[University of Defence] at 01:15 24 May 2016

Downloaded by [University of Defence] at 01:15 24 May 2016 Recovering from Earthquakes Downloaded by [University of Defence] at 01:15 24 May 2016 ii Ú Recovering from Earthquakes Downloaded by [University of Defence] at 01:15 24 May 2016 Recovering from Earthquakes Response, Reconstruction and Impact Mitigation in India

editors Shirish B Patel Aromar Revi Downloaded by [University of Defence] at 01:15 24 May 2016

LONDON NEW YORK NEW DELHI First published 2010 by Routledge 912–915 Tolstoy House, 15–17 Tolstoy Marg, New Delhi 110 001

Simultaneously published in the UK by Routledge 2 Park Square, Milton Park, Abingdon, OX14 4RN

Routledge is an imprint of the Taylor & Francis Group, an informa business

Typeset by Star Compugraphics Private Limited D–156, Second Floor Sector 7, Noida 201 301

Printed and bound in India by Sanat Printers 312, EPIP, Kundli Sonepat 131 028, Haryana

All rights reserved. No part of this book may be reproduced or utilised in any form or by any electronic, mechanical or other means, now known or hereafter invented, including photocopying and recording, or in any information storage and retrieval system without permission in writing from the publishers. Downloaded by [University of Defence] at 01:15 24 May 2016 British Library Cataloguing-in-Publication Data A catalogue record of this book is available from the British Library

ISBN: 978-0-415-56297-3 Contents

List of Tables vii List of Maps and Figures ix List of Abbreviations and Expressions xiii Acknowledgements xix

Introduction Earthquakes in India: Response, Recovery and Impact Mitigation 1 Shirish B Patel 1. Disasters in India: Policy Overview of Vulnerability, Risks and Human Impact 14 Debarati Guha-Sapir and Philippe Hoyois 2. Th e Uttarkashi Earthquake: Management of Relief and Rehabilitation 39 Rajeev Kapoor 3. Earthquake Reconstruction in Maharashtra: Impact on Assets, Income and Equity 47 Krishna S. Vatsa 4. A Decade of Lessons from Marathwada: Earthquake Vulnerability, Politics and Participatory Housing 79 Alex Salazar and Rohit Jigyasu 5. Women Take the Lead: Turning a Crisis into an Opportunity for Development 118 Prema Gopalan 6. Compounding Disasters — First Natural, then

Downloaded by [University of Defence] at 01:15 24 May 2016 Man-made: Failed Interventions We Can Learn From 142 Sandeep Virmani 7. Planning the Reconstruction of Bhuj 159 B.R. Balachandran 8. Participative Rehabilitation: A Dilemma 204 M.N. Joglekar 9. Lessons from the Gujarat Experience: Disaster Mitigation and Management 223 V. Th iruppugazh and Sudhir Kumar vi Ú Recovering from Earthquakes

10. Disaster Management in India: Policy Interventions and Recent Initiatives 238 Anil K. Sinha, Dr Satendra and Shikha Srivastava 11. Seismic Retroﬁ tting of Existing Non-Engineered Load-Bearing Structures 255 Rajendra Desai 12. Medical Management of Earthquake Disasters 291 Brig KJ Singh 13. Trauma after Earthquakes: Mitigating the Psychosocial and Mental Eﬀ ects 316 R. Srinivasa Murthy 14. Catastrophe Risk Assessment and Management: Developing Risk Models for the Insurance Industry 337 Haresh C. Shah and Weimin Dong

Afterword by Aromar Revi 384

About the Editors 405 Notes on Contributors 407 Index 415 Downloaded by [University of Defence] at 01:15 24 May 2016 List of Tables

1.1 Distribution of Natural Disasters by Category — India and Neighbouring Countries, 1974–2003 15 1.2 Determining Characteristics of the Human Impact of Principal Natural Disasters 19 1.3 Occurrence of Natural Disasters in India and Neighbouring Countries, 1974–2003 23 1.4 Occurrence of Technological Disasters in India and Neighbouring Countries, 1974–2003 26 1.5 Comparing Disaster Types in India, Using Ratio Indicators of Human Impact, 1974–2003 27 3.1 Eff ective Project Financing Plan 50 3.2 District-wise Benefi ciary Households in Diff erent Categories of Villages 52 3.3 Scheme of Housing Entitlements under the MEERP 53 3.4 Construction of Houses in Category ‘A’ 56 3.5 Information on Diff erent Types of Old Houses 57 3.6 Provision of Bathrooms and Toilets 59 3.7 Provision of Electricity and Water Supply 59 3.8 Contributions Made by Benefi ciaries 63 3.9 Change of Agricultural Assets with Households 64 3.10 Change of Livestock Assets with Households 65 3.11 Change of Household Appliances with Households 65 3.12 Change of Vehicle Ownership with Households 66 3.13 Change in Income Levels of Households 66 3.14 Classifi cation of Households According to Caste and Type of Old and New Houses (%) 68 3.15 Classifi cation of Households According to Downloaded by [University of Defence] at 01:15 24 May 2016 Occupation and Type of Old and New Houses (%) 69 3.16 Post-Rehabilitation Status of Women 70 4.1 Sample List of Problematic Post-Disaster Relocation Projects, Prior to 1980 80

6.1 Number of Cases of Corruption in Debris Removal Registered in Gujarat 153 viii Ú Recovering from Earthquakes

7.1 Timeline for Development Plan Preparation 167 7.2 Public Consultation Exercises 170 7.3 Timeline for Preparation of Town Planning Scheme 189 8.1 Rehabilitation Game — Macro Models 206 8.2 Classifi cation of Damage to Buildings 218 9.1 Losses Resulting from the Gujarat Earthquake, 2001 224 9.2 Packages for Livelihood Restoration 231 11.1 Recent Earthquakes and their Impact 255 11.2 Building Systems and Materials of Structures Aff ected by Recent Earthquakes 262 11.3 Indicative Losses of Property (buildings) in Past Earthquakes 265 11.4 Retrofi tting Measures 268 11.5 Retrofi tting Rural Dwellings in North-East Himalayas — Unit Cost (Rs/sq. m.) and Retrofi tting Index 270 11A Weaknesses in Non-Engineered Construction 277 11B Earthquake Damage Resulting from Absence of Disaster Resistant Features 278 12.1 Surgical Operations Conducted at Military Hospital Bhuj 306 13.1 Features of PTSD 327 14.1 Top 10 US Insured Property Losses (in US$ billion) 338 14.2 Long-Term Statistics, 1950–2002 339 14.3 Underwriting Experience (1970–94, including Northridge) — Earthquake Insurance Premiums and Losses (in US$ million) 344 14.4 Loss Allocation 360

Downloaded by [University of Defence] at 01:15 24 May 2016 14.5 Event Loss Table 362 14.6 Simpliﬁ ed Event Loss Table 363 14.7 Event Loss Table for the Company’s Current Business 368 14.8 Event Loss Table for Company’s Revised Portfolio 369 14.9 Pricing and Risk Exposure Information for Some Recent Catastrophe Bonds, with Ratings by Standard & Poors/Moodys/Fitch 379 List of Maps and Figures

Maps 3.1 Districts of Maharashtra 48 4.1 Location Map of the Marathwada Earthquake 82 11.1 Traditional Building Systems of Gujarat 260

Figures 1.1 Distribution of Natural Disasters in India, 1974–2003 15 1.2 Proportion of Deaths Caused by Natural Disasters in India, 1974–2003 16 1.3 Distribution of People Aff ected by Natural Disasters in India, 1974–2003 16 1.4 Natural Disasters — Average Number of Deaths per 100,000 Inhabitants in India, 1974–2003 (fi ve-year intervals) 17 1.5 Natural Disasters — Average Number of People Aff ected per 100,000 in India, 1974–2003 (fi ve-year intervals) 17 1.6 Number of People Aff ected by Droughts in India, 1974–2003 20 1.7 Number of People Aff ected by Earthquakes, Floods and Windstorms in India, 1974–2003 21 3.1 Households’ Perception of Civic Amenities 60 3.2 Attributes of New Houses 61 3.3 Scores for Relocation Villages in Category ‘A’ 73 3.4 Recovery and Reconstruction Tree 77 4.1 Section and Plans of a Typical Wada House, Showing Downloaded by [University of Defence] at 01:15 24 May 2016 its Use and Expansion over time 83 4.2 Typical Village Plan — Babalsur Village, 1994 84 4.3 Temple at Banegaon Village 85 4.4 Reconstruction Work at Banegaon Village 91 4.5 New Babalsur Village’s Layout Plan made by Town Planners 94 4.6 Typical Housing Units Constructed by MHADA 95 4.7 Abandoned Shopping Complex at New Killari Village, 2001 97 x Ú Recovering from Earthquakes

4.8 Abandoned Neighbourhood at the New Killari Village 99 4.9 Typical Retroﬁ tting Techniques 102 4.10 Mogarga Village, 2001 105 4.11 Addition of Stone Wall and Room Constructed in the Traditional Style in New Babalsur Village, 2001 108 7.1 Institutional Framework of Reconstruction Process 163 7.2 Overall Process Flowchart 166 7.3 Bhuj Development Plan, 1976 168 7.4 Base Map Preparation — (a) Extract from Old Base Map; (b) Fresh Cadastral and Topographical Survey Output 168 7.5 Examples of Studies Carried Out for Preparation of Development Plan 169 7.6 Stakeholders in the Planning Process 170 7.7 Stakeholder Consultations 171 7.8 Proposed Road Network for Bhuj 175 7.9 Proposed Land-Use Plan for Bhuj 176 7.10 Water Bodies in Bhuj 177 7.11 Heritage Walk Brochure 178 7.12 Relocation Sites 183 7.13 Conceptual Layout of Streets Inserted in the Post-Earthquake Fabric 184 7.14 Plot Reorganisation 188 7.15 Town Planning Scheme No. 2 191 7.16 Walled City — Final Layout 195 7.17 Walled City — Destruction Caused by the Earthquake 196 7.18 Infrastructure Implementation in the Walled City 198 7.19 Bhuj — Rising from the Rubble 200 (a) A Walled City street coming back to life (b) Sarpat Gate after restoration (c) Restoration site — completed and occupied houses

Downloaded by [University of Defence] at 01:15 24 May 2016 8.1 Actors in a Rehabilitation Game 205 8.2 Extent of Consultations and the Number of Beneﬁ ciaries 208 8.3 Proposed Layout — Trombey Village 211 8.4 Layout of Trombey Village as Prepared by the Village Surveyor 212 8.5 Housing Project for Gas Victims at Bhopal 214 9.1 Institutional Set-up of GSDMA 225 9.2 Co-ordination of Stakeholders 226 List of Maps and Figures Ú xi

11.1 A Kachchh Bunga 256 11.2 A Typical Uttaranchal House 257 11.3 A North Gujarat Mud-Th atch House 257 11.4 A Typical Latur House 258 11.5 A Central Gujarat House 258 11.6 Delamination of Random Rubble Masonry 263 11.7 Racking Shear Cracks 264 11.8 Delamination of Gable Wall 264 11.9 Gable Wall Failure 265 11.10 Shake-Table Demonstration — Confi nement Reinforcement 274 11.11 Shake-Table Demonstration — Single RC Band 274 11B.1 Knee-Braces for a Timber Frame 279 11B.2 Knee-Braces at Timber Column to Beam Connections 279 11C.1 A Retrofi tted Non-Engineered Structure 280 11C.2 Cast In-Situ RC Bond Element 281 11C.3 Installation of Seismic Belt 282 11C.4 Seismic Belts 282 11C.5 Seismic Belt Installation 283 11C.6 Installing Bond Elements 283 11C.7 Casting Bond Elements 284 11C.8 Tie Rod for Seismic Belt 285 11C.9 FC-Belt Mesh Installation 285 11C.10 Gable Guy-Lines 287 11C.11 Installation of In-Plane Bracing for Roof Diaphragm 288 11C.12 Roof Diaphragm Bracings 288 14.1 Losses due to Great National Catastrophes Worldwide (a) by Year and (b) by Decade 340 14.2 Losses due to Signifi cant Natural Disasters in the US (a) by Year and (b) by Decade 341 Downloaded by [University of Defence] at 01:15 24 May 2016 14.3 Distribution of Northridge Recovery and Reconstruction Funds by Sources 342 14.4 Stakeholders in the Management of Risk 342 14.5 Major Faults in California 347 14.6 Historical Hurricane Tracks 348 14.7 Attenuation of Ground-Shaking with Distance 348 14.8 Loss Ratio of Building Classes 349 14.9 Historical Data for Wind Damage 350 xii Ú Recovering from Earthquakes

14.10 Depiction of an Insurance Portfolio 351 14.11 Depiction of the Four Types of Treaties and the Use of Facultative Reinsurance 355 14.12 Loss Diagram Displaying Sample Loss as an Area 356 14.13 Insurance Structure Used in Example 357 14.14 Dwelling Loss Experience for the 1971 San Francisco Earthquake 359 14.15 Diff erence between Expected and Distributed Loss Allocations 359 14.16 Allocation of Loss to Participating Parties — Simplifi ed Loss Diagram 361 14.17 Aggregate Loss Exceeding Probability (AEP) Curve 366 14.18 Comparison of Aggregate Loss Exceeding Probability (AEP) and Occurrence Exceeding Probability (OEP) Curves 367 14.19 Exceedance Probability Curve 370 14.20 Expected Loss and Standard Deviation for Diff erent Layers 373 14.21 Spatial Interpolation of Measured Hazard Values 374 14.22 Logic-Tree Branches for Modelling the Hazard Contribution of a Fault 376 14.23 Approximation of a Modelled Loss Curve by a Parametric Trigger Step-Function 377 Downloaded by [University of Defence] at 01:15 24 May 2016 List of Abbreviations and Expressions

ACEO Additional Chief Executive Oﬃ cer ADA Area Development Authority ASAG Ahmedabad Study Action Group ashlar coursed shaped stone masonry set in cement mortar AUDA Ahmedabad Urban Development Authority BADA Bhuj Area Development Authority BAPS Bochasanvasi Akshar Purshottam Swaminarayan basti settlement BBC Burnt-brick masonry in cement mortar BBM Burnt-brick masonry in mud mortar BHADA Bhuj Area Development Authority BIS Bureau of Indian Standards block A unit of development covering several villages. One or more blocks comprise a taluka. BMTPC Building Materials and Technology Promotion Council bhukamp Gujarati and Hindi word for earthquake CARE Cooperative for Assistance and Relief Everywhere CB Concrete Block CBI Central Bureau of Investigation CBO community-based organisation CBRN Chemical, Biological, Radiological and Nuclear CEO Chief Executive Oﬃ cer CEA Chief Executive Authority (administrative head of a development authority in Gujarat) CEPT Centre for Environmental Planning and Technology chulah wood-burning stove

Downloaded by [University of Defence] at 01:15 24 May 2016 CGI Corrugated Galvanised Iron CII Confederation of Indian Industries CLRI Central Leather Research Institute CM Chief Minister CPC Community Participation Consultants CRED Centre for Research on the Epidemiology of Disasters CRF Calamity Relief Fund crore Indian term used in counting, meaning 10,000,000 xiv Ú Recovering from Earthquakes

CRUTA Foundation for the Conservation and Research of Urban Traditional Architecture (India) CSO civil society organisation CSR Corporate Social Responsibility CSSS Centre of Studies in Social Sciences, Pune DDMA District Disaster Management Authority DFID Department for International Development (UK) DIC District Industries Corporation DILR District Inspector of Land Records district Th e next smaller unit of area of administration in a State. India has 29 States divided into 604 districts. DM District Magistrate DP Development Plan DRDA District Rural Development Agency DRM Disaster Risk Management DRR Disaster Risk Reduction DVET Directorate of Vocational Education and Training EERI Earthquake Engineering Research Institute EIA Environmental Impact Assessment EM-DAT An international emergency events database on disasters EOC Emergency Operations Centre EPC Environment Planning Collaborative, Ahmedabad EQR Earthquake Resistant FC Ferrocement (weld mesh in rich cement plaster) FEMA Federal Emergency Management Agency GCCI Gujarat Chamber of Commerce and Industry GDCR Gujarat Development Control Regulations GEC Gujarat Ecology Commission GEERP Gujarat Emergency Earthquake Rehabilitation

Downloaded by [University of Defence] at 01:15 24 May 2016 Programme GIDM Gujarat Institute of Disaster Management GI Sheets Galvanised Iron Sheets GIS Geographical Information System GoG Government of Gujarat GoI Government of India GoM Government of Maharashtra GQ Golden Quadrilateral List of Abbreviations and Expressions Ú xv

Gram Panchayat Is the local government at the village or small town level. It is the primary unit of Panchayati Raj Institutions. Gram Sevak Th e lowest level village functionary in the Panchayat system GROOTS Grassroots Organizations Operating Together in Sisterhood GSDMA Gujarat State Disaster Management Authority GSHAP Global Seismic Hazard Assessment Program GUDC Gujarat Urban Development Company Limited HPC High Power Committee IAEE International Association for Earthquake Engineering IAS Indian Administrative Service IASC Inter-Agency Standing Committee ICMR Indian Council of Medical Research IDA International Development Agency IDNDR United Nations International Decade for Natural Disaster Reduction IEC Information, Education and Communication IFRC and RCS International Federation of Red Cross and Red Crescent Societies IHBAS Institute of Human Behaviour and Allied Sciences IIT Indian Institute of Technology INGO international non-governmental organisation IRCS Indian Red Cross Society IS Indian Standards ISDR International Strategy for Disaster Reduction ISET Indian Society for Earthquake Technology ISR Institute of Seismological Research ISRO Indian Space Research Organisation

Downloaded by [University of Defence] at 01:15 24 May 2016 ITBP Indo-Tibetan Border Police J&K Jammu and Kashmir jatra see yatra jhilla see zilla Mahila Mandal A women’s organisation Mai-baap Literally ‘mother–father’, a patronising form of government MCE Mass Casualty Event xvi Ú Recovering from Earthquakes

mela a fair melava a traditional village get-together melavni Gujarati word meaning ‘matching’ or ‘ﬁ tting together’ MEERP Maharashtra Emergency Earthquake Rehabilitation Programme MIMH Maharashtra Institute for Mental Health MMK Mahila Mahiti Kendra (Women’s Information Centre) MPT Mangalore pattern tile NCCBM National Council for Cement and Building Materials NCCM National Centre for Calamity Management NCPDP National Centre for Peoples Action in Disaster Preparedness NDMA National Disaster Management Authority NDMRC National Disaster Mitigation Resource Centres NDRF National Disaster Response Force NGO non-governmental organisation NGRI National Geophysical Research Institute NIAS National Institute of Advanced Studies NICEE National Information Centre for Earthquake Engineering NID National Institute of Design NIDM National Institute of Disaster Management NIFT National Institute of Fashion Technology NIMHANS National Institute of Mental Health and Neurosciences NSS National Sample Survey OECD Organisation for Economic Co-operation and Development

Downloaded by [University of Defence] at 01:15 24 May 2016 OFDA Offi ce of US Foreign Disaster Assistance Panchayat Literally, assembly of fi ve wise and elder members chosen and elected by the village. Panchayati Raj A decentralised form of government where each village is responsible for its own aff airs. Panchayat Samiti Is the local government body at the taluka or tehsil level. Patil, Patel traditional head of the village patwari rural area offi cial List of Abbreviations and Expressions Ú xvii

PCS Property Claims Services PMU Project Management Unit PO People’s Organisation PoP Plaster of Paris PPPP Public–Private Partnership Programme PRA Participatory Rural Appraisal PRI Panchayati Raj Institutions PSI People’s Science Institute PTSD Post Traumatic Stress Disorder pucca durable, permanent (construction) R&R Repair and Reconstruction R&S Repair and Strengthening RCC Reinforced Cement Concrete RCS see IFRC and RCS RRM Random Rubble masonry RRS Repair, Reconstruction and Strengthening RRSP Reconstruction, Repair and Strengthening Programme RSS Rashtriya Swayamsevak Sangh Samvad Sahayak Village communication assistant SAARC South Asia Association of Regional Cooperation Sarpanch Elected head of the village SCG Savings and Credit Group SDMA State Disaster Management Authority SDMC SAARC Disaster Management Centre SEWA Self Employed Women’s Association SHG Self Help Group SOP Standard Operating Procedure SPARC Society for Promotion of Area Resource Centres SPIPA Sardar Patel Institute of Public Administration SSA Sarva Shiksha Abhiyan

Downloaded by [University of Defence] at 01:15 24 May 2016 SSB Seva Sauhard Bandhutv SSP Swayam Shikshan Prayog (Self-Education Experiment) taluka, taluq sub-unit of a district (zilla) TDO Taluka Development Oﬃ cer tehsil see taluka TISS Tata Institute of Social Sciences ToI Times of India TP Town Planning xviii Ú Recovering from Earthquakes

