and consultants capacity for needs and options assessment processes. ■ Publicise the acceptance of the Commis- sion’s principles, criteria and guidelines Professional associations and in corporate policy and company litera- agencies ture. (International Commission on Large Dams, ■ Abide by the provisions of the anti- International Commission on Irrigation and bribery convention of the Organisation

314 Dams and Development: A New Framework for Decision-Making Beyond the Commission – An Agenda for Change

for Economic Co-operation and Devel- rate social responsibility The example of the WCD opment. policies and statements. shows that establishing a ■ Adopt integrity pacts for all contracts ■ Use the Commission’s basic measure of trust and procurement, as developed by guidelines as social and among actors in a conflict- Transparency International. environmental screens for ridden environment is time- ■ Develop and adopt voluntary codes of evaluating support for, and consuming and costly, but conduct, management systems and investment in, individual launching a sustainable certification procedures for best ensuring projects. mechanism for consensus and demonstrating compliance with the building and standards Commission’s guidelines, including, for Bilateral aid agencies setting requires no less. example, through the ISO 14001 man- and multilateral Wolfgang Reinicke and agement system standard. development banks Francis Deng, Critical Choices ■ For consulting companies, refine the use of the tools proposed by the Commission ■ Develop programmes to so they become standard industry prac- help countries, especially those with a tice. These include distributional analy- significant existing or potential dam sis, multi-criteria analysis, risk and population, formulate a response to the sensitivity analysis, rights-and-risks Commission’s report and find ways to approach, and environmental flow implement its recommendations. assessments. ■ Ensure that any dam options for which ■ Put in place mechanisms to ensure that financing is approved emerge from an designers of dams either participate in or agreed process of ranking alternatives at least receive evaluations of predicted and respect the Commission’s guidelines. social, environmental, financial, and ■ Accelerate the shift from project- to economic performance five years after sector-based finance, especially through construction in order to learn from their increasing financial and technical experience. Make these evaluations support for effective, transparent, and available to the public. participatory needs and options assessment, and the financing of non-structural Private financiers alternatives. ■ Develop criteria for innovative bond- ■ Review the portfolio of past projects to rating systems for use in financing all identify those that may have under- options, including large dams, in the performed or present unresolved issues water resources and electric power and share in addressing the financial sectors. burden of such projects for borrower ■ Recognise the risk reduction opportuni- countries. This may include, for example, ties inherent in proper options assess- cancelling the outstanding debt related ment and develop legally binding ar- to them, converting debt repayment into rangements on environmental and social development assistance targeting affect- matters in assessing insurance, equity, ed areas, or providing new support to and bond premiums. help borrower countries address unresolved ■ Incorporate the principles, criteria, and economic, social, and environmental guidelines of the Commission in corpo- problems.

The Report of the World Commission on Dams 315 Chapter 10

■ Review internal processes ■ Review technical guidelines, norms, and At the beginning many and operational policies in practices regarding water and energy doubted the Commision's relation to the Commission’s resources development to integrate the independence but if you've recommendations to deter- principles, guidelines, and criteria from followed it closely, you will mine changes needed in the the Commission’s report. realise the Commission has selection of projects for ■ done excellent work and its Provide support to improve national lending portfolios; the independence has been capacity for options assessment in appraisal process; and imple- respected.There was a spirit developing countries. mentation, monitoring, and of openness, of evaluation. United Nations Environment understanding, a dialogue Programme between partners. Export credit agencies ■ Actively promote the results of the Aboubacry Mbodji, ■ Introduce and adopt com- CODESEN, a coalition of Commission in its co-ordinating and mon environmental, social NGOs concerned with catalysing environmental work within development and transboundary criteria the UN family of organisations. for financial guarantees and ■ strengthen institutional capacity to Examine how the Multilateral Environ- appraise projects against such criteria. mental Agreements that it administers might serve to strengthen the spirit and ■ Improve co-ordination among agencies advance the proposals of the Commission. at the international level to ensure that dam projects refused by one agency are Academic and research bodies not accepted by others. ■ Require private-sector applicants for dam ■ Assist in the evaluation of further dam case projects to meet due diligence criteria or studies following WCD methodology. voluntary codes of conduct that conform ■ Undertake research on alternatives to to the Commission’s recommendations. dams such as demand side management ■ Promote consultation and information and ensure these are available to deci- disclosure as normal procedure. sion-makers through the options assessment process. Intergovernmental ■ Assist in improving the WCD organisations Knowledge Base as outlined in Box 10.1. The United Nations ■ Facilitate the adoption of a General Continuing the Dialogue Assembly resolution on the Commis- The previous section identified a small sion’s findings and encourage the inclu- selection of specific recommendations aimed sion of the report in the Rio+10 process. at the major groups of participants in the dams debate. It focused on actions that UN technical agencies could be taken immediately and that, in (World Health Organisation, Food and many cases, are specific to the target group Agriculture Organisation, United Nations they are directed at. Development Programme, United Nations Educational, Scientific and Cultural Organi- We choose not to go beyond this for two sation, etc.) principal reasons. First, the Commission has