TPO Town Planning Oﬃ cer UCRC Uncoursed Rubble masonry in Cement mortar UCRM Uncoursed Rubble masonry in Mud mortar UCRL Uncoursed Rubble masonry in Lime mortar UEVRP Urban Earthquake Vulnerability Reduction Programme ULB urban local body UN-ISDR see ISDR UNDP United Nations Development Programme UNHCR United Nations High Commissioner for Refugees UNICEF United Nations Children’s Fund USAID United States Agency for International Development USAID-FIRE(D) USAID Financial Institutions Reform and Expansion — Debt UT Union Territory vyapar trade w&d wattle & daub wall WHO World Health Organization WLRP Women’s Livelihood Restoration Project yatra, jatra religious pilgrimage, often done on foot YUVA a Mumbai-based NGO zilla see district Downloaded by [University of Defence] at 01:15 24 May 2016 Acknowledgements

The feelings of inadequacy and ignorance that sparked this book emerged at a meeting of young professionals called by Anuradha Parikh- Benegal and Himanshu Burte in Mumbai a few months after the Bhuj earthquake of 2001. We had diﬃ culty in ﬁ nding a publisher, until Vijay Kulkarni, then Editor of Indian Concrete Journal, agreed to bring out the book. However, the publication did not take place for a variety of reasons best forgotten. Vijay Kulkarni continued to help in putting together the manuscript. Finally, Routledge, New Delhi, agreed to publish the book, and the editors would like to thank them for the care and professionalism they have brought to the project. Downloaded by [University of Defence] at 01:15 24 May 2016 xx Ú Recovering from Earthquakes Downloaded by [University of Defence] at 01:15 24 May 2016 Introduction Earthquakes in India Response, Recovery and Impact Mitigation Shirish B Patel

The driving force for this book has been the embarrassment of ignorance. We do not know — even as professionals working on earthquake-related issues, we are not always fully aware of — the complex issues that arise when responding to earthquakes. We hope this book will dispel some of that ignorance, not least among professionals, be they administrators or technocrats. Greater awareness of related issues, regardless of whether or not it directly impacts their work, can only help improve its quality. We know that we have not been able to cover the entire range of issues that arise when preparing for or while dealing with the aftermath of an earthquake. We also know that all the topics we have covered here will not be of interest to everyone — some of the later chapters are fairly specialised. Our reason for including them is not that everyone should be able to digest them, but that at least any interested reader should be able to get a sense of what is going on with regard to earthquakes in each area of specialisation. And if the reader happens to be an expert in a particular ﬁ eld, he should ﬁ nd the material in the chapter covering his specialisation engaging. Most of the book, however, should be of wider general interest, if only to comprehend how so many people, almost invariably acting with the best intentions, can go so dreadfully wrong. We have sought also, towards

Downloaded by [University of Defence] at 01:15 24 May 2016 the end of this Introduction, to extract some pointers and lessons for the future from experiences of the past. Th e ﬁ rst thing to understand is that earthquakes have the potential to destroy communities, not just the physical environment they inhabit. How we cope with the aftermath of an earthquake determines the extent of lasting damage that it can cause. Th at mental health can be gravely aﬀ ected, even for supposedly ‘normal’ people, is only now beginning to be recognised. We need, therefore, to think about how to change people’s 2 Ú Shirish B Patel

thinking, from dwelling on the trauma of past experiences towards the possibility of new beginnings and a much improved future. One interesting discovery has been that the devastation wrought by an earthquake also presents an unexpected opportunity: along with reconstructing houses it is possible to initiate changes in social attitudes and social structures as well. For instance, when reconstruction takes place, women have the opportunity to become leaders, with a completely transformed role compared to their earlier status in a traditional society like India. So far, admittedly, the changes that have taken place have been more or less inadvertent. As the awareness sinks in that reconstruction can bring about changes beyond the purely physical, we can surely expect the direction of the social change that will take place after future disasters to become an important area of contention. Who will lead the change and to what extent should disaster victims become owners of the reconstruction process? We know today that the more successful programmes so far have been the ones where the benefi ciaries — or victims, depending on how you look at it — have had a major say in deciding what is to be done. While this kind of approach is bound to be popular, is it the best we can achieve? Take the seemingly small example of toilets built within the house or even within the residential compound. Left to themselves, because traditionally there were no toilets and everyone went to the fi elds to defecate, it is possible that, when off ered as a reconstruction choice, local people might not opt for toilets (especially if women are not part of the decision-making process), whereas any forward-looking undertaking would defi nitely suggest otherwise. So to what extent should ‘modernisation’ — in this case, in the form of private toilets — be forced on people, and, alternatively, can this be accomplished painlessly? Th is brings us to the question of modernisation: what it means, what it does to the built environment, and how it can destroy traditional knowledge and building practices that have evolved over centuries. Too often, modernisation is understood as unthinkingly imitating the Downloaded by [University of Defence] at 01:15 24 May 2016 West. It seems to imply contempt for the traditional, with no eff ort being made to broker a deal between the two: to take what is best from the past and marry it to the best that scientifi c understanding and new inventions have to off er. With new materials like cement and steel being recommended, it is all too easy to knock down old, stone masonry structures, which are strongly built and have already withstood earthquakes, in favour of new, Reinforced Cement Concrete (RCC) buildings that have been specifi cally ‘designed for’ and may therefore survive an earthquake, but are unlikely to endure. Introduction: Earthquakes in India Ú 3

What emerges also, from one experience after another, is that local citizens’ groups are probably our best ﬁ rst responders in a time of crisis. We need, therefore, to encourage the formation of such local associations and systematically equip them with all the necessary emergency survival skills. In the context of that conclusion, what is particularly instructive (and alarming) is the government’s limited understanding of how best we should organise ourselves in order to cope with future crises.

An Overview of Disasters in India

Th e book begins with Debarati Guha-Sapir and Philippe Hoyois providing an overview of natural disasters of all kinds. Th ese are decidedly more frequent in poorer countries — not the natural events themselves, which are of course randomly distributed, but the fact of such an event turning into a disaster (which requires external intervention). A disaster occurs not only on account of the natural event but also because of the vulnerability of the aﬀ ected population, which is a consequence of the precariousness of their daily existence. Th e authors conclude with the admonition that ‘Disaster relief should be pulled out of its comfortable niche of charity and emergency care’. Th ey emphasise the importance of steering disaster management policies away from plans for immediate response and more towards reducing the vulnerability of the population. Th is would mean a huge shift in strategy, addressing, essentially, long-term changes. Th e chapters that follow describe our experience with the recent earthquakes in India, arranged more or less in chronological order, so that we can see how (if at all) our response to earthquakes has evolved over the years.

Uttarkashi (20 October 1991) Downloaded by [University of Defence] at 01:15 24 May 2016 Rajeev Kapoor describes the relief and rehabilitation work at Uttarkashi — what went wrong and why. He comes to the conclusion that there is much truth in the old adage ‘haste makes waste’. According to him, the initial survey work should have been done more carefully, even if it required more time. Rehabilitation work should have been planned properly, with reference to building skills and building materials, and thought through much more carefully in terms of the long-term consequences. It is also important to take the media, the opposition 4 Ú Shirish B Patel

political parties and the local people into conﬁ dence, and work with them in developing a reconstruction programme.

Marathwada (30 September 1993)

Krishna Vatsa’s focus is on assessing the reconstruction programme in terms of its impact on rebuilding assets and livelihoods, and the eff ect, if any, on equity and empowerment. His chapter is followed by a severe critique of the reconstruction programme by Alex Salazar and Rohit Jigyasu. Between them, these two chapters raise a variety of issues central to the discussion on recovery and reconstruction after an earthquake. Th ere is still serious diff erence of opinion regarding which is better when all the houses in a village have collapsed totally: reconstruction on a diff erent site or rebuilding in-situ? In Marathwada the possibility of rebuilding in-situ was given short shrift on account of the declaration by more than one respected Indian scientifi c organisation that the existing locations were particularly hazard-prone and that it was dangerous to build on the black cotton soil that is common in the region. Th is is far from correct, but coming as it did from worthy sources, it was suffi cient to defeat the well-known objections to relocation: that rebuilding on a diff erent site causes serious physiological, psychological and sociocultural stress. Rebuilding on a new site does of course open up the possibility of proper ‘town planning’ with wider streets and layouts that assume, however, the presence of nuclear families living separately from each other. Th is is contrary to the reality of peasant joint-family households and much more aligned with what Salazar and Jigyasu call the ‘unfi nished business of modernisation’. It is well understood that reconstruction following an earthquake presents a signifi cant opportunity for social change. Vatsa notes that the programme had a positive impact on the empowerment of women. Prema Gopalan takes this argument further as she describes the work of

Downloaded by [University of Defence] at 01:15 24 May 2016 her women’s organisation, Swayam Shikshan Prayog (SSP). She tells us how by becoming masons women not only participated in and improved the quality of reconstruction but also changed their own roles and status in society.

Kachchh (26 January 2001)

Sandeep Virmani notes that the reconstruction programme in Kachchh has been regarded as particularly successful. But his work redirects the Introduction: Earthquakes in India Ú 5

focus on what all what went wrong, and why. Although the involvement of non-governmental organisations (NGOs) is one of the important reasons for the success of the Kachchh programme, there are NGOs and there are NGOs. Th ey are not all equal and identical. Many have specifi c vested interests and are pushing their own agendas, and we need to be well aware of this. Th ere are thus important lessons here for us to learn about what not to do in the future. B.R. Balachandran, speaking of the reconstruction of Bhuj, is appalled both at the dismal state of property records and at the lack of any eff ective mechanism by which the government can interact and engage with people. He thinks undue energy was expended on seeking people’s opinion, instead of which more eff ort should have gone into the proactive building of a consensus. Like Virmani, he sees continuity in decision- making as one of the most crucial factors in helping to secure a successful outcome. For urban reconstruction, maintaining accurate and up-to-date maps is vital, as is the existence of mechanisms for interaction between the people and the decision makers.

Chamoli (29 March 1999)

M.N. Joglekar reports a case study of a slum that was successfully reconstructed after the Chamoli earthquake in Uttaranchal. His article falls outside the chronological sequence but was placed here so that the Marathwada and Kachchh programmes can be viewed in the context of the success story he describes. An important point we should note, and this has not been discussed by the author, is the question of scale. We have much to learn from examples of successful reconstruction like this one, but if these represent a ‘boutique’ situation, requiring signiﬁ cant technical inputs, we need to realise that on a larger scale this kind of intervention may not be possible. Also, it could be that we simply do not have the kind of technical resources we need to pour into the process. So, the success or Downloaded by [University of Defence] at 01:15 24 May 2016 failure of the government’s responses have to be judged in context, in relation to the scale of what is being tackled.

Government’s Perspectives

V. Th iruppugazh and Sudhir Kumar give us the Gujarat government’s view and Anil K. Sinha, Dr Satendra and Shikha Srivastava present the central government’s perspective on what is being done and what should be done with regard to disaster mitigation and management in India. 6 Ú Shirish B Patel

In Kachchh, allowing the benefi ciaries a wide range of choices and insisting that each village community take the fi nal decision about relocation were important steps in decentralising and delegating decision-making authority. It led, as we know, to widespread satisfaction with the out-come. But what is alarming is that there seems to be no awareness that programmes and policies, however carefully planned and conceived, need to be verifi ed and validated by a dispassionate interpretation of reports from the ground. Th ere appears to be no awareness that while intentions are one thing, outcomes are another. It is now widely recognised that communities themselves are the best fi rst responders at the time of a disaster. Th ey are available on the spot and can swing into action at once if they are already aware of what it is they need to do. We know that the prior existence of active local community organisations is of great value in the aftermath of a disaster. Moreover, it does not matter much what a particular organisation does. All of them can be vehicles for improving disaster preparedness and training ordinary people in the basics of disaster management. Sadly, the objectives defi ned for the Gujarat Institute for Disaster Manage- ment (GIDM), Gandhinagar, do not include anything on deepening community involvement and training people in handling disasters. We wholeheartedly welcome the central government’s all-India approach to disaster management when we are told that there has been a paradigm shift away from relief towards prevention and mitigation. But how is this shift to be achieved? A High Power Committee has been set up, whose aim is to create a disaster-free India. It has come up with a number of recommendations, including precision Geographical Infor- mation Systems (GIS) and digital maps, and state-of-the-art control rooms and emergency centres. While all this very welcome, the question is, is all this actually taking shape? Sadly, we have no means of knowing. Many activities have built-in feedback mechanisms to measure success or failure. A manufacturing company has a market share against Downloaded by [University of Defence] at 01:15 24 May 2016 which to judge its own performance, and the profi ts it makes. A professional can judge himself by the fee that clients are willing to pay him, or the number of repeat clients he has. How does a government body assess its performance (assuming it wants to)? Self-assessment is worthless and cannot be allowed. Should it report once a year to the public (as corporations do to their shareholders)? Should it report on what it has done in the past year, show how this compares to what it promised Introduction: Earthquakes in India Ú 7

a year ago, and set out what it plans to do in the coming year? Th e more abstract and rarefi ed an activity (and disaster mitigation certainly falls within that category), the more important it becomes to devise and implement ways to measure output and performance. And surely, when we are told that the Government of India has taken many initiatives, and ‘Th e most signifi cant among such initiatives is the re-designating of the National Centre for Disaster Management as the National Insti- tute of Disaster Management (NIDM)’, we have to confess our inability to appreciate the magnitude of what has been achieved. One cannot but wonder if concentrating on the structure of these high-level organisations is anything more than mere shadow-boxing. None of the concerns expressed throughout the rest of this book and none of the improvements in methods or processes that have emerged from experience seem to have found their way into offi cial thinking. Th ey say the proof of the pudding is in the eating. Perhaps we too should judge the systems in place on the basis of our most recent experience, such as the state’s response to fl oods in the river Kosi. How much of an improvement was that over responses to disasters in the past?

Medical Interventions

It is a relief to turn from such lofty policy issues to more mundane matters. Th e last four articles are increasingly technical, but should still be of interest even to a lay reader. Th e fi rst of these, by Brig KJ Singh, is about medical relief in an immediate post-disaster situation. Th e Bhuj experience revealed some unexpected aspects of disaster management, like the importance of managing information, crowd control and media management. Th ese are normally not thought of as problems that may come up at the time of a disaster, but clearly they are aspects of such a situation for which disaster management personnel need to be trained and prepared. Downloaded by [University of Defence] at 01:15 24 May 2016 R. Srinivasa Murthy discusses the growing awareness that is visible today of the mental health consequences of natural disasters. Most people are aff ected by a disaster in one way or another, even if at diff erent times, and they can all do with help in understanding that the changes they experience — in their feelings, their bodily functions (like sleep), even behaviour — are normal. Th ey need to know that their’s is a normal response to an abnormal experience and they need to be told 8 Ú Shirish B Patel

what corrective actions they can take to help them recover. Establish- ing a routine and taking part in social and cultural activities also help in coping with the psychosocial distress experienced by an individual post-disaster. It is important to keep the family together, not separate the men from the women and children, and, in cases of relocation, to try and keep the larger community together as well. While mental health professionals can help, the most important support is best provided by the community itself, and to do this it would be ideal to train one person from within the community for every 25–30 families. In fact, it is rather interesting to trace the history of disasters in India, and see how mental health professionals have increasingly come to understand how best they can prepare people for a disaster, and then, when it happens, within the framework of their limited numbers, how the same professionals can help people cope with the disaster.

Non-Engineered Construction

Rajendra Desai argues that the only way to cope with the problem of such a large number of our existing buildings being ‘non-engineered’ is to ‘retrofi t’ them. Th e trouble is that people have to fi rst be convinced that retrofi tting actually works, and this means convincing not only the owners but also the artisans who will carry out the work. One of the best ways to do this is by using shake-tables to demonstrate how an earthquake brings about a building’s collapse. Th e author’s work in this regard has been pioneering. On a large enough, fl at steel sheet, mounted on rollers, you construct two houses, side-by-side, built at half-scale, i.e., half-size models built using the same materials and construction technology as the real building. One model is built using ‘normal’ traditional construction technology, while the other incorporates those additional features that are demanded of earthquake-resistant construction. Th e table is struck a heavy horizontal blow, which causes

Downloaded by [University of Defence] at 01:15 24 May 2016 it to shake. Th e model houses on top of the table shake with it. The blows are repeated and one can see the increasing damage that the shaking of the ‘ground’ below causes to the model houses. Finally, the house built in the traditional way collapses, whereas the one that incorporates earthquake-resistant features, however damaged it may be, does not collapse. A demonstration of this kind is both vivid and dramatic. Also, because the model houses have been built by local masons and the show takes place at a large public gathering, the demonstration has the impact. Introduction: Earthquakes in India Ú 9

Insurance

Th e article by Haresh Shah and Weimin Dong is also rather technical, located as it is at the frontier of its ﬁ eld of specialisation, but we hope it will be of some general interest to everyone; and in particular, that it will encourage those who work in ﬁ nancial markets to rethink their role in catastrophe risk management. Th e authors note that after the 1994 Northridge earthquake in the US, two-thirds of the loss was covered by insurance companies. In contrast, after the Kobe earthquake in Japan less than 5 per cent of the loss was covered by insurance. Despite such variability, the fact remains that in developed countries insurance has carried a larger and larger burden of natural disaster losses. Th is is not true for developing countries like India, where the need for such a cover is in fact greater. Th e authors’ concern is with how we should spread the cost burden of recovery far and wide, and distribute it between property owners, insurers, reinsurers, the capital market and the tax-paying public. Th ey talk of catastrophe bonds, instruments in which the risk of loss is low, say 1 in 250 (0.4 per cent) — implying a return period for the catastrophic event of once in 250 years — but when the loss does occur it is total. Regardless, risk sharing can be structured in such a way that it becomes more widely dispersed.

Conclusions

It might perhaps be instructive to end this introduction by reﬂ ecting on what we have learnt about disaster management and what we think should be the way forward. First, with regard to immediate response, it should be swift and uncluttered. We should immediately put in command a local person who has experience in dealing with such a situation. Th e restoration of transport and communications links is top priority. With regard to the

Downloaded by [University of Defence] at 01:15 24 May 2016 latter, a huge boon today is the widespread use of mobile phones. Th e last thing we need in the area is VIP visits, which only divert local resources away from what should be their primary focus, i.e., rescue and immediate relief. Th e media too must be managed, and this must be recognised as a central and important task. Unwanted supplies, particularly medical supplies, must not be allowed to add to the confusion. What is required is the standard set of medical supplies for this type of emergency. Doctors and nurses are needed urgently because the heaviest medical workload is in the hours immediately following the disaster. Meanwhile, rescue 10 Ú Shirish B Patel

is the most diffi cult task, particularly since equipment and machines are required to cut and reach through debris. Not only do they need be moved to the location, which is time-consuming and cumbersome, but operators who know how to use the equipment must also be available. And the earlier we can get mental health professionals on-site the better. What we must also remember is that relief workers themselves will be under considerable stress, and need to be properly housed, fed and rested, and perhaps even counselled psychologically. All this calls for a high level of pre-disaster preparedness so as to be able to respond at very short notice. Water and power supplies must be quickly restored for which skilled workers in these fi elds are needed at once. Th e distribution of food supplies and construction of temporary shelters are also urgent. Planning in advance about how these activities will be conducted will defi nitely speed up response. We need to decide well before an actual event what each person will be called upon to do, what supplies will be needed, who will be sent to the site to help with the work, and who will be in charge. Reconstruction, as we have seen, is not an urgent requirement. Damage assessment therefore must be carefully and reliably carried out, once and for all, so that the base document prepared is universally accepted and disputes are minimised. Compensation packages need to be simple and fair. Retrofi tting and repairs, or reconstruction in-situ are always to be preferred to relocation. But there is an opportunity here which should not be missed, and that is to introduce modern systems of water supply and sanitation, and where possible widening the road. Professional advice on village layouts is also welcome. What is needed above all is the rebuilding of self-confi dence and self-reliance, and this is best done by involving people as much as possible in reconstruction, especially in the decision-making process. In the case of towns, the removal and disposal of debris is a major task that calls for careful thought. For reconstruction, if our pre-disaster

Downloaded by [University of Defence] at 01:15 24 May 2016 mitigation eﬀ orts have proceeded as planned, there will be maps and other records that can establish ownership rights as they existed prior to the devastation. Th ese must be made the basis of a fair division of the redevelopment, which may or may not have boundary lines exactly matching the earlier arrangement. Th is will depend on what opportunities present themselves for redevelopment, with better common amenities and improved public services. Such opportunities will need to be discussed with the property owners. Hopefully, if improved layouts are possible, there could well be a consensus on reorganised boundaries. Introduction: Earthquakes in India Ú 11

Th ere is one urgent aspect of reconstruction, whether in towns or in rural areas, and that is the rebuilding of people’s economic activities. Th ey need to start earning as soon as possible and feel economically independent. It is also possible in the process to seize the opportunity to train women in skills hitherto reserved for men. Th e skills of local masons and craftsmen should be upgraded so that they know how to carry out retrofi tting work and understand why it is important to do so. At what stage shake-table demonstrations need to be introduced also requires careful thought: we want to convince people about the improved technology without making them relive the trauma of what they have been through. Showing video fi lms of such shake-table demonstrations could also serve the purpose, even if less forcefully. Re-establishing community activities and regular routines helps restore people’s mental and psychological equilibrium. Th ese aspects of recovery also matter, and addressing people’s mental and emotional needs should be acknowledged as a key constituent of the rebuilding process. With recovery complete, what about the mitigation of future disasters? How do we reduce people’s vulnerability and how do we prepare them for the next disaster? Wherever we are in the country we know that something major and unexpected can happen at any time. How can we ensure a less disastrous outcome? First, we know that the people who can help most quickly and eff ectively are already available on the spot: they belong to the aff ected community. So it is local community organisations, including women’s organisations, that need to be encouraged and strengthened. It probably does not much matter what they do, just that they meet regularly and work together towards a common goal. Th ese are the groups that the government needs to work with in order to train the local people. Th ese could be fi rst-aid skills, for example. They could also form savings groups and work towards taking house improvement loans. Th e point is to develop self-reliance, and self-confi dence our rstfi

Downloaded by [University of Defence] at 01:15 24 May 2016 bulwarks against calamity. We have to persuade people to think of retroﬁ tting their buildings. Th ere is nothing like a shake-table demonstration to convince them of the need for this. Convincing masons and local artisans of the value of this will be easier if we approach them with an understanding and appreciation of their traditional technologies. Th is is the hard part. An eﬀ ort needs to be made in our engineering colleges, where students learn about traditional building technologies, to understand what is wrong with them and what is right about them. From this we can move on to 12 Ú Shirish B Patel

strengthening the methods, principles and practices of construction. Th e same training can then be extended into the countryside, to reach masons and other building craftsmen. At present, unfortunately, retrofi tting is a word that has come to be associated in professionals’ minds with horrifying complexity and absurd costs. Th is is wrong. Retrofi tting does not mean — and should not be taken to mean — bringing an old structure to a state of full compliance with current codes of practice that govern the design of new buildings. Such a strict requirement may be well-nigh impossible to achieve. Instead, we should understand retrofi tting as a process where we deal with the weaknesses of the old structure perhaps one at a time, as funds permit, to make it stronger and stronger, evem if we may never be able to make it as strong as the current codes of construction practice demand from a new structure. Of course, the important the old building, more thorough the retrofi thing should be. Th e focus of training in engineering needs to shift, so that we are not only looking at new structures where we have great freedom of choice but also looking for weaknesses in old structures and learning how best to address these within a given framework of constraints. Calamity insurance is not something which we can expect our economically vulnerable populations to subscribe. Th eir means are stretched thinly enough as it is. Th ey also do not have the training and skills needed to deal with insurers, either before or after the event. But what might be possible is insuring a block of vulnerable people around their own location, undertaken as part of corporate social responsibility (CSR). Th e insurance can be handled by knowledgeable people, and should a disaster take place funds will be locally available to support reconstruction, independent of outside support or aid packages that may be subsequently announced. And, fi nally, what of the government’s role in all this? First, we need some mechanism by which we can objectively assess whether disaster Downloaded by [University of Defence] at 01:15 24 May 2016 response preparedness is really increasing, and if so to what extent. Before embarking on courses of action, as with any enterprise, we fi rst need to set out how success will be measured. Th is may seem diffi cult to do but it is crucial. Without this we have no means of knowing whether we are getting anywhere. Intentions, however noble and clearly expressed, are not to be confused with results. If the recent experience of the Kosi fl oods is any indication, our disaster response mechanism does not seem to have been much of an improvement over past performances, Introduction: Earthquakes in India Ú 13

although admittedly this is impressionistic and hard data is diﬃ cult to come by. Second, the government often seems to us on the outside to be still operating in the patronage mode, i.e., we continue to be recipients of its benevolence, not partners in a joint enterprise. Ours is still a parent– child relationship, not one between comrades-in-arms. Now that would be a real paradigm shift. Nevertheless, we must accept that the leadership with respect to disaster risk mitigation has to come from the government. Th ere is no escape from this. Th e issues we are dealing with revolve around policy-making and the shaping of attitudes. We depend on the government for leadership in this, but perhaps more would be achieved if the rest of us could also all be made partners in the undertaking. Downloaded by [University of Defence] at 01:15 24 May 2016 1 Disasters in India Policy Overview of Vulnerability, Risks and Human Impact Debarati Guha-Sapir and Philippe Hoyois

South Asia is one of the most vulnerable regions of the world with respect to natural disasters and man-made catastrophes. High population densities, rampant and unplanned urbanisation, widespread poverty, and malnutrition are all further aggravating factors that make the populations even more vulnerable to the disasters that befall them on a regular basis. All of these characteristics apply fully to India, which, with 305 natural disasters reported for the last 30 years, qualifi es as one of the countries most at risk of devastating natural catastrophes. For the Centre for Research on the Epidemiology of Disasters (CRED), ‘a disaster is a situation or event which overwhelms local capacity, necessitating a request to national or international level for external assistance’. In co-operation with the Offi ce of US Foreign Disaster Assistance (OFDA), CRED maintains an international Emergency Events Database (EM-DAT) on disasters. Criteria for a disaster to be registered in the database are as follows: at least 10 deaths (persons confi rmed as dead and/or persons missing and presumed dead) or 100 aff ected (people that have been injured, left homeless or requiring immediate assistance during a period of emergency, i.e., requiring basic survival requirements such as food, water, shelter, sanitation, and immediate medical assistance after a disaster) or request for assistance at a national or international level. Downloaded by [University of Defence] at 01:15 24 May 2016 Certain kinds of natural catastrophes are more common, both in numbers of events and frequency, as shown by Table 1.1, which sets out the profi le in fi ve countries of the region by broad disaster groups as recorded in EM-DAT. Floods and cyclones are the natural disasters that occur most frequently in this region. Although much rarer, earthquakes and droughts also have a signifi cant impact on the population. Heat waves and cold Disasters in India: Policy Overview 15

Table 1.1: Distribution of Natural Disasters by Category — India and Neighbouring Countries, 1974-2003

Bangladesh Bhutan India Nepal Pakistan Sri Lanka Total Droughts/famines 4 0 13 1 4 10 32 Earthquakes 6 0 16 2 16 0 40 Extreme temperatures 14 0 31 3 12 0 60 Floods 53 2 115 21 34 33 258 Slides 0 0 32 11 10 3 56 Windstorms 96 1 94 6 16 3 216 Others* 1 14 2 1 0 9 Total natural disasters 174 4 305 46 93 49 671 Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be Note: *Others includes insect infestations, waves/surges and wild fires.

waves cause loss of lives and marginalisation of large population groups by affecting subsistence agriculture. Other catastrophic events, such as landslides, avalanches, snow, and fires threaten smaller proportions of the populated world. More precisely, since 1974, India has suffered over 100 natural disasters per decade, with floods accounting for nearly 40 per cent of all disasters and storms for 30 per cent. For the entire period, in this country, natural disasters have collectively amounted to a significant number of lives lost (more than 121,000), people affected (more than

Figure 1.1: Distribution of Natural Disasters in India, 1974-2003 Others Slides 1% 10% Extreme temperatures Droughts 10% ~ - 4% Earthquakes 5% Downloaded by [University of Defence] at 01:15 24 May 2016

Windstorms 31%

Floods 39%

Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be 16 Guha-Sapir and Hoyois

Figure 1.2: Proportion of Deaths Caused by Natural Disasters in India, 1974-2003 Others 4% Extreme temperatures 9% Earthquakes 27%

Windstorm: 31%

Floods 29% Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be

Figure 1.3: Distribution of People Affected by Natural Disasters in India, 1974-2003

Others Windstorms „ 0% 4%

Floods 34% Droughts 60% Downloaded by [University of Defence] at 01:15 24 May 2016

Earthquakes 2%

Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be Disasters in India: Policy Overview 17

Figure 1.4: Natural Disasters — Average Number of Deaths per 100,000 Inhabitants in India, 1974-2003 (five-year intervals) 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 74-78 79-83 84-88 89-93 94-98 99-03 Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be

Figure 1.5: Natural Disasters — Average Number of People Affected per 100,000 in India, 1974-2003 (five-year intervals) 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 0 74-78 79-83 84-88 89-93 94-98 99-03 Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be

1,832 billion) and also major economic losses (more than Rs 1,961 billion or US$ 43 billion [in 2003 terms].1 Downloaded by [University of Defence] at 01:15 24 May 2016

1 Data on deaths, people affected and economic losses are unequally reported according to the type of disaster and the estimated impact. Considering the total number of disasters which occurred in India during the period 1974-2003, deaths are reported for 88 per cent of the disasters, people affected for 56 per cent and economic damages for only 27 per cent. For earthquakes, the reporting is somewhat better: 87.5, 68.8 and 62.5 per cent respectively. 18 Ú Guha-Sapir and Hoyois

Characteristics and Human Impact of Natural Disasters

Th e destructive agents in the six main classes of natural disasters mentioned above are wind, water (a lack or excess thereof) and tectonic forces. All these cause structural damage and also have a signifi cant mortality and morbidity impact on human and animal populations. Th e disaster cycle can be diff erentiated into fi ve main phases, extending from one disaster to the next. Th ese are:

(i) Th e warning phase, indicating the possible occurrence of a catastrophe and the threat period during which the disaster is impending. (ii) Th e impact phase, when the disaster strikes. (iii) Th e emergency phase, when rescue, treatment and salvage activities commence. (iv) Th e rehabilitation phase, when essential services are provided on a temporary basis. (v) Th e reconstruction phase, when a permanent return to normalcy is achieved.

Th e human impact diff ers between these phases and is also a function of the prevailing health and socio-economic conditions of the aff ected community. As a result of this, global statistics on disasters seem to indicate signifi cantly higher frequencies of natural disasters, of larger scale with greater human impact (measured by number of deaths and people aff ected) in the low and lower-middle income countries than upper-middle and high income countries.2 The phenomenon tends also to arise within countries. In other words, in poor countries such as India, the relatively poorer sections of the disaster-aff ected communities suff er greater damage than their wealthier neighbours. A disaster may, therefore, be defi ned by the vulnerability of the popu-

Downloaded by [University of Defence] at 01:15 24 May 2016 lation to a natural event and not only by the mere fact of its occurrence. Th e four major disaster types, as measured by their impact on a population, diﬀ er on relative scales of lethality, predictability, onset time, and impact scope. Th is ranking provides some guidance towards understanding the variation in human impact noted among disaster events across time and space. Table 1.2 displays the four scales with the relative positions of the disaster types.

2 As deﬁ ned by the World Bank. Disasters in India: Policy Overview Ú 19

Table 1.2: Determining Characteristics of the Human Impact of Principal Natural Disasters

Predictability Direct lethality Scope Onset Earthquakes Low Very high Low Sudden Cyclones Medium High High Medium Flash ﬂ oods/ﬂ oods Medium/high High/medium Low/high Sudden/slow Drought/famine High Low High Slow Source: Author.

Although a drought-related famine is a very special class of disasters, it nevertheless falls within the general paradigm which characterises natural disasters. Famines are disasters of high predictability. Th ey usually develop over months and food shortages make themselves felt slowly over time. Th e Great Bengal Famine of 1941 was appealing in its human mortality (nearly 3 million are reported to have died as a direct result of malnutrition), as was the famine in Bangladesh in the early 1970s. Both events gathered force over time and in the face of non-action by the state and other public authorities. Two of the largest famines since World War II have taken place in countries with normal or more than normal food production during the famine year. Amartya Sen (1981) has observed that Ethiopia was a net exporter of food in 1973, and both Bangladesh and Bengal produced more grain in 1974 and 1941 respectively than in the preceding years. Drought or any other natural disaster which aﬀ ects harvest, as recently in Malawi (Devereux 2002), sometimes serve as a trigger mechanism for a famine, but the disaster remains a largely poverty-related catastrophe with a very weak causal relationship to food supply. Famines provide an excellent illustration of the fact that the knowledge of impending disaster does not imply that a community can or will take responsive action. Unlike an earthquake, this is a phenomenon that can be averted if adequate preparation exists. Today in India, due to close monitoring of markets, rationing systems

Downloaded by [University of Defence] at 01:15 24 May 2016 and effi cient transport networks, famines have been avoided in most cases or remain at a level of food shortages only. Similarly, the impact of other disasters is a function of the physical and economic resistance of the population. Floods can diff er in their onset characteristics. Th ey can develop slowly and be fairly predictable, such as the annual fl oods in the plains of the Ganga or Padma rivers in India and Bangladesh, but nevertheless regularly cause a number of deaths and a certain amount of damage. Violent fl oods have caused large losses of human lives both in Bangladesh and in India. Acute and catastrophic Downloaded by [University of Defence] at 01:15 24 May 2016

Figure 1.6: Number of People (in million) Affected by Droughts in India, 1974-2003 350,000

300,000

250,000

200,000

150,000

100,000

50,000

0 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be Downloaded by [University of Defence] at 01:15 24 May 2016

Figure 1.7: Number of People (in million) Affected by Earthquakes, Floods and Windstorms in India, 1974-2003 128 50

0 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03

Earthquakes Floods Windstorms

Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL-Brussels, Belgium, http://www.em-dat.be 22 Ú Guha-Sapir and Hoyois

fl oods are usually generated by cyclones or tidal waves; examples are the ones in Tamil Nadu (1994), Philippines (1984) and Bangladesh (1985). Large, slow onset fl oods may cause lower mortality than other disasters, but the scope of damage is generally wider and more pervasive. Large crop areas are lost giving rise to serious long-term implications for subsistence farming communities. At the other end of the spectrum, earthquakes tend to be the least predictable of disasters, striking with little warning (see discussion between Geller et al. 1997a, 1997b; Wyss 1997; Wyss, Aceves and Park 1997; US Geological Survey [USGS] 1999; Wilson, n.d.). Japan is one of the few high-risk countries that have an eff ective warning and evacuation system, as well as excellent community education programmes (Nakano et al. 1974). Th e earthquake in Niigata, Honshu (16 June 1964) registered 7.7 on the Richter scale. Although 20,000 houses were destroyed, only 13 people were killed and 315 injured. Due to the quality of its preparatory programmes, Japan suff ers very limited mortality despite the high number of seismic shocks it registers. However, the Kobe earthquake (17 January 1995) showed that mitigation policies have to be constantly improved and carried on (Wieland 2001). In terms of lethality, earthquakes present the greatest risk of death to those aff ected. Onset delay is the shortest in this type of disaster, which is related, to a certain extent, to its low predictability. However, the relationship between the severity of an earthquake and its con- sequent mortality is far from linear, being dependent on many diff erent factors. Data on the human impact of disasters in India when analysed as ratios illustrates the above discussion and are addressed later (Table 1.4).

Natural and Man-Made Disasters: Trends and Differentials

Downloaded by [University of Defence] at 01:15 24 May 2016 Table 1.3 displays comparative time trend data from India, its fi ve bordering countries and for the entire region over the last 30 years. Although, EM-DAT data shows a marked increase in disasters over time for the region and India, the data should be interpreted with caution. First of all, reporting has improved enormously over the last few years, mainly due to widespread and easy telecommunications and mass media, aff ecting statistics. Secondly, disaster databases, such as EM-DAT, have moved from passive data collection approaches to active data collection, ranking up events that otherwise would not have been registered. Disasters in India: Policy Overview Ú 23 (Continued) India Sri Lanka Bhutan Pakistan 00071113 Nepal Bangladesh 000000041018 4001001022 10 100004622 110484000 012228000 597881861314 335244210 0512410111 011010000 310000445 024000664 1120 22 30 20 4 6001362731 0 0 2 24 47 44 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 Downloaded by [University of Defence] at 01:15 24 May 2016 Table 1.3: Occurrence of Natural Disasters in India and Neighbouring Countries, 1974–2003 Droughts/famines Floods Slides Windstorms Others* Total Natural Disasters 34 59Earthquakes Extreme temperatures 8 100481107117 Floods Slides Windstorms Others* Total Natural Disasters 10 20 16 18 27 48 15 17 17 Droughts/famines Earthquakes Extreme temperatures 24 Ú Guha-Sapir and Hoyois India and Neighbouring Countries res. 1974–83 1984–93 1994–2003 Downloaded by [University of Defence] at 01:15 24 May 2016 e OFDA/CRED International Disaster Database, UCL–Brussels, Belgium, http://www.em-dat.be (Continued) Note : *Others includes, insect infestations, waves/surges and wild fi Droughts/faminesEarthquakesExtreme temperaturesFloodsSlidesWindstorms 14Others* 9Total Natural Disasters 11Th Source : EM-DAT: 158 54 59 11 0 7 18 17 230 99 67 18 4 11 33 12 283 105 90 27 5 Disasters in India: Policy Overview Ú 25

This being said, certain disasters, such as floods and slides, are probably really increasing due to urbanisation, deforestation and other man-made actions aggravating environmental destruction. Th e trend from the EM-DAT statistics clearly shows a signifi cant increase in fl oods/ slides, extreme temperatures and wind-related disasters, especially since 1980. Th e disaster of high priority in India is fl oods, not only due to their frequency but also because they aff ect large and densely populated regions of the country such as the Gangetic plains in the east. Man-made disasters have also taken on major importance in the last two decades, as shown in Table 1.4. Accidents of an urban and/or industrial nature have increased substantially, pointing to the critical need to address issues related to regulation, conformity to standards and urban planning.

Immediate Impact of Disasters

On a global level, the mortality generated by natural disasters shows some interesting tendencies, creating the beginnings of an analytical framework within which specifi c impacts may be systematically analysed for robust indicators, effi cient needs assessment or preparedness and rehabilitation planning. Th e mortality from disasters is a function of risk, development and coping or adjustment capacity (preparedness). Table 1.5 presents some simple ratios to help rationalise comparisons between the impact of diff erent disasters. Earthquakes and other acute events such as windstorms have low aff ected/dead ratios, indicating that victims tend to either survive without long-term distress, or die. By contrast, in fl oods and droughts the aff ected/dead ratio is large. Th ese disasters have less immediate tragic outcomes, but by impacting large areas of land and destroying means of livelihood, they leave many people in a precarious situation, sometimes for long periods. Downloaded by [University of Defence] at 01:15 24 May 2016 Th e ratios clearly indicate that approaches to improving response and preparedness are fundamentally diff erent for various kinds of disasters. Th ose that have a sharp eff ect, killing large numbers in short periods, require stronger emergency responses and specialised intervention immediately after the event (Belba 1996), whereas those that do not result in a large number of deaths, and are therefore less spectacular, systematically marginalise signifi cant parts of the population in the aff ected area and require longer-term solutions. 26 Ú Guha-Sapir and Hoyois India Sri Lanka Bhutan Pakistan India and Neighbouring Countries Nepal Bangladesh 000169001 010147010 015000112329 03400053132 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 1974–83 1984–93 1994–2003 Downloaded by [University of Defence] at 01:15 24 May 2016 Table 1.4: Occurrence of Technological Disasters in India and Neighbouring Countries, 1974–2003 Transport accidentsMiscellaneous accidents Total Technological Disasters 0 0 11 10 15 15 3 1 46 36 72 56 1 1 7 6 11 10 Industrial accidents Industrial accidentsTransport accidentsMiscellaneous accidentsTotal Technological Disasters e OFDA/CRED International Disaster Database, UCL–Brussels, Belgium, http://www.em-dat.be Source : EM-DAT: Th 30 12 12 6 335 30 265 40 409 44 322 43 Transport accidentsMiscellaneous accidents Total Technological Disasters 4 4 65 61 5 900022206252 5 00006152191 Industrial accidents Disasters in India: Policy Overview Ú 27

Table 1.5: Comparing Disaster Types in India, Using Ratio Indicators of Human Impact, 1974–2003

Deaths People aff ected Aff ected Occurrence (mean (mean number per killed (total number per disaster)3 (mean number number) per disaster)1 (‘000) per disaster)5 Droughts/Famines 13 2322 136,397 i.d. Earthquakes 16 2,297 3,358 6,159 Extreme temperatures 31 345 d.n.a. d.n.a. Floods 115 333 7,963 24,771 Slides 32 91 2734 2,3214 Windstorms 94 481 1,202 10,123 Source: EM-DAT: The OFDA/CRED International Disaster Database, UCL–Brussels, Belgium, http://www.em-dat.be Note: i.d.: insuffi cient data d.n.a.: data not available 1 : based on reporting rates of 85–100 per cent for number of deaths 2 : data available for only four droughts 3 : based on reporting rates of 61–68 per cent for number of people aff ected 4 : data available for 14 slides 5 : based on reporting rates of 56–65 per cent for both number of deaths and of people aff ected.

Immediate mortality and structural destruction, highest in earthquakes, have been studied with regard to the risk factors that explain these rates. Population density (Lechat 1984), structural quality (Glass et al. 1977; Wieland 2001), time of strike (De Bruycker et al. 1983, 1985), intensity and characteristics of seismic activity (Alexander 1985, 1996; Earthquake Engineering Research Institute [EERI], 2003), absence of mitigation programmes (Federal Emergency Management Agency [FEMA] 2002) and timing of rescue operations (de Ville de Goyet et al. 1976; Guha-Sapir and Carballo 1999) seem to be the main risk factors, but they fail to explain adequately the high mortality in earthquakes.