316 Dams and Development: A New Framework for Decision-Making Beyond the Commission – An Agenda for Change

found that most problems associated with In fact, this is already happening. Whether dams result from faults in the process of or not connected to the WCD process, the options assessment and decision-making on debate around dams continues. For example, energy and water development choices. It policies and institutional responses are evolv- follows that the solution lies in making ing, due to improved awareness, in Brazil, improvements to the process – improve- Sri Lanka, United Kingdom, Thailand and ments that will make it more even-handed, Nepal. Actors continue to respond with equitable, transparent, and inclusive. How tailor-made solutions to local issues such as: the process is designed and conducted will ■ proposals for a national commission on depend, to a large extent, on country or dams, modelled on the WCD; setting. There is no universally applicable model – only basic principles. Box 10.1 Priorities for strengthening the knowledge base A major constraint the Commission faced was a lack of comparative data on the The second reason is more significant. development effectiveness of large dams and the actual direct and indirect Unlike other Commissions, whose reports consequences of such dams on local, regional, and national development, and more specifically on affected people and environments. Much information is were essentially aimed at governments or either not available or not shared by those who hold it. Detailed studies are the international community, our report has needed that are comparative, comprehensive, integrated, long-term, cumulative, and adaptive. There are numerous areas about which the Commission a broad and diverse target. It is aimed found insufficient information on which to base strategic recommendations. The priorities for more information and understanding through specific equally at governments, international research, data collection, monitoring, and evaluation include: organisations, multinational companies, ■ Studies to compare the direct and indirect benefits and costs of multipur- financiers, consultants, NGO networks, pose dams with alternative options. indigenous communities, and locally organ- ■ Research to inform strategic thinking on the cumulative impacts of a cascade ised groups of people affected by dams. of dams or interbasin transfers on river basin environments and populations and on local, regional and national development. There is no natural heir to our work other ■ Information on the potential for multiplier effects relating to food produc- than the complex and multi-faceted com- tion, water use efficiency, poverty alleviation, and non-farm linkages of dam- munity of those concerned with dams issues. supplied major irrigation projects with other irrigation and agriculture options. ■ The impacts of main-stem and tributary dams on downstream ecosystems, It is appropriate that we seek to build no including deltas and the livelihoods of people using those ecosystems. lasting international edifice but instead ■ A better understanding of the extent to which managed floods can offset the prefer to send our report for implementation impacts of dams on downstream ecosystems and livelihoods. where it belongs – to the regional, country, ■ Improved understanding of how dams impact on women and gender relationships. basin, community, and dam-specific level. ■ Improved understanding of how to influence rural-urban migration and the We hope and expect that the report will requirements of meeting needs through decentralised rural development lead to regional and national processes of compared to the challenge of servicing mega-cities for water and energy. dialogue, to discussions concerning specific ■ Baseline studies of greenhouse gas emissions from pre-project rivers for comparison with emissions from post-project reservoirs on those rivers, to projects or issues, to new research and facilitate study of how the change from a natural to a human-modified training programmes, to new networks – in system influences greenhouse gas emissions. short, to a multiplicity of results and develop- ■ In anticipation of global warming, research and strategic thinking on the ments that are decentralised, dispersed, and impact of past periods of drought on the ability of large dams to deliver expected services in arid, semi-arid, and drought-prone areas. highly diverse in character. We hope that they ■ Identification of techniques for recharging aquifers on a larger scale as a will include many new partnerships across storage technique for avoiding water loss through evaporation, and for sectors where earlier divides existed, connect- improving conjunctive and sustainable use of surface and ground water. ing to understand differing interests and to ■ Development of improved policies and practices to ensure that cultural heritage is dealt with in the planning and implementation of dam projects, identify shared objectives. This would be in given the importance of river basins for the origins of human civilisation. keeping with the spirit of the Commission.

The Report of the World Commission on Dams 317 Chapter 10

■ multi-stakeholder workshops convened needed to resolve them is still not available. to receive a debriefing following the last The process of information gathering, meeting of the WCD Forum and to analysis, learning, consultation and review develop a national dialogue; must go on (see Box 10.1). ■ independent mechanisms to review and resolve escalating conflicts on dam Moreover the Commission, in the course of projects; its work, generated a wealth of ideas for the future. These ranged from policy, regulatory ■ reviews of export credit policies follow- and institutional tools to best practice ing controversy on particular dams; and experience at the community level. Many of ■ reviews of dam safety and compensation these ideas inspired the principles and policy for dam-affected people. guidelines set out in our report. But there are also many that proved too detailed, too We expect that such examples will multiply as context-specific, or too untested to include our report is disseminated and becomes part of here. They should not be lost. For example, the ongoing discourse about dams and devel- the WCD Knowledge Base can be used to opment. develop a series of practical tools in as many language versions as possible. Publishing a The debate about dams began regular State of World Dams Report is Unlike other well before the Commission, and another possible project. Ideas for national Commissions, whose it will continue well beyond it. or international recourse mechanisms and reports were essentially We hope that one of the lasting co-ordinated monitoring of compliance aimed at governments results of the WCD process will include creating national dams inspection or the international have been to change the tenor of panels, developing a register of accredited community, our report that debate from one of lack of independent experts for review panels, and has a broad and diverse trust and destructive confronta- establishing a system for assuring that target. It is aimed tion to co-operation, shared goals individual dams or stages in dam projects equally at and more equitable development conform with the Commission’s guidance, governments, outcomes. We must not, however, through an independent and iterative international give the impression that the verification process along the lines of the ISO organisations, Commission has laid the dams 14001 third-party certification mechanism. multinational debate to rest. Several important companies, financiers, elements of that debate must None of these follow-up ideas, nor much of consultants, NGO continue in order to carry the what lies in the body of the report, will networks, indigenous discussion on dams forward in the advance very far without a concerted and communities, and context of meeting development large-scale effort to build the necessary locally organised goals. expertise and institutional capacity for groups of people implementing them. This point cannot be affected by dams. For one thing, the Knowledge stressed enough. It is one thing to propose a Base is not complete. Extensive model or set of actions that will sweep away as the information gathered, structured, and the problems encountered in pursuing analysed by the Commission may be, there energy and water development, especially are still gaps in our knowledge, experience, when dams are involved. It is quite another and understanding. Many issues cannot to put the solutions patiently in place. finally be settled because the information Calling upon developing countries to slot