Downloaded by [University of Defence] at 01:15 24 May 2016 Further research with better data and case-by-case analysis is urgently required to clarify the social and behavioural risks of mortality in these disasters. Th ese factors are important to understand because they lend themselves to low-cost policy initiatives. Droughts and famines have the highest rates for the number of people aff ected per disaster, accounting for more than 100 million per event. Many of the victims are counted more than once since these are multi- year events and EM-DAT registers the event according to the calendar year, if it is still continuing. Th e fi gures refl ect the ‘caseload’ of victims at 28 Ú Guha-Sapir and Hoyois

a given period and therefore serve as a statistically adequate indicator of the comparative weight vis-à-vis the other disasters. Typically, farmers with no access to water or with only inadequate irrigation, and those living on their own produce or using fragile varieties susceptible to excess heat or moisture, are the bulk of the victims of droughts and fl oods. Th e ratios presented here indicate that disaster preparedness and prevention strategies should be different for such phenomena as earthquakes, slides and windstorms, which cause high mortality, have short lifespans and lower levels of aff ected, compared to fl oods and droughts which have a lower direct mortality but have a major long- term impact. It is becoming increasingly clear that in poorer or less developed rural economies, such as in South Asia, local conditions play a much bigger role than expected in determining the human impact of disasters. Th e toll taken by a disaster is generally correlated to the economic standing of a community, which determines its resistance and recuperative powers. Mortality rates, controlling for the number of disaster events, are substantially higher in poor countries than in the richer ones. Th is important infl uence of the prevailing socio-economic conditions on the impact of a disaster has been convincingly argued by Cuny (1983) and Shah (1985), and more recently in the UNDP report on the reduction of disaster risks (2004). For predictive and needs assessment purposes then, these factors and the health conditions prevalent in the aff ected community could be better determinants of the human impact of an event than its physical characteristics.

Disease and Injuries after Disasters

Th e data on morbidity (injuries and disease) after a disaster are remark- able for their absence or incomparability, despite being the most important impact of a disaster on human populations. Th e deﬁ nition Downloaded by [University of Defence] at 01:15 24 May 2016 of injury, when registered, is largely unstated, and the reporting of disease incomplete. Th is has resulted in a variety of observations, some anecdotal, some systematic, but nearly all fragmentary. Th ere is clearly an urgent need for standardised reporting of injuries and the cause of death (preferably using a standard format such as the International Classification of Disease). Without such standardisation, disaster planning and management remains an ad hoc activity and analyses can only be undertaken superﬁ cially. Disasters in India: Policy Overview Ú 29

Injury Proﬁ les of Natural Disasters

Th ere are some recorded fi gures available on injuries sustained in earthquakes where authorities registered and published morbidity data. It is uncertain what qualifi ed as injury and, more importantly, the bias introduced by those who were not treated in hospital. Th ere are even fewer data on non-traumatic morbidity. Classifi cation biases and general incomparability pose important problems for those undertaking analyses for programme or policy purposes. In the case of earthquakes, the disaster type most prone to causing traumatic injury, fractures constitute the most prominent type of injury. Fractures of the extremities of the body seem to be signifi cantly more frequent than any other sort, based on data gathered from earthquakes in other parts of the world. In certain earthquakes, the few instances where injuries were classifi ed according to type, fractures of the limbs accounted for over 60 per cent of all injuries (Shoaf et al. 1998; Kazzi et al. 2000). Most injuries, be they lacerations suff ered in cyclones or fractures in earthquakes, tend to occur during the catastrophe itself or in the very immediate post-impact phase. Clearly, in both earthquakes and cyclones, the structural quality of housing is a major determining factor of the extent and type of injury, which is in eff ect a proxy variable for the socio-economic level of the community or household (Haas et al. 1977; Armenian et al. 1997).

Disease Proﬁ les of Natural Disasters

Major epidemics are fairly rare events that may follow natural disasters (de Ville de Goyet and Lechat 1976; Seaman et al. 1984; Alexander 1996; Morgan 2004). However, in South Asia, where sanitation is poor and endemicity of many communicable diseases is normally high, disease rates do increase dramatically.

Downloaded by [University of Defence] at 01:15 24 May 2016 Th ere has been an unprecedented increase in the number of shanty towns and slum areas in the burgeoning metropolises of several high-risk South Asian countries, making them especially fragile. Th e enormous population density in these slums has contributed to the increase in the number of victims in recent years. In the slum areas of Dhaka and Calcutta, for example, fl oods frequently cause the canals-cum-latrines to overfl ow into the living quarters of the slum-dwellers. Signifi cant increases have been noted in the incidence of typhoid, cholera, skin and gastro-intestinal endemic diseases, which are frequent consequences 30 Ú Guha-Sapir and Hoyois

of fl ooding (Siddique et al. 1991; Mondal et al. 2001; Biswas et al. 1999; Wade et al. 2004), the risk of malaria may be exacerbated as well in such conditions, in the period following (Kondo et al. 2002). Th ese diseases also have a serious impact on infant and child mortality and morbidity, especially when children were already malnourished before the disaster struck. Cholera and malaria epidemics or even signifi cant outbreaks have also been documented as short- and mid-term consequences of fl oods and droughts, resulting from climatic abnormalities (DME 1999; Gagnon et al. 2002; Kovats et al. 2003) and of earthquakes and possible subsequent fl oods (Saenz et al. 1995). As mentioned above, disaster-related disease outbreak patterns diff er according to the socio-economic status of the aff ected groups. Th e incidence rates during a cholera epidemic in Bangladesh were correlated to education and income. Th e poorer sections of the aff ected region used canal water for drinking and washing purposes and presented a generally lower resistance to infections. Th e incidence rate of the disease per 1,000 families with no schooling was 16.3 whereas it was 8.2 among families with at least one high school graduate (Levine et al. 1976). In famines, the synergy between malnutrition and infectious diseases gives it an altogether diff erent dimension compared to other disasters. In a famine, communicable and nutritional defi ciency-related diseases are, in fact, the principal manifestations of the event. Finally, short and long-term psychological consequences of natural disasters; the role of social support as a protective mechanism against adverse psychological outcomes has now been documented (Tierney 2000; Armenian et al. 2002; Woersching and Snyder 2004).

Long-term Impact of Natural Disasters

Th e long-term impact of disasters, possibly the most pervasive and

Downloaded by [University of Defence] at 01:15 24 May 2016 destructive phase, expresses itself variously. In India, where much of the population is involved in subsistence farming, disaster-induced death and the disability of an earning member of a family implies a lifetime’s loss of revenue and possible destitution. Th e informal sector is an important source of revenue for a large proportion of the population and social security systems here are weak. In these conditions such losses can be fatal for the surviving members of the family. In ﬂ oods, typically the coastal ﬂ oods experienced by Tamil Nadu or Orissa, salt-water contamination of subsistence and marginal land Disasters in India: Policy Overview Ú 31

implies the loss of not one but several harvests. For nutritionally and economically fragile populations, this results in a rise in mortality, especially among children — a secondary eff ect of the disaster. Similarly, the death of the breeding stock of herdsmen and the loss of capital or tools of trade due to water damage, cyclones or earthquakes eff ectively destroys the means of livelihood of these families. Finally, maternal deaths have a devastating impact on small children, particularly in poor families, raising the morbidity levels among them (Patil and Koshy 1982). Many of these eff ects, although recognised by fi eld workers, are yet to be empirically measured and studied. Impact evaluation typically stops at counting the dead, and the signifi cant longer-term eff ects remain unaddressed by policy mechanisms.

Lessons from Gujarat: The Four C-Words

Th e earthquake in Bhuj came as a devastating shock to India, and more so to the State of Gujarat.3 Like other disasters it too can serve as a lesson for the future for those willing to learn. In this section we present four c-words as practical lessons to be learnt from this tragic event. Th e fi rst c-word is Co-ordination. Importing seismologists, doctors, and logistics and communications experts into India is mostly unnecessary; this country produces some of the world’s best. Importing manpower is both costly and ineffi cient, especially when India’s own plentiful and skilled human resources could be mobilised for a fraction of the cost of fl ying in and catering to foreign experts. Experiences in other countries have frequently shown that heat, food and other environment-related factors can generate problems for foreign workers, creating an additional burden on local systems. Instead, a strong central co-ordination unit would be able to orient international donor contribution towards mobilising local personnel as far as possible. Th e second c-word is Coverage. Emergency relief is famously in- Downloaded by [University of Defence] at 01:15 24 May 2016 equitable and the poor frequently get excluded. Typically, those who get relief aid are those who can push through crowds or walk to distribution points. Th e more vulnerable people, such as widowed mothers alone with

3 For the last 30 years, according to EM-DAT, it is the Indian natural disaster with the greatest number of people dead or missing (more than 20,000) and reported economic losses (Rs 219 billion or US$ 4.8 billion). Only six droughts, 14 ﬂ oods and one earthquake (Udaipur Gahri earthquake, August 1988) aﬀ ected more people (16 million). 32 Ú Guha-Sapir and Hoyois

small children, or the illiterate elderly are unable to benefi t from such aid. Th ese groups require specially targeted relief coverage. Th e problem can persist even after the emergency phase. For example, property ownership is often the basis on which reconstruction aid or relocation homes are provided. In the earthquakes of Guatemala (1977) and Nicaragua (1972), wealthy slum landlords received most of the relief for reconstruction when aid poured in after the quake. Administrative processes excluded the dirt-poor slum-dwellers because they had no leases, much less any property rights. Specifi c eff orts to ensure that aid gets shared equitably, bearing in mind the caste system, may reduce such inequalities in distribution. Reaching those who do not ask is diffi cult but they are also the most vulnerable and may have lost all their possessions and savings along with their homes. Th e third c-word is Construction Code. Gujarat is a State where the annual per capita income is about Rs 11,000 (approximately US$ 250). Imposing building codes that can raise the cost of construction substantially, as a measure to withstand seismic shocks that may occur within the next expected return period (approximately 187 years) is not economically realistic. On the other hand, strengthening community- based preparedness routines to allow local inhabitants to mitigate and manage the aftermath of a disaster can be achieved. Experience has shown that families and neighbours are the main rescuers in all such situations and that they help with the rebuilding as well. Familiarising neighbourhoods with extrication techniques and simple building skills may be a more cost-eff ective solution for poor communities where in any case it is diffi cult to enforce regulations. Th e last c-word is Change. Natural disasters have a way of opening the doors to political change. Political regimes have fallen as a consequence of major disasters: General Somoza in Nicaragua after the Managua earthquake of 1972, Haile Selassie and Mengistu after the terrible Ethiopian famines of 1972 and 1985. Th e independence of Bangladesh Downloaded by [University of Defence] at 01:15 24 May 2016 followed the terrible cyclone of 1972. Less spectacularly, new legislation, the redistribution of assets and political accountability can also be the positive fruits of a massive disaster. Some changes were introduced in the wake of the Bhuj and Latur quakes and less so, following the Orissa cyclone. Th e real challenge is to achieve institutional paradigm shifts towards vulnerability reduction within development programmes rather than one-off initiatives focussing on small groups of people, led mainly through short-lived project approaches. While the Disasters in India: Policy Overview Ú 33

non-governmental movements are critical to a democracy, they are not a long-term solution to disasters, especially in countries like India which require systemic solutions. So, right now, the Indian authorities and humanitarian agencies need to make sure that the poorest, struggling to survive at the margins, are given the help they need, both now and in the future, when rehabilitation starts. In the longer term, India will need substantial and practical disaster management plans with clear lines of authority and responsibilities. Th e window of opportunity for introducing changes to the system is still open.

Conclusions and Policy Implications for Disaster Preparedness and Management

The increasing interest being shown in the impact of disasters by researchers in disciplines other than engineering, geology and meteorology has had the salutary eff ect of raising questions about current international and national disaster policies and relief action. Th e growing body of literature emphasising the importance of discriminating between the geophysical event and its human consequences is prompting organisations and governments to take another look at disaster relief. Th is is an encouraging turn of events given the large calamities witnessed in the last three years and the worrisome increase in the number of victims, dead or destitute. Furthermore, with non-emergency developmental aid and co- operation between the First and Th ird Worlds grinding to a slow halt, the substantial resources generated by public appeals for disaster assistance demands effi cient and cost-eff ective use, oriented towards a long-term resolution of the problem instead of an emergency, stop gap measure. Disaster relief should be pulled out of its comfortable niche of charity and emergency care. Th is will force public planners to address structural

Downloaded by [University of Defence] at 01:15 24 May 2016 issues regarding the availability of resources, equity and appropriateness with respect to the community in question. A widening of scope would, for example, lead to the inclusion of disaster health planning in the normal health plans of the region. Disaster response could be successfully incorporated into the training of health workers as well as their activities. Th e fundamental tenets of primary healthcare can be applied to disaster preparedness and prevention programmes involving the community at risk, with multi-sectoral objectives, and while promoting the use of local resources. 34 Ú Guha-Sapir and Hoyois

Local community involvement is especially relevant in India where in some regions people could be living in isolated mountainous areas and in others in illegal settlements or precarious rural conditions. External (including national) emergency assistance in the event of a disaster is rarely, if ever, either on time or particularly appropriate. Th is is not due to any sluggishness on the part of the distant agencies, rather it is the result of inadequacies in communications, assessments of need, accessibility and other diffi culties. In the Gujarat and Latur earthquakes, families, friends and neighbours were among the fi rst to search, evacuate and extricate their people in the immediate aftermath of the disaster. In most major disasters, by the time external relief teams become functional on the site, a very large majority of the total that are eventually declared dead would have already died (Guha-Sapir and Carballo 2000). Th ose who die do so within the fi rst few hours of the event, and De Bruycker et al. (1983) have shown that nearly 80 per cent of the survivors were those extricated within the fi rst 24 hours. It is a fact that local inhabitants provide immediate emergency rescue and care, most often using their bare hands or simple tools to rescue people (Noji et al. 1993). Finally, the survivors, contrary to expectations, are rarely in a status of panic or disorganised. In fact, they generally act with common sense and to the best of their abilities, manage their own aff airs. External emergency relief tends to be based on myths that suggest otherwise and frequently, therefore, is not particularly eff ective. Th is, however, does not imply that external or state-led disaster relief should be abandoned simply because local resources exist for people needing help. It indicates, rather, that external disaster relief should focus on reducing population vulnerability and invest in structural changes in healthcare organisation and in improving the accessibility of the population. It should also provide training and education at the local level for emergency measures such as evacuation, fi rst-aid and so forth. Th is rationalisation of policy would provide the communities with Downloaded by [University of Defence] at 01:15 24 May 2016 the tools and knowledge required to try and defend themselves against future hazards. Again, taking the example of the health sector, disaster assistance resources can be deployed to expand the primary healthcare structure and train their personnel in emergency shelter management, rapid epidemiological surveillance and control, food distribution and needs assessment and, recording and registration. Health workers are trained to perform most of these activities under normal circumstances; in a Disasters in India: Policy Overview Ú 35

crisis situation the operating principles remain fundamentally the same, requiring only some adjustment due to the urgency and peculiarity of crisis situations. Disaster policies articulated along these lines would use existing, local infrastructure and local human resources and reduce dependence on external emergency assistance. Disaster relief has traditionally been based on a policy formulated from charitable motives, drawing on critical and emergency care approaches. Th is has made it a primarily medical activity, involving surgical units fl own out with specialist teams and fi eld hospitals equipped with sophisticated life-saving apparatus. On the other hand, research and development in natural catastrophes has concentrated almost exclusively on climate monitoring, radar tracking, fl ood barriers and other middle to high-technology devices. Finally, the charitable nature of disaster relief permits policy-making to be dictated by the principle that any aid is good aid, thereby generating anecdotes in the literature on the superfluous, inappropriate and frequently absurd relief packages. Even volunteers, who tend to descend upon the scene in large numbers, can be a serious liability in a crisis situation, due to their inexperience and redundancy. In conclusion, apart from exceptionally violent events, the major eff ects of disasters on people refl ects the inability of current disaster management policies to reduce the vulnerability of a community. Th is relationship, between a disaster and its impact in complex situations such as those in India, is not a simple one. Diffi cult sociopolitical and economic considerations have to be factored into response planning and preparedness programmes. Th e essence of the vulnerability issue lies in the fact that the communities can often, as Berfolt Brecht put it, ‘cope with earthquakes but not with their fellows’.

References

Downloaded by [University of Defence] at 01:15 24 May 2016 Alexander, D. 1985. ‘Death and injury in earthquakes’, Disasters 9 (1): 57–60. ———. 1996. ‘Th e health eﬀ ects of earthquakes in the mid-1990s’, Disasters 20 (3): 231–47. Armenian, H., A. Melkonian, E. Noji and A. Hovanesian. 1997. ‘Deaths and injuries due to earthquake in Armenia: a cohort approach’, International Journal of Epidemiology 26 (4): 806–13. Armenian, H., M. Morikawa, A. Melkonian, A. Hovanesian, K. Akiskal and H. Akiskal. 2002, ‘Risk factors for depression in the survivors of the 1988 earthquake in Armenia’, Journal of Urban Health 79 (3): 373–82. 36 Ú Guha-Sapir and Hoyois

Belba, G.J. 1996. ‘Problems of Emergency Medical Care at the Time of the Great Hanshin–Awaji Earthquake’, Annals of Burns and Fire Disasters 9 (4): 235–39. Biswas, R., D. Pal, S. Mukhopadhyay. 1999. ‘A Community-based Study on Health Impact of Flood in a Vulnerable District of West Bengal’, Indian Journal of Public Health 43 (2): 89–90. Cuny, F.C. 1983. Disasters and Development. New York, Oxford: Oxford University Press. De Bruycker M., D. Greco, I. Annino, M.A. Stazi, N. De Ruggiero, M. Triassi, Y.P. De Kettenis and M.F. Lechat. 1983. ‘Th e 1980 Earthquake in Southern Italy: Rescue of Trapped Victims and Mortality’, Bulletin of the World Health Organization 61 (6): 1021–25. De Bruycker M., D. Greco and M.F. Lechat. 1985. ‘Th e 1980 Earthquake in southern Italy: Morbidity and Mortality’, International Journal of Epidemiology 14 (1): 113–17. Department of Medical Entomology (DME), US Naval Medical Research Unit No. 2. 1999. ‘El Nino and associated outbreaks of severe malaria in high- land population of Irian Jaya, Indonesia: A Review of Epidemiological Per- spectives’, Southeast Asian Journal of Tropical Medicine and Public Health 30 (4): 608–19. Devereux, S. 2002. State of Disaster: Causes, Consequences and Policy Lessons from Malawi. ActionAid Report, June. de Ville de Goyet, C., E. del Cid, A. Romero, E. Jeannée and M.F. Lechat. 1976. ‘Earthquake in Guatemala: Epidemiologic Evaluation of the Relief Eﬀ ort’, Bulletin of the Pan American Health Organization 10 (2): 95–109. de Ville de Goyet, C. and M.F. Lechat. 1976. ‘Health Aspects in Natural Disasters’, Tropical Doctor 6: 152–57. Earthquake Engineering Research Institute (EERI). 2003. Preliminary Observa- tions on the Tokachi-Oki, Japan, Earthquake of September 26, 2003. Special Earthquake Report, December. Federal Emergency Management Agency (FEMA). 2002. Earthquake Hazard Mitigation Handbook, http://www.conservationtech.com/FEMA-WEB/FEMA- subweb-EQ/(accessed 5 May 2004). Gagnon, A., K. Smoyer-Tomic and A. Bush. 2002. ‘Th e El Nino Southern Oscil- Downloaded by [University of Defence] at 01:15 24 May 2016 lation and Malaria Epidemics in South America’, International Journal of Biometeorology 46 (2): 81–89. Geller, R., D. Jackson, Y. Kagan and F. Mulargia. 1997a. ‘Earthquakes cannot be predicted’, Science 275: 1616–17. Geller, R., D. Jackson, Y. Kagan and F. Mulargia. 1997b. ‘Response to Wyss, Aceves and Park’, Science 278: 488–89. Glass, R.I., J.J. Urrutia, S. Sibony, H. Smith, B. Garcia and L. Rizzo. 1977. ‘Earth- quake Injuries Related to Housing in a Guatemalan Village’, Science 197: 638–43. Disasters in India: Policy Overview Ú 37

Guha-Sapir, D. and M. Carballo. 1999. ‘Th e disaster in Turkey: Lessons for Health Preparedness (Letter to the Editor)’, Lancet 354 (9190): 1649–50. ———. 2000. ‘Medical Relief in Earthquakes’, Journal of the Royal Society of Medicine 93: 59–61. Haas, J.E., R.W. Kates and M.J. Bowden. 1977. Reconstruction Following Disaster. Cambridge, Massachusetts, London: MIT Press. Kazzi, A., M. Langdorf, N. Handly, K. White and K. Ellis. 2000. ‘Earthquake Epidemiology: Th e 1994 Los Angeles Earthquake Emergency Department Experience at a Community Hospital’, Prehospital Disaster Medicine 15 (1): 12–19. Kondo, H., N. Seo, T. Yasuda, M. Hasizume, Y. Koido, N. Ninomiya and Y. Yamamoto. 2002. ‘Post-ﬂ ood Infectious Diseases in Mozambique’, Pre- hospital Disaster Medicine 17 (3): 126–33. Kovats, R., M. Bouma, S. Hajat, E. Worrall and A. Haines. 2003. ‘El Nino and Health’, Lancet 362 (9394): 1481–89. Lechat, M.F. 1984. ‘Natural and Man-Made Disaster’, in W.W. Holland et al. (eds), Oxford Textbook of Public Health, Vol. 1. London: Oxford University Press. Levine, R.J., S. D’Souza, M.R. Khan and D.R. Nalin. 1976. ‘Failure of sanitary wells to protect against cholera and other diarrhoeas in Bangladesh’, Lancet 2 (7976): 86–89. Mondal, N., R. Biswas and A. Manna. 2001. ‘Risk Factors of Diarrhoea Among Flood Victims: A Controlled Epidemiological Study’, Indian Journal of Public Health 45 (4): 122–27. Morgan, O. 2004. ‘Infectious Disease Risks from Dead Bodies following Natural Disasters’, Revista Panamericana de Salud Pública 15 (5): 307–12. Nakano, T., H. Kadomura, T. Mizutani, M. Okuda and T. Sekiguchi. 1974. ‘Natural Hazards – Report from Japan’, in G.F. White (ed.), Natural Hazards – Local, National, Global. New York: Oxford University Press. Noji, E., H. Armenian and A. Oganession. 1993. ‘Issues of Rescue and Medical Care following the 1988 Armenian Earthquake’, International Journal of Epidemiology 22 (6): 1070–76. Patil, B. and S. Koshy. 1982. For Evaluation of the Composite Programme for Women and Pre-school Children in Kerala (an analysis of Care India data), UNICEF. Downloaded by [University of Defence] at 01:15 24 May 2016 Saenz, R., R. Bissell and F. Paniagua. 1995. ‘Post-Disaster Malaria in Costa-Rica’, Prehospital Disaster Medicine 10 (3): 154–60. Seaman, J., S. Leivesley and C. Hogg. 1984. Epidemiology of Natural Disasters: Contributions to Epidemiology and Biostatistics, Vol. 5. Basel, New York: Karger. Sen, A. 1981. Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford: Oxford University Press. Shah, G. 1985. ‘Social Work in Disaster’, Indian Journal of Social Works 45: 462–71. 38 Ú Guha-Sapir and Hoyois

Shoaf, K., H. Sareen, L. Nguyen and L. Bourque. 1998. ‘Injuries as a Result of California Earthquakes in the Past Decade’, Disasters 22 (3): 218–35. Siddique, A., A. Baqui, A. Eusof and K. Zaman. 1991. ‘1988 fl oods in Bangladesh: Pattern of Illness and Causes of Death’, Journal of Diarrhoeal Diseases Research 9 (4): 310–14. Tierney, K. 2000. ‘Controversy and Consensus in Disaster Mental Health Research’. Working paper. Disaster Research Center, University of Delaware. United Nations Development Programme (UNDP). 2004. Reducing Disaster Risk: A Challenge for Development. Geneva: UNDP. US Geological Survey (USGS) Working Group on California Earthquake Prob- abilities. 1999. Earthquake Probabilities in the San Francisco Bay Region: 2000 to 2030 – A summary of fi ndings. USGS Open File Report 99–517. Wade, T., S. Sandhu, D. Levy, S. Lee, M. LeChevallier, L. Katz and J. Colford. 2004. ‘Did a Severe Flood in the Midwest cause an Increase in the Incidence of Gastrointestinal Symptoms?’, American Journal of Epidemiology 159 (4): 398–405. Wieland, M. 2001. Earthquake Alarm, Rapid Response and Early Warning Systems: Low Cost Systems for Seismic Risk Reduction. Proceedings of International Workshop on Disaster Reduction. Reston, VA, August. Wilson, D.C. ‘Earthquake prediction: Methodology and feasability’. Working paper. Department of Earth and Environmental Science, New Mexico Institute of Mining and Technology, http://www.ees.nmt.edu/davew/rweb/ eqpaper.htm Woersching, J. and A. Snyder. 2004. ‘Earthquakes in El Salvador: A Descriptive Study of Health Concerns in a Rural Community and the Clinical Implications. Part III — Mental Health and Psychosocial Eff ects’, Disaster Management and Response 2 (2): 40–45. World Bank. 2004. List of Economies, http://www.worldbank.org/data/ countryclass/countryclass.html (accessed April 2004). Wyss, M., R.L. Aceves and S.K. Park. 1997. ‘Cannot Earthquakes be Predicted?’, Science 278 (5337): 487. Downloaded by [University of Defence] at 01:15 24 May 2016 2 The Uttarkashi Earthquake Management of Relief and Rehabilitation Rajeev Kapoor

Uttarkashi, a large but sparsely populated district in the hills of (the then) Uttar Pradesh was hit by an earthquake of medium intensity (6.1 on Richter scale) in the early hours of 20 October 1991. Although the epicentre was at Agoda, the area in and around Bhatwari (26 km from Uttarkashi, en-route to Gangotri) suff ered the maximum damage. More than 650 lives were lost, the houses of more than 30,000 families were either badly damaged or completely destroyed, and in the worst aff ected areas community assets like irrigation channels, water sources and school buildings suff ered widespread damage. To put things in perspective, the population of Uttarkashi district was 225,000 (approximately 50,000 families) and as per the survey conducted by government functionaries, more than 50 per cent of the population of the district was aff ected by the earthquake — approximately 8,000 families had their houses fully destroyed while the houses of more than 22,000 families were partially damaged.1 Th e only bridge that connected Bhatwari to Uttarkashi had been destroyed, rendering relief administration extremely diffi cult in the fi rst few days after the quake. More importantly, Uttarkashi being a backward and extremely remote hill district (the elevation of aff ected areas being in the range of 4,500–9,000 ft) the infrastructure available was inadequate and it was extremely diffi cult to reach relief supplies to most of the villages. Th e quality and speed of relief and rescue operations has to be seen from Downloaded by [University of Defence] at 01:15 24 May 2016 this perspective. Th e initial rescue operation was managed well by the local administration, with full co-operation of the Army, the Indo-Tibetan Border Police (ITBP) and Seva Sauhard Bandhutv (SSB, an organisation under

1 Th e number of houses that were damaged was much less because more than one family lived in the same house. Th is generated great controversy and will be commented upon later. 40 Ú Rajeev Kapoor

the cabinet secretariat of the central government). Th e State government also showed great maturity and fl exibility in managing immediate relief and rescue measures. However, things changed dramatically within the fi rst month of the quake with complaints of relief material not being reached to interior areas, of corruption, of no eff orts being made for the reconstruction of houses. A scare was created that a major catastrophe would occur in the ensuing winter that was only one month away. Th e opposition political parties took the cue and found this to be a good opportunity to settle political scores. It was at this time that the government panicked and decided to change the district magistrate, who, till then, had done an excellent job of managing the aftermath of the quake; and further decided that all assistance (both cash and material) for the reconstruction of houses be distributed within the next 20 days. In this brief article, an account is presented of the relief and rehabilitation work done from December 1991 to June 1992. An attempt is also made to analyse the possible reasons for the failure of the rehabilitation eff orts of the government.

Relief Policy and Programme

Th e scale of relief announced by the government for Uttarkashi earthquake was:

z For families whose house was fully damaged: cash relief of Rs 20,000 (grant of Rs 10,000 and loan of Rs 10,000) and material (cement, GI sheets and steel) worth Rs 15,000 per family. z For families whose house was partially damaged: cash grant of Rs 5,000 per family. z Gratuitous relief: this was Rs 500 per family aﬀ ected by the earthquake. Th is was only to be given to families rendered destitute and was in addition to the relief that had been distributed immediately

Downloaded by [University of Defence] at 01:15 24 May 2016 after the quake by way of tents, blankets and food, etc.

A survey was completed by local patwaris in the month of November, and the government posted 30-odd additional fi eld staff (sub-divisional magistrates/block development offi cers) to assist the administration in the distribution of relief. In compliance with the instructions of the government, arrangements were made to distribute relief to all the identifi ed families over 20 days in December 1991. Almost 95 per cent of the families were thus given Uttarkashi Earthquake Ú 41

cash relief during this period, but the material component could not be distributed, as it could not reach Uttarkashi in time. Although this cash relief cooled political tempers and was well commented upon by the media as well, the whole programme was fl awed, as will subsequently be discussed in more detail. Th e fi rst reaction from the families was that they were either not given any relief or were given relief suitable for a partially damaged house although they felt that their houses were fully damaged. Demonstrations were conducted at various places in the district over the next one month and after several rounds of discussions and the intervention of local leaders it was decided to invite complaints from people in a standard format. More than 10,000 complaints were received and over the next fi ve months offi cers of the rank of sub-divisional magistrates inquired into them. Only 50 new cases were registered and they were distributed relief in June 1992. Meanwhile, the material component — approximately 300,000 galvanised iron [GI] sheets, more than 50,000 bags of cement and 6,000 tonnes of steel bars — was distributed during April–June 1992. Th us, by June 1992, the declared relief had been almost fully distributed to the aff ected families identifi ed and at this stage the demand for any further relief had also quietened. Relief/rehabilitation was over as far as the government was concerned. But was it truly over for the people as well? Not quite, for despite the huge amounts of money doled out, public satisfaction was close to zero, houses had still not been reconstructed and the vulnerability of the public to any such future event remained high. In the following sections are outlined the major reasons for this failure.

Political Failures

Th e earthquake was used as a tool to score political victories which served to defeat the public cause. While initially it was the opposition that was responsible for this, the ruling party too was guilty of not tackling the

Downloaded by [University of Defence] at 01:15 24 May 2016 political sloganeering. Th eir response to the opposition’s complaints of poor relief operations was to pour in loads of money without much planning or a programme. Th is did shut the opposition up but meant that in the long run the public was left to fend for themselves. Instead of engaging the opposition and the aﬀ ected public in a dialogue and designing a relief policy based on a careful assessment of the local conditions and after consulting the aﬀ ected people, money was thrown in without any design. In their hurry the government forgot that winters would be the worst time for any house construction activity in the hills, with the result that 42 Ú Rajeev Kapoor

a large number of beneﬁ ciaries used up the cash given by the State for other purposes, and continued to live in houses that were badly damaged. Even the model design of houses recommended by the government (that included cement and iron, which had been sparingly used in the past), did not take into account the local realities that masons in the hills did not know how to construct houses using cement and concrete. Had the politics been more mature, the relief and rehabilitation work would have helped the earthquake aﬀ ected to rebuild their homes and lives. Instead, relief funds were utilised largely for consumption, and the people, ultimately, did all the rebuilding at their own slow pace over a long period.

Policy Failures

Th is was the most serious failure. Although, this was partly a result of the political failure, but in large measure it was also a bureaucratic failure. Th e relief and rehabilitation policy was ill-conceived, impractical and devoid of an awareness of the local reality. Th e following are some comments on the major policy-related shortcomings.

No Clarity Regarding Relief Entitlement

Th e major relief provided was for house repairs/rebuilding. No thought was given to the basis on which a village patwari would classify a house as partly or fully damaged. Would a house that might still be standing but which would have to be pulled down for repairs classify as partly or fully damaged? Th is, for instance, was never explained, and led to major anomalies and was a prominent cause for discontent. Equally diﬃ cult was the question of deﬁ ning a family. Relief was given on a per family basis. As a result, at many places patwaris counted all married adults living in the same house as separate families, while in many other places the lists included only the house owner. In fact, the government added to the con-

Downloaded by [University of Defence] at 01:15 24 May 2016 fusion mid-way by issuing a circular that relief was to be given per house and not per family, but by that time most of the money had been already distributed. Lack of clarity regarding entitlements was the major cause for the disturbances and agitations that followed relief distribution.

Design of Relief Package

As noted earlier, the relief package consisted of both a cash and a material component. Th e material component was introduced so as to ensure Uttarkashi Earthquake Ú 43

that the new houses be built with cement concrete bands to make them better resistant to future earthquakes. However, planners and policy makers did not consider the following facts while prescribing the design and the relief package:

(i) Th ere existed no local expertise on the use of iron and cement. Improperly used, they would render the houses even more unsafe. (ii) It would be extremely expensive to transport these items to high altitude areas not connected by road. In these areas, water availability was another constraint that would make the use of cement concrete impractical. As a result, a large quantum of the relief material provided found its way back into the local market. (iii) No thought was given as to where the villagers would store cement, susceptible as it is to setting, and given that they were living in tents. (iv) Considering that the material was not to be made available till April 1992 why the government insist that the cash component be handed over in December 1991? By the time the material was made available the cash relief had been consumed already.

Earthquake Shelters

Another example of an impractical approach adopted was the decision to construct 300-odd earthquake shelters over January–February 1992. Th ese were patterned on shelters that are used by the army during a war and were constructed using tin sheets and an angle iron structure. First, they were totally unsuitable for hilly areas — extremely hot in summer and extremely cold in winter — and second, it was not clear why they had to be constructed in such a hurry. If it was being done to help the victims

Downloaded by [University of Defence] at 01:15 24 May 2016 it was too late for they had already settled into tents or other temporary structures and if it was to meet a future contingency, more research and local consultation would have helped in designing structures that were better suited to local conditions. As a result, more than Rs 6 crores were spent and lots of managerial attention was consumed but without any commensurate gain. Th e money and the eﬀ ort could have been used better to provide immediate relief to the aﬀ ected families. It appears that the senior management was in a hurry to get the relief distribution work over with and was unconcerned with whether, 44 Ú Rajeev Kapoor

ultimately, the quake-aff ected got properly settled or not. Th is was refl ected in a lack of appreciation for the time that it would take for a proper survey in a diffi cult hill district like Uttarkashi. A properly designed programme would have earmarked at least two to three months for a survey and for crosschecking, and that period should have been used to fi nalise the relief package. Th is time, however, was not given. As a result a huge amount of money was distributed on the basis of a survey done by patwaris over a 10-day period and without crosschecking survey results at any level. Th is led to widespread resentment and, despite lots of money pouring in, the public felt dissatisfi ed with the outcome. Th e fact that the politicians were in a hurry explains the urgency but this was no excuse to launch a hasty and poorly designed relief operation that resulted in a colossal waste of public money. It seems that the senior administration had no clue as to how earthquake rehabilitation work should be planned and managed. Th e UP administration was more used to managing fl ood relief, where the entire operation is over in a month or so. Th e same approach was used to deal with the rehabilitation of earthquake victims, which is a long-drawn aff air, and it is not surprising that it failed the way it did.

Implementation Failures

A large part of the implementation failure was a result of policy failures. Relief distributed on the basis of improper surveys and at the wrong time was bound to generate dissatisfaction. However, the following factors contributed to the panic that was felt at the State level and may have led to the knee-jerk responses of the policy makers.

Media Management

In the aftermath of the quake the entire national media (as well as

Downloaded by [University of Defence] at 01:15 24 May 2016 many from the international media) descended on Uttarkashi. Th ey were new to the hills and the local people, similarly, were new to them. Local, small-time politicians found an opportunity to see themselves in print (and on TV) and the media also found it more interesting to highlight all the negative reactions without crosschecking any facts. As a result, despite the immediate rescue and relief operations having been managed relatively well, media reports were extremely negative. Th is led to a national uproar and a panic reaction set in at the top- most administrative level. Had the media been better managed, and Uttarkashi Earthquake Ú 45

had the oﬃ cials in charge made arrangements for regular interactions with them, the results might have been diﬀ erent. In fact the State media cell became very active when relief was being distributed in December and the media reports of that period were very positive but by then the damage had already been done.

Inadequate Feedback

Th e local administration should have been fi rmer in representing the ground realities and should have insisted on correcting the policy. Th is was not done. Understandably, it would have been diffi cult for the junior administrators to question the wisdom of the government, but this would have been possible only if the questioning was done by a representative consultative body comprising infl uential politicians, social workers and aff ected people. Th e absence of such a body led to the administration implementing policy directives that were impractical, and also having to defend them before the public.

Management of the NGO Effort

A large number of social organisations came forward to oﬀ er relief immediately after the quake. Many of these were not professional and some were dubious in their approach. Many of them came with material (such as blankets, clothing and, in some cases, food) and cash relief. Th ey were new to the hills and, as mentioned earlier, most of them had nothing more than social concern to guide them. Unfortunately, the eﬀ orts of these organisations lacked direction and were not co- ordinated by the district administration either. As a result, most of the relief material was distributed in the few villages that were closer to the district headquarters and were accessible by motor transport, while genuinely needy villages in the interior areas were deprived of

Downloaded by [University of Defence] at 01:15 24 May 2016 this assistance, which would have been very welcome in the immediate aftermath of the quake. Not only this, these social organisations also reported negatively about the administration, as they were not guided well. Th is corroborated the reporting by the media and added to the panic of the State government. During the rehabilitation phase also many NGOs came forward to provide help in the reconstruction activities. Of these, a few, such as the People’s Science Institute (PSI), Dehradun, helped the administration create an awareness about improved techniques for constructing 46 Ú Rajeev Kapoor

low-cost but safe houses in the hills and another NGO made a video fi lm on the same subject that was circulated to various villages. Th is technical support was helpful and their eff orts were co-ordinated with the district administration. However, many NGOs came forward to reconstruct the houses in several villages adopted by them. It was here that the administration was nonplussed. Although government relief had already been provided to all the aff ected villages some of them had their houses constructed by the NGOs while others had to manage with the government relief only. Had the distribution of relief been delayed, there would not have been any need to provide government assistance for the villages adopted by the NGOs, and the quantum of relief to other villages could have been enhanced. Th is too led to lots of resentment and dissatisfaction amongst the public.

Conclusion

It is sad but true that despite the large sums that were distributed as relief and the eff orts put in by the administration, the earthquake- aff ected people perceived the government and the administration as being insensitive and unhelpful. Politics took precedence over propriety and even the bureaucracy displayed an incapacity to advise the political leadership and to envision the reconstruction task with clarity. Th e result was a policy that was poorly conceived, impractical and ill-suited to local conditions. One wonders if things would have been diff erent had the quake-aff ected areas been located in the plains, with which the bureaucrats were more familiar, and which were more accessible and would therefore be easier to administer relief. One also wonders if this was not also a refl ection on the general inadequacy of the regulatory administration in implementing social development schemes: the immediate rescue and relief operations that required a quick and urgent response were managed rather effi ciently, both by the top manage-

Downloaded by [University of Defence] at 01:15 24 May 2016 ment and the administration, but when it came to tackling rehabilitation, which required a more long-term, process-oriented and consultative approach, the top management and the administration proved unequal to the task. 3 Earthquake Reconstruction in Maharashtra Impact on Assets, Income and Equity Krishna S. Vatsa

Post-disaster reconstruction has become big business today with billions of dollars in investment pouring in. Due to the volume of these investments, the guiding principles behind reconstruction have become the subject of serious academic debate. Th is article presents a case study of an earthquake reconstruction programme from India, which was implemented in the Marathwada region of Maharashtra after the Latur earthquake that took place on 30 September 1993. Th e objective is to analyse the impact of reconstruction on communities and households in terms of building assets and livelihoods. It seeks to assess the contribution of the reconstruction programme in building household assets, expanding economic opportunities and reducing the vulnerability of earthquake-aﬀ ected people. It also explores certain considerations of equity and empowerment that could be introduced through a reconstruction programme. When an earthquake of magnitude 6.3 on the Richter scale struck the State of Maharashtra in India on 30 September 1993, it turned out to be the most devastating natural disaster ever experienced by the State. It left 7,928 people dead and more than 16,000 injured. Th e mass collapse of houses in the area rendered a large number of families homeless. In the two districts of Latur and Osmanabad, which were

Downloaded by [University of Defence] at 01:15 24 May 2016 the worst-aﬀ ected, 70 villages were completely destroyed, and more than 1,500 villages reported extensive damage to houses and public property. Another 11 districts of Maharashtra suﬀ ered damage, though to a much smaller extent.1 The total property loss was estimated

1 Besides Latur and Osmanabad (which belong to the region known as Marathwada), the other districts that suﬀ ered damage and destruction were Nanded, Parbhani, Aurangabad, Beed (Marathwada); Solapur, Sangli, Satara, Kolhapur, Pune, Ahmednagar (west Maharashtra); and Nashik (north Maharashtra). 48 Krishna S. Vatsa

Map 3.1: Districts of Maharashtra

N I Dhule Bhand-ar-a Nagpur JaLpaou Ainc.i'.Mfi

Wardha Buldhan! Akob Clun rlr^pur Nashik itoocnuJ ^ 1h.inc

A!illi.|..[siij',,l[ Fa^hani >o ficvd s Nanded Kiigad l.-iuir Epicentral Chinj SnIjJ'.iUF region R.itnjfiiri ^5^ CJsm^n.ib^d 4*% [ndl.i

SangjT TtongLidesh

•::-.:; n.liii.

Sri Lanka ?0 0 20 40 Miles /

Source: Maharashtra Remote Sensing Applications Centre (MRSAC), Nagpur. Note: The districts that are shaded participated in the earthquake reconstruction programme.

to be approximately US$ 300 million, which is rather low compared to the loss of life, suggesting poor housing stock in the area. Soon after the earthquake, the Government of Maharashtra (GoM) undertook a large-scale damage assessment exercise and began designing the reconstruction programme. It was effectively facilitated by the World Bank's offer of support with emergency reconstruction credit. Over the next six years, the Maharashtra Emergency Earthquake Rehabili- tation Programme (MEERP, as it came to be known) was implemented, with credit from the World Bank and assistance from a number of other donor organisations.2 In terms of housing units, it was one of the largest 3 Downloaded by [University of Defence] at 01:15 24 May 2016 post-disaster reconstruction programmes ever to be implemented.