318 Dams and Development: A New Framework for Decision-Making Beyond the Commission – An Agenda for Change

proposals into a framework that does not without compromising individual values or exist, or that is incapable of accommodating losing a sense of purpose. But it also demon- them, is a recipe for frustration. The Com- strates that all concerned parties must stay mission is convinced that helping to build together if we are to resolve the issues the capacity to consider fully all options for surrounding water and energy resources water and energy development is as impor- development. It is a process with multiple tant as implementing any choices that are heirs and no clear arbiter. We must move made. Often the latter cannot take place forward together or we will fail. without the former. We do not assume, of course, that there will We hope that the momentum generated be no further disagreement. Dynamic debate during the past three years – from the Gland leads frequently to better outcomes, as meeting, through all the papers and reviews controversy can often be written for the Commission, to the actions empowering. We do, however, taken to launch and disseminate the report believe that we will not turn Sri Lanka and other – will be sufficient to carry the process the corner and put the princi- countries have already forward. The establishment of the Commis- pal conflicts around water and benefited from the WCD sion opened an important space for debate energy development behind us process... For example, Sri and dialogue, and that space has been fully unless the participants in the Lanka has appointed a used. Further spaces must now be created debate decide to work together committee to examine the and filled. locally, nationally, and global- grievances of dam-affected ly, in the spirit of the Commis- people and to compensate A Call to Action sion. victims, as well as to take action on dam safety issues The work of the World Commission on The world never stands still. not undertaken earlier. Dams is over. But the concerns that led to The context in which deci- Tilak Ranaviraja, Ministry of its establishment are still with us. Dams sions on energy options and Mahaweli Development have too often left a legacy of social injus- water development must be tice and environmental damage, and that made is in constant evolution. Technology legacy does not disappear because we have is advancing, democracy and governance identified a better way of doing things in reform is spreading, the market is changing future. Early and resolute action to address and the demand for greater equity is growing some of the issues arising from the past will more resolute and persistent. As noted in go a long way to building the trust required Chapter 1, this century will see increasing to enable the different actors to work stress on water resources worldwide. The together. So, too, would an assurance to assessment of water and energy options and countries still at an early stage of economic of the role dams play, increasingly takes development that the dams option will not place against a background of competing be foreclosed before they have had a chance sectors and interests. Growing needs and a to examine their water and energy develop- dwindling resource base, in both quantity ment choices within the context of their and quality, will require many countries to own development process. fundamentally reassess their water management policies. We are convinced that The experience of the Commission demon- business as usual will not prove to be a strates that common ground can be found viable strategy.

The Report of the World Commission on Dams 319 Chapter 10

The Commission has not performance of dams and their contribution to The work of the World sought to divert these global development. We have done this through an Commission on Dams is trends or to predict their future inclusive process that has brought all signifi- over. But the concerns scale and direction. We have, cant players into the debate. And we believe that led to its instead, tried to bring them we have shifted the centre of gravity in the establishment are still together, in some form of dams debate to one focused on options with us. Early and harmony, on a score card that assessment and participatory decision-making. resolute action to address the range of actors in the The rights-and-risks approach we propose will some of the issues arising debate can use. We believe this raise the importance of social and environ- from the past will go a report is a milestone in the mental dimensions of dams to a level once long way to building the evolution of dams as a develop- reserved for the economic dimension. trust required to enable ment option. We have con- the different actors to ducted the first comprehensive We have told our story. What happens next work together. and global review of the is up to you.

320 Dams and Development: A New Framework for Decision-Making Comment

A Comment – Medha Patkar

1. The process of the World Commission 3. The frequent failure of large dams to on Dams was unprecedented in bringing provide their claimed benefits and this together so many of those involved in poor performance needs to be recognised debates and conflicts over large dams. and accepted. There is no reason for Many peoples’ movements and NGOs – optimism on the feasibility of improving the constituency that first proposed a the poor performance of dams and comprehensive and independent review mitigating their impacts. A major – actively participated. Our final Report question is the feasibility of just rehabili- has synthesised enormous amounts of tation with land for land lost by agricul- information and diverse opinions and turists and alternative, appropriate delivered many important findings and sources of livelihood for other displaced recommendations. I hope it will be a people. In large scale displacement, the reference for all those concerned about experience shows a clear failure. Within large dams. While signing the Report the value framework the Commission because of its many positive aspects, I propagates – equity, sustainability, still feel I must put forth this opinion on transparency, accountability, participa- some fundamental issues that are missing tory decision-making, and efficiency – or not given the central place they large dams have not helped attain, but deserve. rather hindered, “human development”.

2. The problems of dams are a symptom of 4. An inclusive, transparent process of the larger failure of the unjust and decision-making with equal status to all destructive dominant development the stakeholders, equal place for local model. It is beyond the scope of our and national needs and plans, equal report or the brief of the Commission to significance to social, environmental, resolve all the underlying problems of technical and financial aspects of plan- global development. But addressing these ning, would be a great advance, but does issues is essential in any attempt to reach not go far enough. Even with rights an adequate analysis of the basic systemic recognised, risks assessed and stakehold- changes needed to achieve equitable and ers identified, existing iniquitous power sustainable development and to give a relations would too easily allow develop- pointer towards challenging the forces ers to dominate and distort such process- that lead to the marginalisation of a es. These developers include multilateral majority through the imposition of institutions like the World Bank that unjust technologies like large dams. have pushed many large dams despite