2 The World Bank loan was provided through the soft loan window of the International Development Association (IDA). The credit became effective on 1 July 1994 and concluded on 31 December 1998. The reconstruction programme continued for another year. 3 In the wake of the Bhuj earthquake of 2001, a larger housing programme has been implemented under the Gujarat Emergency Earthquake Reconstruction Programme. Earthquake Reconstruction in Maharashtra Ú 49

Data Description

Th e development credit agreement signed between the World Bank and the GoM stipulated that a baseline survey of the programme benefi ciaries be carried out. Th e Tata Institute of Social Sciences (TISS) was engaged to carry out a rapid survey of benefi ciaries’ preferences. A proper household survey was not carried out until 1998–99, when the GoM engaged the services of the Centre of Studies in Social Sciences (CSSS), Pune, to prepare a database of programme benefi ciaries. Th e data was collected through a baseline survey of the benefi ciaries on a census basis (full head count) in category ‘A’ (i.e., 27,415 benefi ciaries in 52 relocated villages).4 Since the total number of benefi ciaries in ‘B’ and ‘C’ category villages was approximately 200,000, in these categories a sample of 5 per cent was considered in the survey. CSSS surveyed 23,498 of the 27,415 benefi ciaries in category ‘A’ villages (88 per cent). Th e remaining benefi ciaries were either dead or had migrated, or were minors, or were untraceable. In category ‘B’, a total number of 451 benefi ciaries from four villages were surveyed, with two villages each from the Latur and Osmanabad districts. Finally, a total of 9,057 benefi ciaries were surveyed in 15 category ‘C’ villages. Th e selection of sample villages for conducting fi eldwork in the category ‘C’ villages was done by following a random sampling method with probability proportionate to the number of households in each sub- category, i.e., C1 and C2. Four sets of questionnaires were prepared by CSSS for the fi eld survey:

(i) Questionnaires for individual beneficiaries in category ‘A’ villages (ii) Questionnaires for individual beneﬁ ciaries in villages in categories ‘B’ and ‘C’

Downloaded by [University of Defence] at 01:15 24 May 2016 (iii) Village-level questionnaire for 52 villages (iv) Questionnaires for the NGOs participating in the MEERP

Besides collecting information through questionnaires, CSSS also conducted village-level meetings in all of the 52 relocated villages and sample villages from categories ‘B’ and ‘C’. Care was taken to ensure

4 Th e total number of beneﬁ ciaries in category ‘A’ increased further to 27,944 at the conclusion of the programme. 50 Ú Krishna S. Vatsa

that elected representatives and oﬃ ce bearers of the Gram Panchayat (village council), women representatives and other members of village- level committees attended these meetings. My research refers to a second data set as well, compiled by the Eco- nomic and Political Weekly (EPW) Trust in 1998 on the basis of a sample survey of 2,600 beneﬁ ciaries drawn from 123 villages. Th is data has been used only in part, wherever the CSSS data did not provide information. Th e data was collected by distributing questionnaires, prepared separately for each category of villages. Th e surveyors also organised group meetings of villagers to supplement the data and information obtained from the household surveys.

Programme Assistance and Household Characteristics

Th e MEERP was implemented at a total cost of US$ 358 million. Th e World Bank provided loan assistance of US$ 221 million. Th e GoM and other donors provided the remaining amount needed for reconstruction. Among the international donors the most prominent was UK’s Department for International Development (DFID). It co-fi nanced the project along with the World Bank with a grant of £ 12.5 million. Th e Asian Development Bank (ADB) and the United Nations Development Programme (UNDP) provided technical assistance. Th e eff ective programme fi nancing plan is shown in Table 3.1.

Table 3.1: Eﬀ ective Project Financing Plan

Agency Amount (US$ million) Percentage of programme cost a. IDA 221.90 61.81 b. GoM 95.91 26.82 c. Donors 40.65 11.37 Total 357.56 100.00 Downloaded by [University of Defence] at 01:15 24 May 2016 Source: Earthquake Rehabilitation Cell (1999: 5).

Schemes for Earthquake Rehabilitation

In 1994, the GoM drafted an Earthquake Rehabilitation Policy, which laid down the implementation strategy and entitlements for the reconstruction programme. Th e primary objective of the rehabilitation policy was the ‘comprehensive and satisfactory resettlement and Earthquake Reconstruction in Maharashtra Ú 51

rehabilitation of aﬀ ected people and villages’.5 Its principal components were the reconstruction of housing and infrastructure, and economic, social and community rehabilitation. Th e policy made special mention of protecting the rights of women and children and sought to ensure full participation of the community in project implementation. Th e Maharashtra government decided to relocate the most damaged villages because the scale of destruction had been massive and the levels of trauma experienced by people were also very high. People in these villages were also keen on relocation, for the earthquake had completely destroyed their villages. On the basis of extensive consultation with all the stakeholders, the government developed a housing policy which laid down the criteria for three categories of assistance:6

(i) Relocation villages (Category A): included 52 villages, in which over 70 per cent of the structures suﬀ ered damage of categories IV or V (collapse or severe structural damage), and where the depth of black cotton soil was more than 2 metres.7 Th ese villages were relocated and constructed on a nearby location. (ii) Hybrid villages (Category B): included 22 villages, which suﬀ ered the same level of damage, but had black cotton soil of less than 2 metres depth.8 Th e policy envisaged that these villages would be resettled in-situ. (iii) Reconstruction, Repair and Strengthening Programme (RRSP or Category C): included all the remaining villages, where the Maharashtra government provided two packages of assistance, depending upon the severity of damage for in-situ reconstruction, repairs and strengthening.

Th e data summary for villages in categories ‘A’, ‘B’ and ‘C’ is given in Table 3.2. Downloaded by [University of Defence] at 01:15 24 May 2016 5 http://www.maharashtra.gov.in/index/stateInitiatives/meer.php (accessed on 15 February 2003). 6 Th e number of deaths as the basis for the categorisation of villages was mentioned in the ﬁ rst proposal prepared by the Maharashtra government. However, immediately after, it was removed as a criterion in the revised proposal. 7 Black cotton is an expansive soil with very poor foundational strength. 8 It originally included 16 villages, but six more villages were later added to this category. 52 Ú Krishna S. Vatsa

Table 3.2: District-wise Beneﬁ ciary Households in Diﬀ erent Categories of Villages

No. of No. of houses Description villages Latur Osmanabad Total Category ‘A’ villages 52 14,327 13,617 27,944 (relocated villages) Category ‘B’ villages 22 4,566 6,056 10,622 (in-situ reconstruction) Category ‘C’ villages 1,500 103,381 60,938 164,319 + 24,794 (repairs and strength- (from 11 other ening programme) districts) = 189,113 Source: Compiled by author from Earthquake Rehabilitation Cell (1999: 8–60), section on housing.

Individual Entitlements

Th e policy laid down individual entitlements in respect of villages divided into categories ‘A’, ‘B’ and ‘C’, though these entitlements evolved further during the course of implementation. In category ‘A’ villages, which were relocated, the individual entitlement available to all households was decided on the basis of the original landholding. Th e size of plots and houses corresponding to landholdings has been described in Table 3.3. Th e need for more space for agricultural operations with the increase in the size of landholdings was the rationale for providing plots and houses of diff erent sizes. Th e principle of equity was observed by laying down that all benefi ciaries would be provided with plots with the fi rst 250 sq. ft. carpet area of a core house free of cost, while the area in excess of 250 sq. ft. in the larger core houses would be treated as a loan to the benefi ciaries. In category ‘B’ villages, all the benefi ciaries were given the same amount of assistance, Rs 62,000 for reconstruction.9 In category ‘C’ villages, there were two levels of assistance depending upon the extent of damage to houses: Rs 17,000 (C1), and Rs 34,500 (C2). Downloaded by [University of Defence] at 01:15 24 May 2016 Women who were widowed in this earthquake and were otherwise not entitled to a plot and a house because the original landholdings were not in their name would be given separate plots of 1,575 sq. ft.

9 All category ‘B’ villages were to be reconstructed in-situ. However, almost all the villages demanded relocation. Initially, the government was reluctant to give permission, because it would also involve building civic amenities and social infrastructure. Later, however, it agreed and allowed all the villages to be resettled on a new site. Downloaded by [University of Defence] at 01:15 24 May 2016

Table 3.3: Scheme of Housing Entitlements under the MEERP

Earthquake-aff ected villages Criteria for classifi cation of villages Classifi cation of benefi ciaries Assistance Category ‘A’ More than 70 per cent of the houses collapsed Landless and those holding land up 1,575 sq. ft. plot and or suff ered severe structural damage; surface to 1 hectare 250 sq. ft. house black cotton soil of more than 2 metres Landholding between 1 and 7 2,520 sq. ft. plot and depth (52 villages in Latur and Osmanabad hectares 400 sq. ft. house districts) Landholding above 7 hectares 5,000 sq. ft. plot and 750 sq. ft. house Category ‘B’ More than 70 per cent of the houses collapsed All eligible benefi ciaries Rs 62,000 per house or suff ered severe structural damage; surface black cotton soil of less than 2 metres depth (22 villages from Latur and Osmanabad districts) Category ‘C’ All other villages, with varying degrees of Th ose whose houses had collapsed Rs 34,500 per house damage (from Latur, Osmanabad and 11 or suff ered severe structural other districts of the State) damage (IAEE Category IV and V) Th ose whose houses have suff ered Rs 17,000 per house slight to moderate damage (IAEE Category I–III) Source: Compiled from GoM (1994) and other project documents. 54 Ú Krishna S. Vatsa

with 250 sq. ft. houses. Ownership of the land and the house were to be in the name of the widows. In all other cases, plots and houses would be provided in joint ownership of the husband and wife, except where the benefi ciary was a widower. In case of a minor orphan becoming entitled to a house because of the death of his parents, who would otherwise have been entitled to the ownership of a house, the Additional Collector of the district was nominated as the ‘Court of Ward’. In addition to housing, the other components of the rehabilitation programme included infrastructure, social rehabilitation, economic rehabilitation, and technical assistance, training and equipment. Most of the infrastructure, civic amenities and social services were provided in category ‘A’ villages. In villages categorised as ‘B’ or ‘C’, assistance was limited to the reconstruction, repair and strengthening of houses. Of the diff erent kinds of assistance provided under MEERP, economic assistance was provided at the level of households on the basis of an assessment of losses in the worst-aff ected villages. Under this programme, agricultural implements, livestock, tool kits and help in dig- ging wells were provided. Assistance for agricultural implements had two categories: major and minor. Th e typical major implements used in agriculture are wooden and iron ploughs, bullock carts, diesel and electric pumps, tractors, trolleys and malanis (harvesters), while the minor implements are tifans (three-bowl seed drills), kolpes (weed removers), auts (drills). Altogether, 24,363 benefi ciaries were provided with major implements and 12,639 benefi ciaries with minor implements. Under the livestock programme, 1,534 cows, 1,812 buff aloes, 1,358 sheep, 8,478 goats and 14 donkeys were distributed to the benefi ciaries. Th e number of livestock to be distributed was arrived at through a survey of cattle deaths. Artisans and small business persons were also helped with the provision of tool kits. Altogether 979 benefi ciaries received tool kits under the programme (Earthquake Rehabilitation Cell 1999). Most of these benefi ts went to households in category ‘A’ villages.

Downloaded by [University of Defence] at 01:15 24 May 2016 Independent of the MEERP, the Chief Minister’s Relief Fund was another important source of personal assistance following the earthquake. Th rough the Fund, households that had lost family members received Rs 50,000 per person as ex-gratia assistance. Th e Relief Fund also provided Rs 25,000 for those who had suffered full disability and Rs 10,000 for people with partial disability. As 8,000 people died and 16,000 people got injured in the earthquake, approximately Rs 194 million was disbursed to households in the worst-aﬀ ected villages, of which 70 per cent was given in the form of savings certiﬁ cates. Earthquake Reconstruction in Maharashtra Ú 55

Household Characteristics

Latur and Osmanabad districts, situated in the Marathwada region of Maharashtra, are characterised by low levels of economic development, urbanisation and literacy, and a high incidence of poverty. More than 70 per cent of the population is dependent on agriculture and allied activities for a livelihood. Agricultural labour in Latur and Osmanabad districts comprises 31 and 41 per cent of the working population, respectively. Th e Scheduled Castes or Dalits, who are socially weak and economically marginalised communities, constitute 19 and 16.34 per cent of the population of Latur and Osmanabad districts respectively, as against the State average of 11 per cent (Census of India, 1991). More than 30 per cent of the population of these ditricts lives below the poverty line, as deﬁ ned by an annual income of Rs 11,000 for a family. In category ‘A’ villages, the total population of the households is 122,355, making the average family size 5.2 persons. In ‘B’ and ‘C’ category villages, the average size of a household is 5.71 and 6.33 persons respectively. Th e average size of households in Latur and Osmanabad is 5.91 and 5.48 (ibid.). The smaller household size in category ‘A’ villages, compared to the other districts and categories of villages, can be explained by the large number of deaths in the earthquake and the rise of nuclear families after resettlement. Th e caste composition and occupational patterns in these villages are consistent with aggregated patterns at the district level. Th e poverty and marginalisation of households in this area needs to be understood within the regional context of Marathwada. In pre- Independence days, Marathwada was part of Hyderabad state and was ruled by the Nizams. Th eir rule was notorious for exploitation and tyr- anny, and establishing a regressive feudal order in the region. Marathas were the dominant caste in the region, and the other, weaker castes bore the brunt of social discrimination. Th e region did not come under the

Downloaded by [University of Defence] at 01:15 24 May 2016 inﬂ uence of the social reform movements, of the late nineteenth and early twentieth centuries, that had played an important role in making Maharashtra one of the more progressive States in India. Th e social lag between other regions of the State and Marathwada has therefore been an important factor contributing to its backwardness (Damle 2001). Marathwada falls in an arid zone and has hot climate, with temperatures in region of 9°c–43°c and moderate rainfall, averaging about 750 mm between June and September. Th e terrain is undulating and the soils are shallow. Deterioration in water management practices in the 56 Ú Krishna S. Vatsa

region has made it prone to frequent droughts. Agriculture is practiced on a subsistence basis, with very little surplus being generated for the market. And with no worthwhile industry, except a few sugar mills, the economic prospects for the region are not bright.

Main Findings and Observations Provision of Housing

With a magnitude of 6.4 on the Richter scale, Killari was not a major earthquake. Yet the number of deaths it caused, approximately 8,000, was staggering. Th e inferior quality of construction and degradation in stone masonry construction practices, in conjunction with the poverty and social backwardness of the region, explain the unusually high number of deaths. Improved and safe housing was, therefore, the principal objective of the rehabilitation programme. Apart from providing much needed shelter, housing was a critical asset for sustaining agricultural operations, livestock and home-based enterprises. Since housing was considered the key to a recovery, it would be well in order to analyse ﬁ ndings about diﬀ erent aspects of housing under the MEERP.

Relocation (‘A’) Villages: Difference between New and Old Houses

In the 52 villages of Latur and Osmanabad districts that were relocated, 27,944 houses were constructed. At the commencement of the rehabilitation programme, 23,000 houses had been planned. Th e increase of more than 21 per cent in the scope of the programme could be attributed to younger members of the household, who were earlier part of the extended household and were now claiming to head nuclear families. Th e district-wise ﬁ gures for all three types of houses are shown in Table 3.4. Downloaded by [University of Defence] at 01:15 24 May 2016

Table 3.4: Construction of Houses in Category ‘A’

‘A’ type houses ‘B’ type houses ‘C’ type houses District (250 sq. ft.) (450 sq. ft.) (750 sq. ft.) Total Latur 7,749 5,831 747 14,327 Osmanabad 6,831 5,625 1,161 13,617 Total 14,580 11,456 1,908 27,944 Source: Compiled by author from Earthquake Rehabilitation Cell (1999: 8–60), section on housing. Earthquake Reconstruction in Maharashtra Ú 57

Relocated villages turned out to be very diff erent from the old villages. Th e diff erences between these settlements can be described in the following ways: Building Materials used in Old and New Houses In the relocation villages, the government took on the responsibility of reconstruction and engaged contractors to build houses. Even the NGOs and donor organisations who took on the responsibility of relocating villages worked through contractors. New houses were constructed using load-bearing masonry construction technology. Solid concrete blocks were used for wall construction. Reinforced concrete slabs were selected as feasible roofi ng structures, replacing the traditional timber-plank roofs. Mud mortar, frequently used as a binder in masonry construction prior to the 1993 earthquake, was replaced with cement-based mortar. Th is was radically diff erent from the old villages, in which approximately 90 per cent of the housing stock was constructed using uncoursed random-rubble stone masonry, with mud-mortar bonding. Besides, the old houses had heavy earthen roofs supported by timber planks and joists. It was the failure of stone masonry walls and the collapse of earthen roofs supported by stone walls, in fact, that produced the most fatalities in the old villages (Earthquake Engineering Research Insti- tute (EERI) 1999; Momin, Nikolic-Brzev and Bajoria 1996). Improvements upon Old House Types We may get a clearer picture of the change in the housing situation by comparing the new houses in relocation villages with the old houses occupied by the benefi ciaries prior to the earthquake. Information about the diff erent types of old houses is presented in Table 3.5. Th e new houses represented a signifi cant improvement for about 37 per cent of the families who had lived in huts and tin sheds in their old villages. Whether these houses represented an improvement upon

Downloaded by [University of Defence] at 01:15 24 May 2016 Table 3.5: Information on Diﬀ erent Types of Old Houses

Type of house Number Percentage Hut 1,356 5.86 Tin shed 7,133 30.83 Wada 8,705 37.63 Independent bungalow 5,940 25.68 Total reporting households 23,134 100 Non-reporting households 166 Source: Centre of Studies in Social Sciences (CSSS) (1999). 58 Ú Krishna S. Vatsa

the wadas that constituted a majority of the traditional dwellings, needs to be seen from the viewpoint of a changing social milieu. A wada represents the architectural style of a feudal society. Th ese are single- storeyed buildings that consist of a central courtyard, small rooms and a colonnaded verandah in front of the rooms. High plinths and stone walls give these houses a monumental look. For social reasons (seclusion of women) and security, these houses were inward-oriented with small doors and window openings. In all new constructions elsewhere in the region, households have discarded this style almost completely, in favour of independent bungalows. Condition of Old Houses Th e condition of housing in the region prior to the earthquake was poor, refl ecting the poverty and economic stagnation of the area. Th e annual growth rate of the housing stock in these two districts was a low 3 per cent (EERI 1999). About two-thirds of the houses in the old villages were more than 20 years old. Area of Old and New Houses Around 60 per cent of the households reported their old houses had an area larger than 400 sq. ft. However, in the new villages only 7 per cent of the households have an area larger than 400 sq. ft. (i.e., 750 sq. ft.). In terms of the built-up area, in the new villages, households have smaller houses. However, the older houses had been occupied by extended households. A number of family units lived in these houses, thereby explaining their larger size. In the new villages, all nuclear family units were entitled to an independent house. So while the area of the houses in the new villages had decreased, the number of houses went up signifi cantly. It also needs to be said that the reconstruction programme never did intend to replace the old houses. Under the MEERP, the GoM undertook essentially to provide a core house, which the benefi ciary could later expand. Th e plot given with the house was much bigger, enabling Downloaded by [University of Defence] at 01:15 24 May 2016 extended construction of the core house. Provision of Bathrooms and Toilets Bathrooms and toilets are provided in almost all the new houses as a matter of policy. Th ese facilities have been one of the most important improvements upon the old houses (see Table 3.6 for a comparison between new and old houses on the provision of these facilities). While most of the households earmarked a private space for bathing, the use of internal toilets in this region was limited to a very small number of Earthquake Reconstruction in Maharashtra Ú 59

Table 3.6: Provision of Bathrooms and Toilets

Amenities New houses (Percentage) Old houses (Percentage) Provision of Bathroom Independent bathroom 96.98 50.13 Facility with some privacy 2.88 47.15 Common bathroom 0.14 2.72 Provision of Toilet Independent toilet 95.77 3.17 Common toilet 1.24 13.61 No facility 2.99 83.22 Source: CSSS (1999).

households. While a large number of households still use these toilets as a storage space in their new houses, with the spread of concepts like sanitation and hygiene their use has steadily increased. Provision of Electricity and Water Supply Th e new houses have also been an improvement in terms of electricity connections, though piped water supply did not achieve its capacity utilisation. At the time of data collection, the work of providing electricity connections in all the relocation villages was in progress, and the data reﬂ ects this incomplete activity (see Table 3.7). For the provision of drinking water, a regional water supply scheme, covering all the relocated villages, was undertaken. Th e scheme ran into serious technical diﬃ culties,

Table 3.7: Provision of Electricity and Water Supply

Amenities New houses (Percentage) Old houses (Percentage) Provision of Electricity Independent connection 64.11 37.34 Common connection 4.18 17.07 No connection 14.45 43.37 Illegal connection 17.25 2.23 Sources of Drinking Water Downloaded by [University of Defence] at 01:15 24 May 2016 Public handpump 24.57 39.71 Public pipe supply (standposts) 61.14 25.65 House connection 10.50 22.06 Well 3.76 12.54 River, stream 5 0.04 Adequacy of Water Adequate 40.61 89.79 Less than adequate 47.35 6.69 Storage in summer 12.05 3.52 Source: CSSS (1999). 60 Krishna S. Vatsa

thereby affecting the water supply. Most of the houses and villages had to depend on temporary water supply schemes and did not have the same level of access to water as in the old villages. It was expected that as the scheme became fully operational, houses would get adequate water supply. However, at present, the inadequacy of drinking water remains the single biggest problem faced by residents of the new houses. Provision of Civic Amenities A number of civic amenities were provided in the relocated villages — some of these existed in the old villages, while some were created anew. In the survey of beneficiaries, most of the amenities provided in the new villages have received high scores. The score ranges between -100, the most unfavourable score, to +100, the maximum favourable score. The value of zero signifies a neutral response. Prominent among the amenities receiving high scores are schools, education, health centres, internal roads, connecting roads, women's centres, Aanganwadi (child feeding centres), Balwadi (Pre-schools), libraries, and community centres. All these amenities have a score of around 80. Some of these high-scoring amenities, like Balwadi, Aanganwadi, women's centres, etc., are new introductions (see Figure 3.1). These amenities, if maintained,

Figure 3.1: Households' Perception of Civic Amenities ABCDEFGHI JKLMNOPQRS 100

40 g u CO 20

0 Downloaded by [University of Defence] at 01:15 24 May 2016 -20

-40 Type of Amenities Source: CSSS (1999). Notes: A: Design of new village; B: Educational amenities; C: Tech. training facilities; D: Health amenities; E: Aanganwadi; F: Balwadi; G: Library; H: Gymnasium; I: Community centre; J: Women's centre; K: Godown; L: Crematorium; M: Street lighting; N: Internal roads; O: Connecting roads; P: Bus shelter; Q: Passenger transport facility; R: Social forestry; S: Office of co-op societies. Earthquake Reconstruction in Maharashtra 61

will contribute to the quality of life in these villages. The three amenities which have negative scores are crematoriums, technical training and street lighting. Out of them, only crematoriums has a large negative score, the remaining two have a score nearer to zero. Safety and Quality of Construction and the Application of Earthquake-resistant Technology Most of the beneficiaries considered the new houses to be safe despite the fear of another earthquake. They considered the use of cement, sand, hollow concrete blocks and earthquake-resistant technology as contributing to the quality of construction and the safety of houses. Figure 3.2 indicates the percentage of response in favour of different attributes of newly constructed houses. Earthquake-resistant technology has been the most valued feature of the new houses.

Relocation Over In-situ Reconstruction: 'B' Category Villages

The category 'B' villages, finally 22 in number, were supposed to be reconstructed in-situ. However, almost all the villages opted for the construction of a new house on a different site. The beneficiaries selected an NGO and authorised it to receive their assistance from the government and reconstruct their houses.

Figure 3.2: Attributes of New Houses 80 70 60 50 I 40

Downloaded by [University of Defence] at 01:15 24 May 2016 u CO 30 20 10 0 A B C D E F Attributes Source: CSSS (1999). Notes: A: Safety of house; B: Quality of construction; C: Area of house; D: Built-up area; E: Sturdiness in construction; F: Earthquake resistance. 62 Ú Krishna S. Vatsa

Th e survey of sample households in this category indicated that reconstruction on a diff erent site was the result of a decision taken by the majority to build houses elsewhere, based on their fear of earthquakes and inadequate space available near the old house. Unlike ‘A’ category, where the government acquired land for relocating people, the policy of in-situ reconstruction precluded any provision for land. Th e benefi ciaries therefore used their own land or asked the NGOs to pay for the land for reconstruction. Once a new house was constructed, benefi ciaries (93.59 per cent) abandoned their old houses completely. Th ese fi ndings clearly show their preference for new settlements over the old ones. In category ‘B’ villages, all the benefi ciaries got money enough to construct just a core house of 250 sq. ft. Th e houses constructed in these villages resembled those in category ‘A’ villages. However, the civic amenities in these villages were made available on a much smaller scale.

Reconstruction, Repairs and Strengthening: ‘C’ Category

‘C’ was an owner-driven category in which, with government assistance that was composed of cash and materials, the benefi ciaries reconstructed, repaired and strengthened their houses themselves. Around 95 per cent of the benefi ciaries utilised the assistance fully for the intended purpose. Th e number of defaulters, who took assistance and did not utilise it, was about 5 per cent. In the ‘C’ category, benefi ciaries overwhelmingly preferred construction of additional rooms adjoining their old house, over the retrofi tting option. Only 6 per cent of the total number of benefi ciaries who reported undertook the retrofi tting of their old house. Th is became a somewhat contentious issue in the beginning, because the government was keen

Downloaded by [University of Defence] at 01:15 24 May 2016 on supporting the retrofi tting of old houses. Th e Ahmedabad Study Action Group (ASAG), an NGO, launched a strong advocacy campaign in favour of retrofi tting, but the benefi ciaries insisted on exercising their choice in constructing additional rooms. Th ere were several reasons for this cited by the surveyed households. Prominent among them were technical diffi culties (58 per cent) and expenses involved in retrofi tting an old house (44 per cent), inadequate living space (50 per cent), and feeling of insecurity in the old house (25 per cent). As close to 190,000 benefi ciaries spread over 1,500 villages in these two districts constructed Earthquake Reconstruction in Maharashtra Ú 63

new rooms with the help of programme assistance in a span of two years or so, the housing stock in the entire area was almost renewed. Th e programme assistance worked as a catalyst for a large number of benefi ciaries. Th ey supplemented the assistance with their own resources in cash, material and labour (see Table 3.8). Even if we ignore the fourth category for being unspecifi c, the per-household investment in the process of reconstruction was substantial. Th is showed that the benefi ciaries looked upon the programme as a signifi cant opportunity for rebuilding and expanding their houses. In category ‘B’ and ‘C’ villages, the level of benefi ciaries’ satisfaction with respect to the quality and safety of construction was very high. In category ‘B’, 78 per cent of the benefi ciaries were satisfi ed with the construction undertaken by the NGO, while in category ‘C’, 95 per cent reported overall satisfaction about the construction/repair work undertaken of their houses. Most of them attributed the use of cement and earthquake- resistant technology as major reasons for the safety of these houses.

Table 3.8: Contributions Made by Beneﬁ ciaries

Total amount Average Nature of contribution Number (Rs) (Rs) Contribution in cash 4,553 42,095,431 9,246 Contribution of material 2,422 18,243,411 7,532 Contribution of labour 1,590 4,893,605 3,078 Nature of contribution not mentioned 811 10,048,605 12,390 Total 75,281,052 32,246 Source: CSSS (1999).

Asset-Building and Income Generation

The survey collected information on various assets of households — before and after the earthquake. Since economic assistance for recovery as well as ex gratia payments for deaths was released mostly to

Downloaded by [University of Defence] at 01:15 24 May 2016 households in the category ‘A’ villages, the information available is only in respect of category ‘A’ villages. Th e total number of assets included in the survey is 43, which could be classiﬁ ed into the following four groups (see Table 3.9):

(i) Agricultural assets (ii) Livestock (iii) Household appliances (iv) Vehicles 64 Ú Krishna S. Vatsa

Table 3.9: Change of Agricultural Assets with Households

Description of assets Before earthquake At present Percentage change Land Land under cultivation (ha) 49,092 47,861 –2.51 Agricultural implements Wooden ploughs 2,722 3,653 34.20 Iron ploughs 4,325 6,687 54.61 Tifans (three-bowl seed drills) 5,522 7,143 29.36 Kolpes (weed removers) 9,948 14,715 47.92 Auts (drillers) 5,408 7,689 42.18 Diesel pump sets 125 108 –13.60 Electric pump sets 2,476 2,838 14.62 Bullock carts 3,175 3,078 –3.06 Tractors 69 108 56.52 Trolleys 74 124 67.57 Malanis (harvesters) 160 234 46.25 Irrigation sources Sprayers 1,777 4,487 152.50 Sprinklers 44 68 54.55 Wells 2,676 2,554 –4.56 Bore wells 706 1,106 56.66 Source: CSSS (1999).

Results in respect of agricultural and livestock assets are mixed. Th e percentage of land under cultivation has declined, which could be attributed to land acquisition on a large scale for the relocation of villages. As agriculture is the mainstay of the local economy, the loss of cultivable land may produce a negative impact. However, considering almost 28,000 families in 52 villages were relocated on new lands, it was a relatively small loss of cultivable land. Th e number of agricultural implements, on the other hand, increased signiﬁ cantly, which could be attributed to the distribution of these implements as part of the economic assistance provided. Th e number of

Downloaded by [University of Defence] at 01:15 24 May 2016 diesel pumps decreased, but was substituted by electric motors. Similarly, the decline in bullock carts was compensated with an increase in the number of tractors, and open dug-wells by a considerable rise in bore wells. It suggested a change in the asset profi le, with households opting for modern and energy effi cient equipment and implements. Th e number of livestock has been arrived at by adding the total number reported by all the respondents in the survey. Th e cumulative fi gures in Table 3.10 show that in the new villages there was a decline in the ownership of livestock assets. Th is could be explained by families staying Earthquake Reconstruction in Maharashtra Ú 65

Table 3.10: Change of Livestock Assets with Households

Description of assets Before earthquake At present Percentage change Bullocks 12,382 11,348 –8.35 Cows 7,687 6,507 –15.35 Improved cows 1,287 1,762 36.91 Buﬀ aloes 7,616 7,423 –2.53 Other cattle 2,840 4,017 41.44 Sheep and goat 20,011 13,136 –34.36 Horses 54 31 –42.59 Pigs 1,338 378 –71.75 Donkeys 139 118 –15.11 Poultry 18,572 8,702 –53.14 Source: CSSS (1999).

in temporary sheds for a long time, which did not provide much space for keeping livestock. In the new villages there has been a certain amount of emphasis on hygiene and sanitation through community education, which may have discouraged the rearing of animals in their houses. Th e number of improved varieties of cows, distributed as part of economic assistance, has however increased. Th ere has been a signiﬁ cant increase in the ownership of household appliances and consumer durables (data presented in Table 3.11). Most noticeable among these appliances is rise in ownership of black-and- white and colour TVs and video cassette recorders (VCRs). Further, the individual ownership of vehicles has gone up (see Table 3.12). Th ough the number of these vehicles is still small, it suggests that these households value their mobility. In addition to economic assistance, a large-scale reconstruction programme has also helped in the circulation of a large

Table 3.11: Change of Household Appliances with Households

Description of assets Before earthquake At present Percentage change Iron cots 9,943 11,596 16.62 Downloaded by [University of Defence] at 01:15 24 May 2016 Steel cupboards 1,360 1,746 28.38 Radios/tape players 1,963 2,958 50.69 Sewing machines 933 1,121 20.15 Colour TVs 271 437 61.25 B/W TVs 1,268 2,644 108.52 Cooking gas connections 180 242 34.44 Electric mixers 415 663 59.76 Emergency lights 486 409 –15.84 VCP/VCR 20 34 70.00 Source: CSSS (1999). 66 Ú Krishna S. Vatsa

Table 3.12: Change of Vehicle Ownership with Households

Description of assets Before earthquake At present Percentage change Bicycles 4,189 5,231 24.87 Mopeds/scooters 162 216 33.33 Motorcycles 248 371 49.60 Rickshaws 32 47 46.88 Jeeps 55 76 38.18 Tempos 10 10 0.00 Trucks 30 43 43.33 Source: CSSS (1999).

amount of money in these two districts. Th e trickle-down eff ect of the investments in reconstruction can be seen in the increased ownership of these assets. Further, it also suggests an improvement in the quality of life for most households. One of the most important variables is the impact on household income. Th e programme provided the aff ected people with economic assistance, mainly to resume agricultural operations and trade-based occupations. Th e objective of providing assistance was to help these households revive economic activities and sustain themselves fi nancially. Th e benefi ciaries in ‘A’ category villages reported a change in income as a result of the implementation of MEERP, summarised in Table 3.13. Th e programme did not produce signifi cant income gains for the benefi ciaries. A number of them have maintained their income levels,

Table 3.13: Change in Income Levels of Households

Number of households (percentage in parentheses) Description Increased Unchanged Decreased Total Change in income from 80 931 771 1,782 profession/self- (4.49) (52.24) (43.27) (100) employment, if continued Change in income 122 416 498 1,036 Downloaded by [University of Defence] at 01:15 24 May 2016 from dairy (11.78) (40.15) (48.07) (100) Change in employment 2,804 3,504 3,460 9,768 opportunities as agriculture (28.71) (35.87) (35.42) (100) labour Change in employment 2,019 1,951 3,152 7,122 opportunities as (28.35) (27.39) (44.26) (100) other labour Change in overall employment 3,111 7,716 5,514 16,341 opportunities (19.04) (47.22) (33.74) (100) Source: CSSS (1999). Earthquake Reconstruction in Maharashtra Ú 67

and a small percentage have reported an increase in their income. Still, there is a large number of benefi ciaries who have reported a decrease in income from all sources. Th ough the project may not have contributed directly to a decline in income, there are reasons for this which need to be explored. A large number of families in the category ‘A’ villages lost their earning members in the earthquake. Many households were out of the labour market as a result and must have experienced a drastic loss in income. Th ere could be other reasons for the decline in income: loss of agricultural land, distance from the newly relocated villages, reduced availability of agricultural labour due to construction activities, loss of livestock, trauma and psychological stress — though it is diffi cult to say what contributed most. According to the EPW Survey, the boom in construction activities in the area created an unprecedented demand for construction labour. While the well-to-do villagers, including the upper castes, refused to accept wage labour in construction work because they felt that it was beneath their dignity, the poor responded to the demand for construction labour mainly to supplement their meagre family incomes. Th e shift of rural labour to house repairs/construction created a temporary shortage of agricultural labour, pushing up agricultural wages, which benefi tted those who continued to work as agricultural labourers. Th e level of employment and wages for the poorer households was thus higher for about 4–5 years, largely due to the demand for employment generated by the MEERP. However, the demand for construction and agricultural labour was limited to the landless communities, which derived their livelihood from performing casual labour. Other segments of society which derived their income from other sources did not benefi t from the project.

Equity and Gender Downloaded by [University of Defence] at 01:15 24 May 2016 One of the most important yardsticks by which a rehabilitation programme can be judged is how it has benefi tted the aff ected region’s weaker classes and castes. In Marathwada, where caste hierarchy and social inequities are deeply entrenched, the issue of equity in programme implementation is even more critical. Th e survey looked at the social distribution of programme benefi ts by applying three categories: caste, landholding and occupation. A cross-tabulation of old and new houses, based on these categories is presented in Table 3.14. 68 Ú Krishna S. Vatsa

Table 3.14: Classiﬁ cation of Households According to Caste and Type of Old and New Houses (%)

Castes Upper Scheduled Other Scheduled Nomadic Types of houses castes castes backward castes tribes tribes Old Houses Huts/tin sheds 21 60 38 68 54 Wadas 50 20 33 19 23 Independent bungalows 29 20 29 13 23 Total 100 100 100 100 100 New Houses Type ‘A’ 32 80 57 47 60 Type ‘B’ 56 19 39 49 37 Type ‘C’ 12 1 4 4 3 Total 100 100 100 100 100 Source: CSSS (1999).

Th e survey shows that a majority of the scheduled castes, scheduled tribes and nomadic tribes lived in huts and tin sheds before the earthquake. Under the MEERP, these sections came to own new houses along with plots of land — defi nitely a vast improvement upon the older structures. Since the scheduled castes formed one of the largest groups of benefi ciaries, they were also the biggest gainers from among the weaker sections. Nomadic tribes living in Laman Tandas, traditionally settled outside the village, constituted the second most signifi cant group of benefi ciaries.10 Th ough it is a matter of subjective preference if the houses built under MEERP are an improvement upon the wadas or not, but given an option, most of the households would opt for independent houses as built under the programme. Among the occupational groups (data presented in Table 3.15), agricultural labour, other labour and the self-employed have been the biggest benefi ciaries of the housing programme. Most of them had earlier Downloaded by [University of Defence] at 01:15 24 May 2016 lived in huts and tin sheds, and after the earthquake they came to own independent houses. A signifi cant section of these groups came to occupy type ‘B’ houses, a category dominated by agriculturists. Th e diff erence between these groups, irrespective of their property, is not glaring. As

10 Laman Tandas are settlements of a nomadic tribe known as Laman. Th ese Tandas are situated at some distance from the main village, showing the Laman’s social exclusion. Earthquake Reconstruction in Maharashtra Ú 69

Table 3.15: Classiﬁ cation of Households According to Occupation and Type of Old and New Houses (%)

Castes Agriculture Other Self- Types of houses Agriculture labour labour employed Service Old Houses Huts/tin sheds 17 56 51 39 25 Wada 50 25 28 37 45 Independent bungalows 33 19 21 24 30 Total 100 100 100 100 100 New Houses Type ‘A’ 13 78 76 68 46 Type ‘B’ 70 22 22 30 47 Type ‘C’ 17 Negligible 2 2 7 Total 100 100 100 100 100 Source: CSSS (1999).

all the households in these villages came to own houses, and diff erences in their entitlements was well within reasonable proportions, it could be argued that the project succeeded in protecting the entitlements of the weaker sections of society. Th e survey conducted by EPW also confi rms that agricultural labourers and marginal farmers, who together account for between 50 to 60 per cent of the rural population, had benefi tted relatively more than others from the programme. According to this survey, poorer groups have benefi tted in many other ways as well, apart from being the proud owners of independent houses. First, as a large number of people from the weaker sections lived in structures made of light material such as bamboo, thatch and other agricultural waste materials, they lost much less in the earthquake compared to those who had used stones and mud mortar in their houses. Second, in addition to the house and plot

Downloaded by [University of Defence] at 01:15 24 May 2016 that these people received, they also gained access to a number of civic amenities such as drinking water supply, electricity, medical centres, etc., which were not available to them earlier. Finally, having occupied new houses alongside the better-oﬀ among the villagers, the poorest, most of whom were living in the periphery of villages in the pre- earthquake period, were now brought into the mainstream of rural life, breaking their age-old isolation (EPW Research Foundation 1998: 126). How have women beneﬁ tted from the programme? Th e results for women are somewhat mixed (see Table 3.16). Th e programme has not 70 Ú Krishna S. Vatsa

been helpful in increasing their prospects for employment. Th ey have also found it diffi cult to fetch water, and do household chores which are related to the availability of water. With gradual improvement in water supply, many of these diffi culties are expected to be alleviated. Due to the lack of suffi cient vegetation in and around the new settlements, women have also found it diffi cult to collect fuel and fodder. On the other hand, women now have access to better social services, including education and healthcare. Th e provision of a number of civic amenities in the relocation villages has rendered these services within the reach of women. Th ey are more exposed to modern conditions of living, and fi nd it easier to clean and maintain their houses. Th ough women’s participation as a group in category ‘A’ and ‘B’ villages was limited, they played a signifi cant role in category ‘C’ villages. Around 7 per cent women from beneficiary households worked as Samvad Sahayaks (communications facilitators) in implementing this programme. A majority of the women (72 per cent of total respondents)

Table 3.16: Post-Rehabilitation Status of Women

No. of households Condition after rehabilitation Description Reporting Worse Same Better Employment opportunities 14,648 32.52 46.68 20.79 for women Self-employment 11,816 37.75 48.69 13.57 opportunities for women Earnings of women 16,582 39.07 46.35 14.58 Condition of women: 17,620 51.05 35.73 13.22 fetching water Condition of women: 17,248 33.31 60.79 5.90 collecting fuel/fodder Facilities for women: 17,261 10.31 34.35 55.33 education Facilities for women: 17,678 6.82 39.73 53.44 Downloaded by [University of Defence] at 01:15 24 May 2016 medical Facilities for women: 17,796 15.16 23.21 61.63 toilet Facilities for women: 16,253 20.11 64.11 15.78 marketing Facilities for women: 14,822 14.61 53.58 31.81 government schemes Women’s exposure to 16,656 6.64 45.31 48.05 modern living Source: CSSS (1999). Earthquake Reconstruction in Maharashtra Ú 71

contributed to the repairs and strengthening programme by bringing water and sand to the construction sites. About 22 per cent women contributed labour to actual construction/repairs of their own houses. Combined with many activities such as the recruitment of women junior engineers, training of women masons and the formation of self-help groups, the programme had a positive impact on the empowerment of women.