The Report of the World Commission on Dams 321 Comment

non-compliance with their own policies. 7. No undue legitimacy should be granted The State, controlled by powerful vested to corporations and international financ- interests, may do the same. Understand- ing agencies. The sovereignty of both ing this takes us beyond a faith in people and the nation-state must not be negotiations to emphasize certain compromised for anything but the basic priorities and primacies. values and goals of humankind. It is necessary to give a serious critique of the Communities, especially those who live privatisation of the water and power on and seek livelihood from their natural sectors and the resulting marginalisation of resource base, such as forest produce local people and corporate domination gatherers, farmers or fisherpeople, should over natural resource-based communities. have the first right to planning, development and management of those resourc- 8. The issues above are those raised by the es. Inequities within communities also peoples’ movements whose role and need to be recognised and addressed. perspectives should be given their due Social and environmental parameters place. Not just with stories of eviction, must have a higher weightage than the repression and confrontation, but with technical and financial aspects in deci- their ideologies, strategies, and vision. sions concerning human development. It is necessary to stress the ‘principle of 9. Over and above all this, I recognise and subsidiarity’, according to which devel- share the Commission’s achievement of opment planning would be based on local to global consultation and, more so, micro-catchments, working from ridge to a humane, well-intentioned, open and river, and from origin to sea. frank dialogue under an able leadership, which needs to be kept alive beyond the 5. A full assessment of the options for short life of this forum. To endorse the meeting water and energy needs as the process and many of our findings and first part of project planning needs to be recommendations, I have signed the supported. But only creating a level Report. To reject the underlying assump- playing field for options cannot suffice. tions of a development model which has We should instead give priority to more palpably failed and to caution against the equitable, sustainable and effective options massive gulf between a statement of good to satisfy basic human needs and liveli- intent and a change in practice by hoods for all before supporting the addi- entrenched vested interests I have asked tional luxuries of the few, unjustified in the for this note to be attached. face of the many who remain deprived. Whatever is missing or could not be ad- 6. The wider context of national and global dressed in the Report should form an agenda political and economic trends obviously for further dialogue and research; but also affect decisions in the water and power for struggles for justice with people at the sector. These trends include the dimin- forefront, people more empowered by this ishing role of the State, the growing report and otherwise to say NO to the marginalisation of national laws and perverted development vision, process and institutions, and the trampling of human projects. rights due to the expanding role of private capital and free trade. While there may be a few welcome instances of progress towards enhanced human rights and equity, to say there is a global trend Medha Patkar towards these goals would indeed be erroneous.

322 Dams and Development: A New Framework for Decision-Making Index

Index

The index does not cover the Executive Summary (pages xxvi-xxxvii) or the glossary (Annex II). Page numbers given in italics indicate illustrations, figures or tables.

A loss in fish production 85 see also Tucurui dam activist groups 18-20 sediment management 81 Earth Summit (1992) 234 ADB 41, 45, 47, 48, 54, 57, 62, 114, 171, Australia 11, 75, 79, 82, 83, 88, 139, 145, financial assistance 172 188 159, 231, 296, 304 participation in decision-making 177 African Development Bank (AfDB) 41, Austria 91, 177 Brazil Ceara Water Resources 128 54, 56, 62, 171, 188 British Columbia 53 Agno River basin 111 B Buffalo Creek dam 64 agriculture 6, 12, 42-6, 100-101, 137-48 Bakolori dam 18, 35, 112 bulk water supply 56-7 Akosombo dam 81, 107, 108, 110, 115, Bakun project 111 Burnett River 296 116, 118 Balbina dam 111 Burundi 104 alternatives to dams 23, 135-64, 178-80, Bangladesh 13, 111, 146, 160 178, 221-24, 262, 263, 266-70 banks 188-90, 315-16 C Amu Darya River 88 Bargi dam 106, 128 Cambodia 145, 145 Andra Pradesh II & III 128 baseline assessments Cameroon 84, 85, 231, 239 aquatic ecosystems 77-83, 90, 234-39 ecosystems 293-94 Canada 4, 11, 14, 31, 59, 112, 128, 226 Aral Sea 88 social conditions 296-97 see also Grand Coulee dam Argentina 106, 117, 144, 305 Batang Ai dam 107 canal lining 140 Arun III dam 19, 172 Bayano dam 111 capital costs 39-41, 48, 49, 70 Asian Development Bank see ADB benefit-sharing mechanisms 243, 253-54, Cariño, Joji ix, 395 Aslantas dam 31, 42, 44, 49, 56-7, 63, 82, 300-301 Case Studies ix, 30, 31, 38, 40-42, 44-45 85, 116 benefits see costs and benefits 53-54, 350-53 alternatives considered 178 Benin 81, 115 see also names of specific dams cost recovery 48, 48 Bhagani Tildeh river catchment 144 Cerro de Orro dam 107 costs and benefits 100, 121, 121, Bhumibol dam 107 Ceyhan Basin 31, 173 124, 125 bilateral financing agencies 187-90, 315 Chad 112 cropping pattern and yields 45-6 bilharzia 118 Chagga people 174 cultural heritage 117 biodiversity Chakma people 111 displaced communities 105, 106, in aquatic ecosystems 77-83 Challawa Gorge dam 113 107 in terrestrial ecosystems 75-7 Chandil dam 18 financial assistance 173 Convention 234, 235, 238 Chashma project 67 irrigation 44, 46, 116, 141, 190 biomass systems 154 Chico River 19 Asmal, Kader viii, 28, 394 birds 78, 83, 87 Chikugo River 59 Asurini people 107, 122 black stilts 78 Chile 19, 111, 143 Aswan High dam 13, 59, 107 blackfly 80 China 4, 7-9, 10, 13, 14, 17, 21, 62, 66, cultural heritage 117, 117 Blackmore, Donald ix, 395 100, 107, 113, 180 displacement 103 Botswana 84 see also Three Gorges dam flood management 59 Brazil 11, 14, 84, 101, 101, 107, 108, 110, benefit-sharing 127, 128 health 118 111, 112, 127, 141, 144, 231 corruption 187

The Report of the World Commission on Dams 399 Index

cultural heritage 118 existing 225-33, 291-93 environmental impacts 21-2, 25, 223-24, displacement of communities 104, in the pipeline 276-77 289, 289, 291-92 107, 108, 109, 110 multi-purpose 39, 44, 62-3, 68 see also ecosystems impacts electricity 152 opposition to 18-21 Epupa dam 19, 117 environmental conditions 190 Danjiangkou project 108 equity 125-29, 204, 205 financial assistance 189 Danube River 177 Ertan project 127 flood management 160, 161 decision-making 169-75, 199, 202, 205, European Union 15 health issues 118 206, 208-11, 263, 263, 280-81 European Wind Energy Association 154 irrigation 140, 141, 144, 144, 146 participation in 176-77, 177, 204, evaluation studies 47, 54, 184, 226-27, reparations 128 215, 217, 219, 222 274, 312 water supply 158 decommissioning 10-11, 92, 184-85, 232-33 existing dams see dams Chinantec Indians 107 demand-side management 148, 149, 150- export credit agencies (ECA) 188, 246, Chixoy dam 18, 106 51, 157-60, 180 312, 316 Churchill Rivers project 111 Denmark 157 civil society organisations 205, 314-16 Dhom dam 106 F climate change 7-8, 22, 53, 60, 64, 69, 74, Diama dam 119 financial agencies 188-90, 315-16 75-7, 149, 223 diseases 115, 118-19 financial internal rate of return see FIRR Colombia 105, 127, 142, 172 displacement of communities 16-17, 102- financial profitability Colorado River ii, 78-9, 92, 125, 143, 145 112, 207, 240 hydropower dams 54-6 Columbia Basin project 44, 45, 47, 47, see also resettlement irrigation dams 46-8 49, 59, 62, 122, 128 dispute resolution 307 water supply dams 58 Columbia River 80, 82, 88-9, 175, 231 distributional analysis 288 findings see WCD findings Columbia Treaty (1968) 175 Dongting Lake 161 Finland 91 Colville tribe 106, 128, 190 drainage 139, 161 Finnish International Development command area 43, 43, 70 Agency (FINNIDA) 174 compensation 105, 106, 107, 114, 124, E FIRR 230, 238, 241, 242 Earth Summit (1992) 234 irrigation dams 46 see also reparations Echo Park dam 19 water supply dams 58 compliance 185-90, 190, 204, 244-50 economic growth 4-5 Fisching Project 177 compliance plans 244, 301 economic internal rate of return see EIRR fish 16, 78, 79, 80, 81, 82-3, 84-6, 119, construction costs 39-40 economic profitability 231, 237, 295-96, 296 construction phase 99-100 hydropower dams 54-6 fisheries 84-6, 90, 113, 295-96 contracts 242-43, 298-300 irrigation dams 47-8 flood control Convention Concerning the Protection & water supply dams 58 benefits 58-59 Integration of Indigenous & Other ecosystem enhancement 86-7, 231 limitations 60, 83 Tribal & Semi-Tribal Populations in ecosystems dams 58-62 Independent Countries (1959) 23-4 baseline surveys 293-94 flood management 14-15, 60-62, 68, 84, corruption 186-87, 249 impacts 10, 15-16, 16, 21-2, 73-93, 160-63, 161, 162, 239 cost 234-39 floodplain ecosystems 83-4, 90, 145, 145, overruns 39-42, 54, 68 Egypt 13, 59, 66, 107, 116-18, 145 147, 237-38 recovery 38, 48-9, 56, 62 EIRR Food and Agriculture Organisation cost-benefit analysis (CBA) 46, 180-82, 186 Ghazi-Barotha project 291 (FAO) 140, 316 Costa Rica 172 irrigation dams 46, 47, 48 foreign assistance 170-73, 173, 174 costs and benefits 37, 21, 120-29 Kariba dam 55 Forestry Stewardship Council (FSC) 246 see also cost-benefit analysis multi-purpose projects 62 forests 75, 76, 83, 161 country studies 30, 350, 351 water supply dams 58 France 4, 10, 81, 91, 92, 231 cropping intensity 43, 45, 70 electricity 14, 101, 101, 116, 121, 128, free, prior & informed consent 218-219, cropping patterns 45 148-56, 150 280-82 Cross-Check Survey ix, 28, 30, 31, 42, see also hydropower Freudenau dam 177 43, 44, 50, 53, 62, 64, 353-56 Embera people 107, 111 Friesach project 177 cultivation techniques 141 employment 101-102, 115 Funtua dam 104 cultural heritage 116-18, 285 Endangered Species Act (1974) 19 energy resources 148-56, 153-54 G D entitlements 240-43 Gansu province 144, 144 Daguangba Multipurpose project 128 environmental flow assessments (EFA) Gariep dam 44, 115, 121, 122, 124 dam safety 63-5, 186, 272 294-95 Gavaio da Montanha people 107, 122 dam-building countries 9, 11 environmental flow releases (EFR) 238-39 gender disparities 114, 116, 216 Damodar Valley Corporation 60 environmental impact assessments (EIA) Germany 4, 14, 154, 158 dams 19, 89, 177, 182-83, 183, 187, 236, Gezira-Managil irrigation schemes 140 debate 2-3, 17-21 241, 282-85 Ghana 81, 107, 108, 110, 115, 146

400 Dams and Development: A New Framework for Decision-Making Index

Ghazi-Barotha project 128, 291 Hoa Binh 107 International Commission on Large Dams Gland Workshop iii, vii, 18, 21, 27, 28 Houay Ho dam 108 (ICOLD) 11, 27, 65, 73, 172, 173, Glen Canyon dam 78-9, 79 human rights see rights 314, 370 global warming see climate change Hungary 185 International Court of Justice 254 Glomma and Laagen Basin 31, 40, 50, 53, Hydro-Québec 128 International Covenant on Economic, 59, 66, 103, 179 Social & Cultural Rights (1966) 24 hydropower 12, 14, 101, 102 alternatives considered 178 International Energy Agency 91, 314 hydropower dams 49-56, 62, 68 costs and benefits 121, 121, 124, International Energy Association (IEA) 73 Hydropower and Environment 91 125 International Finance Corporation (IFC) fish passes 82 188, 230 political decision 170 I International Hydropower Association (IHA) 73, 173, 314 services 121, 122 Ibaloy people 111 International Labour Organisation Goldemberg, José ix, 395 ICOLD World Register of Dams 314 (Convention 169) 216, 219 governance 22-3, 205, 209 Iguacu River 177 Ilisu dam 19, 188 International Organisation for Standardi- Grand Coulee dam 31, 44, 49, 50, 51, 53, sation (ISO) 246, 314, 315, 318 66, 100 impact assessments 241, 282-85, 306-307 see also environmental impact International Rivers Network (IRN) 27, 41 alternatives considered 178 assessments International Water Management cost recovery 62 environmental flow assessments Institute 68 costs and benefits 47, 121, 122, 123- 294-95 investment 11, 13, 249-50 24, 124, 125, 190 impoverishment risk analysis 241, 297-98 Iran 10 cultural heritage 117 Inanda dam 117 Ireland 295 displacement of communities 105, India 4, 7, 9, 10, 10, 17, 19, 21, 41, 58, irrigated area 43, 43, 44-5 106, 113, 115 66, 104, 172 irrigation 12-13, 13, 100-101, 137-48 dams 42-9, 62, 68 ecosystems 80, 81, 82, 85, 86, 88-9 see also Sardar Sarovar project systems 138-43 employment 99, 102 costs and benefits 123-24, 127 ISO 14001 management system standard flood control 59 cultural heritage 117, 118 displacement of communities 18, 257, 315, 318 performance 55, 55 Israel 141 political decision 170 106, 107, 108, 110, 115 ecosystems 75 Itá dam 107, 110 reparations 128 electricity 152 Itaipu dam 127 Great Fish River 80 environmental conditions 190 IUCN see World Conservation Union greenhouse gases (GHG) 74, 75-7, 76, 77, financial assistance 171-72, 188 90, 92, 122, 149, 223, 287-88 flood control 60 J gross value of production 45-6 health issues 118 Jain, Lakshmi Chand viii, 395 groundwater abstraction 146 irrigation 13, 44, 100, 140, 141, 144, Jama’are River 113 groundwater pumping 139, 139 145, 146 Japan 9, 10, 59, 60, 127, 157, 158, 231 Guatemala 18, 106 reparations 128 Japanese encephalitis 118 Gwembe Tonga people 18, 106, 107, 114, water supply 157, 158 Jordan 141 119 indigenous peoples 19, 38, 105, 106, Gwembe Tonga Rehabilitation and 110-12, 207, 216, 218-20, 256 K Development Programme 128 Indonesia 4, 105, 108, 113 Kainji dam 85, 109 Indus Basin Irrigation System (IBIS) 44, Kano River project 113 48-9, 67, 122, 122, 140 Kao Laem dam 105, 107 H Indus River 59, 90 Kaptai dam 111 Hadejia River 113 industrial usage of water 13-14 Kariba dam 18, 31, 40, 42, 50, 52, 53, 55 Hainburg dam 177 Innu people 112 access to electricity 128 Hanoi 14 Inspection Panel 19 alternatives considered 178 Harvey Basin Restoration Trust 75 Integrity Pacts 249, 305 costs and benefits 121, 121, 124, 125 health issues 100, 115, 118-20, 236, 241, Inter-American Development Bank 41, displacement of communities 103, 284 171, 172, 188, 216 106, 115, 190 Intergovernmental Panel on Climate Hells Canyon dam 82 ecosystems 75, 85, 86 Change (IPCC) 7 Helsinki Rules (1996) 252, 254 employment 99, 102 International Association for Impact Henderson, Judy viii, 395 financial assistance 173 Assessment 314 Hendrik Verwoerd dam 170 health 118, 118-19 International Bank for Reconstruction & political decision 170 Hillsborough dam 86 Development (IBRD) 170 reparations 128 Himba people 117 see also World Bank Kedung Ombo dam 108 HIV/AIDS 100, 115, 119 International Commission on Irrigation Kelly Barnes dam 64 Ho Chi Minh City 14 & Drainage (ICID) 27, 173, 314 Kenya 84, 104, 107, 146

The Report of the World Commission on Dams 401 Index

Kiambere Reservoir 104, 107 Morocco 66, 145 options for energy & water needs see Kolyma River 113 Morse Report (1992) 26 alternatives to dams Korea 237 Mozambique 86 Orange River 44, 80, 81, 88, 125, 170, 178 Kotri barrage 88, 112 Mubuku project 109 Development Project 102, 122 Kpong dam 110 Muda Irrigation Scheme 141 Pilot Study 80 Kuna people 107, 111 multi-criteria analysis 223, 224, 236, 285- O’Shaugnessy dam 18 Kyoto Protocol 287 86 multi-purpose dams see dams P L multilateral financing agencies 187-90, 315 Pak Mun dam 19, 31, 40, 51, 55, 113, 115 Lake Chad 84 Multilateral Investment Guarantee alternatives considered 178 Lake Roosevelt 122 Agency (MIGA) 230 costs and benefits 55, 121, 124, 125 Laos 20, 75, 108, 113, 145, 183 Mun River 84 cultural heritage 117 large dams Mur River 177 displacement of communities 104 functions 12, 12 Murray Darling Basin 88, 231 ecosystems 82, 84, 85 regional distribution 8 Murray River ii, 79, 83, 140 environmental impact assessment types 11 183, 190 legislation 127-28, 185, 186-87, 189, 200, N financial assistance 173 216, 226, 242-43, 249, 255, 298-301 Nam Ngum River 59-60, 61, 256 Pakistan 7, 13, 31, 67, 113, 122, 128, 140, Lesotho 231 Nam Theun II 20, 75 291 Lesotho Highlands Water Project 19-20, Namibia 19, 117 see also Indus Basin; Tarbela dam 119, 187 Nangbeto dam 107, 115 financial assistance 173 licensing procedures 185, 226, 232-33, Narmada River 128 irrigation 141, 145, 146 272, 274 Narmada Sagar dam 117 Panama 107, 111, 117 life-cycle assessment 286-87 Narmada Water Disputes Tribunal 104 Panama Canal 102 Lindahl, Göran ix, 395 National Environmental Protection Act Pangani Falls Redevelopment Project 174 Lingjintan project 113, 190 (1969) 19 Pangue project 111 Liu-Yan-Ba project 107 natural vegetation 139, 158-59 Papaloapan River Commission 18, 106 Logone-Chari system 84 needs assessment 222, 262, 263, 264-65 Paraguay 106, 144 Lubuge project 127 negotiated outcomes 208-211, 208 Parakaná people 106, 107, 122, negotiations 218, 219, 280-81 participation in decision-making see M Nepal 13, 19, 53, 172, 256, 317 decision-making Madden dam 117 Netherlands 160 Patkar, Medha viii, 396 Maguga project 119 New Zealand 78 Pecos valley 145 Mahaweli Development Programme 105, Niger 112 Pehuenches people 111 114, 115, 319 Niger River 83, 84, 85 performance 21, 38, 45 Maheshwar dam 188 Nigeria 18, 58, 84, 104, 109, 112, 113, 140 Columbia Basin project 47 malaria 118, 119 Nile Delta 81, 85, 145 evaluations 184 Malaysia 107, 111, 141, 162, 162 Nile River ii, 59 findings 68-9 Mali 83, 84, 112 non-dam options see alternatives to dams flood control dams 58-62 Manantali dam 84, 112, 119 NORAD 174, 183 hydropower dams 49-56 Manibeli Declaration 26 Norway 11, 31, 91, 178, 185, 190, 237 improvement 138-40, 227-28 maps 31 see also Glomma and Laagen Basin multi-purpose dams 62-3 Marshall Plan 170 nuclear power 153 water supply dams 56-8 Matsubara dam 59 Nya Heun people 108 performance bonds 247-48, 301, 303-304, Maya Achi people 18, 106 304 Mazatec people 18, 106 O Philippines 19, 58, 101, 110-11, 188 Mekong River 84, 145 O&M costs see operational & mainte- plankton 81 mercury levels 118, 119 nance costs Poland 61-2 Mexico 18, 49, 106, 109, 111, 145 OECD Convention on Combating political decision-making 169-70, 170 micro-irrigation systems 141 Bribery of Foreign Public Officials in Pollan dam 295 migration 82-3 International Business Transactions pollution 7 Miguel Aleman dam 18, 106 (1997) 187, 249 Pongolapoort dam 83, 84, 85 Mississippi River 160 OECD countries 4, 30 population 3, 17, 17 Missouri River 111 Ogallala aquifer 146 Porto Primavera dam 84 mitigation measures 82, 90, 91, 93, 241-43 Okavango Delta 84 Position Paper on dams and the Environment see also compensation; resettlement Ontario Hydro Study 65 (1997) 27 mitigation, resettlement and development operation issues 183-84 private sector 205, 207, 314-15 action plan (MRDAP) 298-300 operational & maintenance (O&M) costs project Mohale dam 231 48-9, 48, 54, 62 benefits see benefit-sharing mecha- Moore, Deborah x, 395 opposition to dams see dams nisms

402 Dams and Development: A New Framework for Decision-Making Index

implementation 262, 272-73 São Paulo 34, 101, 101 Suriname 305 operation 263, 274-75 Sardar Sarovar project 19, 26, 104, 105, Sweden 91, 237 preparation 262, 270-71 108, 114, 115, 118, 172 Swedish International Development project-level impact assessments 283-85 Saudi Arabia 146 Agency (SIDA) 174 public acceptance 177, 215-20 schistosomiasis 118, 119 Switzerland 91 Scudder, Thayer viii, 396 Syr Darya River 88 R sediment management 81-2, 293 rain-fed agriculture 143-45, 144 sedimentation 10, 16, 44, 62, 65-6, 65, 69, T rainwater harvesting 158, 158 75, 81, 138-39, 228 Tana River 84, 104 Rajasthan 144 Senegal 84, 112, 145, 231, 239 tank systems 144-45 Ralco dam 19, 111 Senegal River 84, 112, 116, 127 Tanzania 172, 174 shared rivers 173-75, 174, 251-56, 306- Ramsar Convention on Wetlands 86, 234, Tarbela dam 31, 42, 44, 46, 50, 51, 52, 53, 307 235, 238 56-7, 99 shared values 199, 199, 202, 206 recycling of water 145-46, 159 alternatives considered 178 Shepparton Irrigation Region 139 Regional Consultations ix, 28, 32, 98, costs and benefits 121, 122, 123, Shimouke dam 59 117, 357 124, 125 regulations see legislation Shuikou project 127 displacement of communities 105, reparations 128-29, 133, 229, 230-31 Silenced Rivers (1996) 27 106, 107, 112, 128, 190 resettlement 17, 20, 22, 103, 105, 106, Singkarak project 113 ecosystems 82, 88 107-110, 128, 241-43, 298-300 Sirindhorn dam 107 financial assistance 173 review panels (IRP) 302-303 Slovakia 179 Rhine River 160 Sobradinho reservoir 112 flood management 59 Rhone River 81, 160 social conditions, baseline assessments gender disparities 114-15, 115 Rift Valley fever 118, 119 296-97 sedimentation 66 rights 19, 20, 22, 24, 200-202, 203, 203, social impacts 16-17, 22, 25, 97-130, 289- tenders 270 204, 206, 256 90, 289, 291-92 Tenughat reservoir 60 rights and risks approach 202, 206-210, socio-economic impacts 99-102, 112-14, terrestrial ecosystems 75, 75 215, 216, 240, 241 182, 223-24, 228-31, 241 Teton dam 64 Rio+10 process 313, 316 Sokoto River 112 Thailand 19, 31, 105, 107, 108, 142, 145, Rio de Janeiro 234 solar photovoltaics (PV) 154 158, 183, 317 Rio Declaration on Environment and solar thermal systems 154 see also Pak Mun dam Development (1992) 24, 200, 201, South Africa 11, 20, 53, 88, 115 Thematic Reviews 28, 30, 32, 33, 356-57 391-393 see also Gariep dam; Orange River Theun Hinboun project 113, 183 Rio Grande project 127 Development Project Three Gorges dams 19, 104, 118, 187, Rio principles 201-203, 207 access to electricity 128 188, 189 risk cultural heritage 117 Tiga dam 113 analysis 241, 297-98 ecosystems 82, 83, 84, 85, 87 Tocantins Basin 173 assessments 290-91, 291 water supply 158-59 Togo 81, 107, 115 risks 26, 186, 207-208, 207, 237 South African National Water Act 231, Tong River 237 river fragmentation 87, 88-9 239, 253 Tonga people see Gwembe Tonga people River Severn 64 South Korea 10 Transparency International 187, 315 rivers see also aquatic ecosystems Spain 9, 11, 15, 185 Trinidad 86 flow regimes 78-81, 81 Spokane tribe 106 Trushuli-Devighat hydropower station 53, restoration 10, 234-36 Sri Lanka 51, 105, 114, 115, 145, 317 256 shared 173-75, 174, 251-56, 306-307 Sri Sailam project 106, 107 transformation by dams 15-16, 15 stakeholder analysis 279-80 trust funds 247-48, 304-305 Role of Dams for Irrigation, Drainage and stakeholders forum 217-18, 279-80, 286 Tucurui dam 31, 42, 50, 53, 102 Flood Control (2000) 27 statistics 368-82 alternatives considered 178 rural electrification 154-55 Steiner, Achim x, 396 costs and benefits 121, 121, 122, Russia 11, 14 Sterling Forest 159 123, 124, 125 Ruzizi project 104 strategic impact assessment 282-83 displaced communities 105, 106, 107 Rwanda 104 submissions 29, 32, 211, 357-58 ecosystems 75, 77, 82, 85-6, 86 Sudan 84, 112, 118, 118, 140 financial assistance 173 S Sudd 84 financial performance 55, 56 safety see dam safety Sulawesi 105 health issues 119, 119, 122 Sahel 83 supply-side options and legislation 127, 190 salinity 66-8, 69, 139-40, 139, 159 energy 151-54 Turkey 10, 19, 31, 44, 45-6, 48, 117, 131, salmon 78, 81, 82, 85, 89, 113, 295 water and irrigation 143-48 173, 188 Salto Caixas project 177 water supply 158-59 see also Aslantas dam San Roque dam 111, 188 surface irrigation 46, 141, 228 Turkmenistan 66

The Report of the World Commission on Dams 403 Index

U Universal Declaration on Human Rights priorities for strengthening 317 Uganda 109, 157-58 (1948) 23, 387-90 WCD findings UN Convention on Biological Diversity Upper Krishna project 118 compliance 190-91 Upper Sinú River 105 decision-making 188-89 234, 235, 238 urbanisation 4 ecosystems impacts 92-3 UN Convention on Law of Non-Naviga- Urrá 1 dam 105, 113 needs 163-64 ble Use of International Watercours- planning 191-92 es 175, 252, 254, 256, 306, 307 socio-economic impacts 129-30 UN Declaration of Human Rights (1947) V Vaal River 88 wetlands 83-4, 86-7, 145, 145 200, 202, 387-90 Wetlands International 86-7 UN Declaration on the Right to Develop- values see shared values Van der Kloof dam 53, 115, 121, 122, 124 wind power 153-54 ment (1986) 24, 200-201, 202, 383-86 Working for Water programme 158-59 UN Draft Declaration on the Rights of vegetation 75, 76, 77, 80, 83, 94, 139, 158-59 World Bank 19, 20, 27, 40, 41, 42, 45, 47, Indigenous People 216 Veltrop, Jan x, 396 48, 49, 54, 58, 62, 67, 171-72, 182, UN Millennium Report 199, 209 Venezuela 144 230 UNDP Human Development Report Victoria dam 51 Operations Evaluation Division (2000) 203, 203 Vietnam 14, 62, 107, 145, 185 Report (OED)(1996) 26, 27, 41, 46, UNDP World Energy Assessment 150-51 Volta River 81, 85, 110 56, 114 United Kingdom 158, 159, 317 policies 188 United Nations 140, 316 World Commission on Dams see WCD United Nations Charter (1945) 23, 199 W United Nations Conference on the Waimiri-Atroari people 111 X Waitaki River 78 Human Environment (1972) 182, Xiaolangdi dam 109-110 201 water appropriation methods 143-45 United Nations Development Programme quality 7, 88, 139-40, 158 Y 316 rights 143 Yacyreta project 106 United Nations Educational, Scientific & supply 156-60 Yakutia 113 Cultural Organisation 316 supply dams 56-8 Yangtze River 62, 104 United Nations Environment Programme tariffs 58, 142, 147, 157 Yantan project 127 316 water-stressed countries 6, 7, 7 Yellow River ii, 107 United States 4, 9, 10, 13, 14, 15, 21, 31, waterlogging 66-8 Yemen 145 61, 111, 169, 170, 185, 226 WCD yields 12, 45-6 see also Grand Coulee dam establishment 25-8 corruption 187 mandate 28, 195 Z ecosystems 91, 93, 231 members viii, ix, x, 394-6 Zaire 84, 104 flood management 160 policy framework 202, 203-206 Zambezi River 85 irrigation 140, 143, 146 process 29 Zambia 31, 107, 114, 121, 128, 145 water policy 237 Secretariat 397-8 see also Kariba dam water supply 159 WCD Forum 27, 29, 31 Zimapan resettlement 109 United States Agency for International WCD Global Review xxviii-xxx Zimbabwe 31, 91, 107, 115, 121, 237 Development (USAID) 173 WCD Knowledge Base 29, 38, 56, 58, 68- see also Kariba dam United States Bureau of Reclamation 173 9, 359-67 access to electricity 127-28

404 Dams and Development: A New Framework for Decision-Making