Conclusions

On the basis of the benefi ciaries’ responses, as analysed above, the achievements of the MEERP could be summarised here. In terms of housing, the programme brought substantial gains to the benefi ciaries. Before the earthquake, most people had occupied old houses, with stone and mud walls, and wooden/tin roofs. Th ey had no toilets, although some did have bathrooms. Dependence on bore-wells for drinking water was substantial. Th e newly built houses under the MEERP are clearly an improvement upon them. Th e built-up area is generally smaller, but the use of cement and concrete in the construction of fl oors, walls and roofs has provided greater structural safety to these houses. Almost all the new houses are equipped with their own bathrooms and independent toilets. It was the regional water supply scheme that was to provide water to the new villages through individual connections or as common standposts that turned out to be the weakest aspect of programme implementation. Many villages did not receive water for a long time, limiting the use of new facilities like toilets and increasing diffi culties for women. Even the layout of the new villages and the design of amenities were an improvement upon the old villages. Due to the increase in family size and the limited space available in the old villages, they were cluttered and congested, with inadequate sanitation and poor hygiene. Most of

Downloaded by [University of Defence] at 01:15 24 May 2016 the new villages were settled over an area eight times that of the old villages, with internal roads and drainage. Out of the total number of respondents, 77 per cent in fact favoured the new layout plans over the old village. One of the most positive outcomes has been the provision of several civic amenities. Th e beneﬁ ciaries have responded to this change very favourably. Some of these amenities such as a Mahila Kendra and a gymnasium have been provided for the ﬁ rst time. Th e MEERP brought about an overall increase in almost all assets at the household level. Th e programme proved to be a catalyst in bringing 72 Ú Krishna S. Vatsa

about a clear shift from possessing traditional means of production (bullock carts, diesel pump sets, and open dug-wells) to modern ones like tractors, electric pump sets, cross-bred cows, etc. For instance, the use of tractors, iron ploughs, trolleys, and sprayers have increased by 56 per cent, 55 per cent, 68 per cent and 153 per cent respectively in the post-resettlement period. Th e number of almost all domestic assets has increased. While the level of assets ownership has gone up, employment and income generating opportunities have shown a decline. Th e fall in income from agriculture and dairy, the two most common sources of livelihood options in the region, refl ects adversely upon the programme. Th ough the dislocation of economic activities due to the earthquake may be responsible for this, it is also true that the programme has not succeeded in addressing this aspect, despite helping households with agricultural implements and livestock. Socially and economically weaker sections of the community have gained signifi cantly from MEERP. Almost 37 per cent of the households who reported that they were staying in huts or tin sheds before the earthquake, are now staying in newly constructed houses. In fact, 27 per cent of the households who were staying in a hut or tin sheds earlier now occupy ‘B’ and ‘C’ type houses with areas of 450 sq. ft. and 750 sq. ft., respectively. For women, the MEERP has brought mixed results. Th ey have found it easier to clean and maintain their new houses. Th ey also have better access to education and health services. However, their diffi culties have increased with respect to the fetching of water and fuel, maintaining cattle and walking to work. Employment opportunities for women have also, according to women respondents, not changed for the better. CSSS asked the benefi ciary households surveyed to give a rating to the reconstruction and rehabilitation programme. Th e latter assigned a score on a scale of 10. For each village, these scores were averaged and a

Downloaded by [University of Defence] at 01:15 24 May 2016 composite average was calculated. Th e lowest score that a village received was 3.9 (Gubal), with the highest score being 9 (Toramba). Around 5 per cent of the respondents gave the lowest value zero, while 13 per cent gave the highest rating of 10. As many as 43 per cent respondents gave a rating of 8 or more. Th ere are village-speciﬁ c reasons for the variation between village scores, which need to be analysed for a more detailed picture. For instance, in all the villages where the water supply scheme had not become operational at the time of the survey, the scores were relatively low. Th e number of villages in diﬀ erent score intervals, from 1 to 10, are Earthquake Reconstruction in Maharashtra 73

shown in Figure 3.3. Between the scores of 5 and 8 are placed 40 of the 52 villages in total. The average rehabilitation score for all the relocated villages worked out to be 6.3. Though it is not a very high score for category 'A' villages, it shows the overall satisfaction of the beneficiary households with the quality of reconstruction and rehabilitation.

Figure 3.3: Scores for Relocation Villages in Category 'A' 18 16 14 12 > 10 8 I 6 2 4 2 0 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 Scores Source: CSSS (1999).

Though assistance in 'B' and 'C category villages was limited to housing, the level of beneficiaries' satisfaction in these villages has been higher than that of category 'A'. In category 'B', the beneficiaries insisted upon reconstruction on a different site in most of the villages. The direct participation of beneficiaries in the decision-making and implementation processes was more helpful in creating satisfaction among them, though these villages did not get the civic amenities or the infrastructure made available to category 'A' villages. In category 'C, owner-driven reconstruction provided direct opportunities to the beneficiaries to Downloaded by [University of Defence] at 01:15 24 May 2016 participate in the programme. In terms of the beneficiaries' perception, it emerged as the most successful component of the MEERP. A large number of the beneficiaries contributed their own resources to what was provided by the government. Morover, they also internalised the application of earthquake-resistant technology. Most of the villagers opted for the reconstruction of additional rooms over retrofitting. Since the programme was undertaken on a very large scale, it led to widespread renewal of housing stock in the entire region. 74 Ú Krishna S. Vatsa

Emerging Policy Issues

A post-disaster reconstruction programme provides a unique opportunity to improve the quality of life and usher in development. Th e large- scale investments that follow a disaster can revive the local economy by providing employment in the construction and service sectors. It can improve access to housing, its quality and structural safety, as well as social and community infrastructures. Peoples’ participation in reconstruction programmes provides them motivation and self- confi dence, and strengthens community networks. A recovery and reconstruction ‘decision tree’ has been presented in the Annexure (see p. 77). Th e shaded areas represent the decision points for a government considering its response to a disaster. At each of these decision points, the government should ask some basic questions regarding the objectives of its intervention, its capacity to undertake programming and its strategy for resource mobilisation. Th e ‘decision tree’ has been structured in such a way that choices that require less government intervention are on the left and those that require more intervention are on the right. However, the process is never as linear as the ‘decision tree’ suggests. In a post-disaster situation, the government may plan for diff erent phases at the same time. A ‘decision tree’ will help the government in enhancing the depth of programming and providing a more comprehensive view of recovery and reconstruction than what is generally understood. When a reconstruction programme is launched, the immediate objectives are to provide shelter to the people and, as soon as possible, restore the pre-disaster situation. Th is is understandable in view of the dislocation and disruption caused by a disaster. Yet, a more long-term vision of development and better human settlements could always be integrated and pursued in a post-disaster reconstruction programme. Th e MEERP suff ered from the lack of balance between the immediate Downloaded by [University of Defence] at 01:15 24 May 2016 priorities of housing and the long-term goal of development. Th e housing programme emerged as the most prominent aspect, while activities aimed at increasing income and livelihood did not receive considerable attention. Similarly, although the educational and health facilities were restored in the relocated villages, there was no real improvement in the quality of these services. In all reconstruction programmes, which can be seen as opportunities for regional development, the need to upgrade educational and health services can hardly be overstated. Earthquake Reconstruction in Maharashtra Ú 75

A reconstruction programme also serves to bring people together. It provides an opportunity for a number of community-based activities. Under the MEERP, several activities were undertaken for organising women. A large number of self-help groups were formed, savings were encouraged and training programmes were organised. Similar activities could have been undertaken for water conservation and social forestry too, and could have changed the landscape of these villages considerably. Instead, the programme spent a great deal of resources on the regional water supply scheme, an ambitious attempt at supplying drinking water to these villages. It prevented local ownership of the scheme, and created a number of problems in implementation and operations. Smaller programmes at the village level for water supply and conservation would have been more successful. Under the MEERP, beneﬁ ciaries were provided assistance without the expectation of any contribution from them. In retrospect, this was not the best strategy. In category ‘C’, where the villagers contributed their own resources towards reconstruction, because assistance was smaller, the results were much better. Households ended up renovating their houses completely by pooling their own savings. Th e government’s assistance worked primarily as a catalyst. In categories ‘A’ and ‘B’, households did not contribute their resources, either in terms of money or household labour. Th is did not increase community ownership of the programme, nor did it help them enhance the asset value of their houses and community infrastructure. With respect to housing, the implementation of the MEERP raised a number of issues. One contentious point has been the relocation of villages (‘A’ and ‘B’ categories) versus in-situ reconstruction (‘C’ category). In-situ reconstruction is generally a preferred choice since it causes less dislocation, ensures continuity of community life and avoids consuming new land. But relocation is at times unavoidable due to considerations of decongestion, difficulties in removing debris or

Downloaded by [University of Defence] at 01:15 24 May 2016 problems of psychological trauma. Relocation also provides the advantage of a better layout and architecture, engineered construction, and more civic amenities. An ideal solution for a reconstruction programme is one which combines positive features of both the options, which means in-situ reconstruction in a decongested area, with a bigger plot, better layout and architectural design, and stronger foundations. It is diﬃ cult to achieve such a balance when a reconstruction programme is undertaken on a large-scale, but it can be facilitated through micro- planning at the village level. 76 Ú Krishna S. Vatsa

Second, there is a debate over construction strategy — between construction through contractors (‘A’ category) and owner-driven construction (‘B’ or ‘C’ category), which is preferable? MEERP pursued a combination of both strategies. While, theoretically, owner-driven construction is a better choice, with better utilisation of resources and greater supervision, it is a feasible strategy only when the households are not hard-pressed to fi nish it within a short timeframe. In cases where the construction is to be done at faster pace or when the households are constrained for time due to their professional commitments, construction through contractors, despite its limitations, is a more practical strategy. Th e third issue has been retrofi tting versus reconstruction. In category ‘C’ villages, most of the benefi ciaries preferred construction of one or two rooms over retrofi tting their old houses. Retrofi tting involves complex techniques and requires closer supervision, and may not be feasible in a large-scale programme, and the households also wanted increased living space for their families through reconstruction. A reconstruction programme in the wake of an earthquake tends to emphasise seismic engineering. While it is important therefore to include earthquake-resistant features, it is even more important to focus on resettlement planning and architecture. Reconstruction programmes generally ignore this, which does not help in improving the overall quality of the habitat. Within its broad framework, the MEERP pursued diff erent strategies, showed great fl exibility and innovation, and provided multiple options to its benefi ciaries. It also has several lessons to off er. Would it have been possible for any programme to demonstrate these strategies and innovations had it not been fi nanced and supported by the World Bank? Th is is the biggest challenge for any post-disaster reconstruction programme, particularly for a developing country like India. All the reconstruction programmes that have been implemented in India Downloaded by [University of Defence] at 01:15 24 May 2016 without the World Bank’s support have been narrow in their scope and weak in terms of technology, procurement methods and implementation. It would be most important to include these technical and institutional aspects, as discussed above, without the external support of a bilateral or multilateral donor agency so as to off er wider choices and alternatives to the aff ected households in keeping with their needs. Downloaded by [University of Defence] at 01:15 24 May 2016

Annexure

Figure 3.4: Recovery and Reconstruction Tree

Disaster event

1. Organise response and relief. 2. Conduct damage assessment. 1. Leave post-disaster care and relief to Emergency 3. Set up a post-disaster programme. NGOs and private households. phase 60 1. Include economic and business activities in Days restoration programme. 2. Mobilise resources through government, 1. Restore all public utilities donors and insurance. Restoration 3. Fix entitlement for disaster assistance. and civic infrastructure. phase 100 4. Seek participation of citizens' groups, NGOs and private sector. days

1. Set up a reconstruction programme. 2. Provide assistance / subsidy / building materials 1. Organise loans from financial to households and businesses. Replacement institutions for households and 3. Institutionalise community participation. and businesses for reconstruction. 4. Provide technical assistance for disaster-resistant 2. Facilitate insurance payment for reconstruction Ito2 construction. insured entities. phase years 5. Legislate for building code and land use plan. 6. Revive economy through local resource investment and employment programmes.

1. Leave it to individual 1. Invest in area development programmes. households and the private sector. Betterment 2. Build disaster-resistant critical infrastructure. 2. Ask individual households and development 3. Invest in microzonation, flood plain private business to take period 2 to 5 management and building code regulations insurance against natural disasters. years 4. Develop local-level capacity in building technology and disaster management. 78 Ú Krishna S. Vatsa

References

Centre of Studies in Social Sciences (CSSS). 1999. Survey of Rehabilitated Households aff ected by Killari Earthquake (Latur and Osmanabad Districts), Vols I & II. Pune: CSSS. Damle, Jasmine Y. 2001. Beyond Economic Development: A Case Study of Marathwada. New Delhi: Mittal Publication. Earthquake Engineering Research Institute (EERI). 1999. Lessons Learned Over Time, Innovative Earthquake Recovery in India. Oakland, California. Earthquake Rehabilitation Cell. 1999. Quarterly Progress Report-22 (Final). Mumbai: Maharashtra Emergency Earthquake Rehabilitation Programme. Economic and Political Weekly (EPW) Research Foundation. 1998. Marathwada Earthquake, A Documentation of the Benefi ciaries’ Experiences: A Survey Report. Maharashtra Earthquake Rehabilitation Documentation Project, Report No. 7. Government of India (GoI). 1996–2002. Census of India, 1991. Delhi: Controllers of Publications, GoI. Government of Maharashtra (GoM). 1994. Earthquake Rehabilitation Policy. Bombay: Earthquake Rehabilitation Cell, Mantralaya. Momin, S.S., S. Nikolic-Brzev and K.M. Bajoria. 1996. ‘Seismic Retrofi tting of Stone Masonry Buildings Damaged in the September 1993 Earthquake in India’, Paper No. 1389, proceedings of the Eleventh World Conference on Earthquake Engineering, 18–23 June, Acapulco, Mexico. Twigg, John. 2002. Technology, Post-Disaster Housing Reconstruction And Livelihood Security, http://livelihoodtechnology.org (accessed on 17 February 2003). Downloaded by [University of Defence] at 01:15 24 May 2016 4 A Decade of Lessons from Marathwada Earthquake Vulnerability, Politics and Participatory Housing Alex Salazar and Rohit Jigyasu

Disasters can inspire the most heroic of eff orts and off er a window of opportunity for people to transform society. Th is is especially true of disasters that directly and dramatically aff ect the built environment, which is when urban policies themselves are brought into question. Here, previously inappropriate land use decisions, the poor enforcement of building codes, ineffective public planning methods, and media attention to such stories can push decision makers into taking rectifi catory action. Also, it is here that community-based organisations (CBOs) and non-governmental organisations (NGOs), if given the political space, can mobilise grass-roots support to demand changes that will truly address the needs of the worst aff ected. Unfortunately, such attempts are hard to accomplish and are often sidelined by cold political calculations, the pressure for decisive action and instant results, and the fi nancial needs of developers and banking institutions — with only lip service being paid to the idea of ‘peoples’ participation’. Over the last 30 years this has been the norm, not just after disasters but in a variety of development activities, such as the construction of dams, aqueducts, frontier development, etc., resulting in the faulty or unnecessary relocation of perhaps tens of thousands of villages and towns. The negative physical, economic and sociocultural impact of relocation on rural communities have long been Downloaded by [University of Defence] at 01:15 24 May 2016 understood, and are best described in the collection of essays edited by Hansen and Oliver-Smith (1982), Bruce Rich’s critique of World Bank policies (1994), a collection of disaster conference papers edited by Ian Davis (1981), and Aysan and Oliver’s paper on post-disaster housing policy (1987). Many of the relocation projects mentioned in these other writings are well-known failures today: abandoned or never occupied as people returned to rebuild their houses near the original settlement sites. 80 Ú Salazar and Jigyasu Citation Oliver-Smith (1982) (1990); Oliver (1987) et al. (1990) et al. (1990) (1984), cited in Coburn et al. (1990). in Davis (1981) Oliver (1992); (1987) Oliver-Smith (1982); Oliver- Smith (1986); Oliver-Smith (1990); Oliver-Smith (1994) Palacio, in Hansen and Kreimer and Echeverria Ladinski, in Awotona (1997) Parsa (1985), cited in Coburn Rezani (1984), cited in Coburn Coburn, Leslie and Tabban Davis (1981); Caldwell et al. Oliver, in Davis (1981) Oliver, in Davis (1981) Aysan and Oliver (1987); Oliver-Smith, in Hansen and Davis (1981) Davis (1981) cantly Synopsis of relocation problems return to old Managua City; 70 per cent abandonment rate at some sites, such as domes, due to culturally inappropriate house forms. ect on the rural building economy eff ect on the rural building economy eff abandoned people continue to live in temporary house additions in-situ changed by occupants and, in some cases, abandoned priate temporary sheds turned into long-term housing; physical vulnerability linked to 500-year process of sociocultural change due to Spanish colonisation Culturally inappropriate housing; despite relocation 1977 Table 4.1: Sample List of Problematic Post-Disaster Relocation Projects, Prior to 1980 Downloaded by [University of Defence] at 01:15 24 May 2016 Earthquake 1963Earthquake Culturally inappropriate relocation housing 1972Earthquake 1973 Urban-based contractor-built housing has a detrimental Technologically and cultural inappropriate housing, often Earthquake Not available Despite attempt to relocate villages, the villagers rebuild Not available Not available New relocation housing dismantled for sale in the marketEarthquake Chisholm, in Davis (1981) 1970Hurricane Relocation to faraway sites rejected; culturally inappro- Not available Culturally inappropriate relocation housing Not available Not availableCyclones Culturally inappropriate relocation housing 1968 and Not available Not available Unsuccessful relocation Type of disaster Year Macedonia Turkey, circa 1973 Carusin, Turkey India Managua, Nicaragua Earthquake 1972Skopje, Republic of Khorasan, Iran Relocated families abandon new housing units and Qir, Iran EarthquakeBingol, Eastern 1978 Urban-based contractor-built housing has a detrimental Central Anatolia, Gibellina, Sicily, ItalyBangladesh EarthquakeGediz, Turkey 1968Peru Earthquake Culturally inappropriate relocation housing 1970Belize Culturally inappropriate relocation housing is signifi Lice, Turkey Andhra Pradesh, Location Greece Lessons from Marathwada Ú 81

After the September 1993 earthquake, in the core disaster-aff ected area of Marathwada, projects followed much the same pattern. While measuring only 6.4 on the Richter scale, the earthquake destroyed thousands of stone masonry homes, and left over 9,000 people dead and many more homeless. By May 1994, with the aid of the World Bank, the Government of Maharashtra (GoM) launched the Maharashtra Emergency Earthquake Rehabilitation Project (MEERP) — a US$ 326 million aid programme that aff ected over 264,500 households in 13 agricultural districts. Initially there were high hopes, especially given the participatory approach to housing design and the promise of using sustainable building practices. By 2001, however, the projects appeared to have resulted in many negative impacts. During fi eld visits to dozens of relocation projects, a majority of the homes appeared to only be used for storage purposes, and in some villages, like Killari, entire neighbourhoods lay completely abandoned. And in the existing villages, where households were to have ‘retrofi tted’ vernacular buildings, over 99 per cent of the work was in the form of new concrete and brick additions (Nikolic-Brzev et al. 1999). Th is chapter unravels how these policy failures took place, which in turn will help us better understand the impact of reconstruction on changes taking place in the social, cultural and environmental patterns in Marathwada. Part I provides the background information. It outlines a vulnerability approach to understanding the disaster, and examines how social and economic changes over time helped marginalise artisan castes and undermine vernacular building practices prior to the earthquake. Second, it describes the international policy context up to 1994, specifi cally with regard to village relocation and reconstruction in-situ work after disasters. Th ird, it analyses political infl uences on MEERP’s participatory housing policies and highlights how misinformation and social pressures marginalised alternative NGO projects from having any infl uence in Marathwada. Part II describes and assesses the outcome Downloaded by [University of Defence] at 01:15 24 May 2016 of MEERP housing policies and focuses on three major programmes: village relocation; in-situ reconstruction; and RRS, that is, reconstruction, repair and seismic retrofi tting. Part III returns to the topic of vulnerability and summarises how the MEERP reinforced the social and economic vulnerability of artisans and local builders in the region, thereby contributing to unsuccessful disaster mitigation measures. And fi nally, in conclusion, Part IV summarises the major points of this chapter and 82 Salazar and Jigyasu

Map 4.1: Location Map of the Marathwada Earthquake

MADHYA PRADESH GUJARAT

DADRA& NAGAR HAVELI M.J^H A^R/A & H t

Mumbai ANDHRA ARABIAN PRADESH

Latur district

KARNATAKA Relocated village area Osmanabad GOA 0 100 200 km Source: Drawn by Alex Salazar.

offers a set of general guidelines for building post-earthquake housing in rural India, policies that would discourage unnecessary village relocation work, support collaborative government-NGO-CBO rehabilitation projects, and promote the meaningful involvement of local artisans and builders.

Part I: The Background

Established disciplines and dominant institutions of government have chosen to treat [disasters] not as crises of modernity or the predicaments modernity creates on the ground, nor as failures of a research paradigm or policies and organization. Rather ... these hazards are placed, intellectually,

Downloaded by [University of Defence] at 01:15 24 May 2016 socially and geographically, at the frontiers, as part of the unfinished business of modernization (Hewitt 1995:117).

The epicentre of the Marathwada earthquake lay along the Terna River, between Latur and Osmanabad — two of the most 'backward', drought- prone districts of the State. Many of the people killed or injured by this moderate tremor of 6.4 were from the Maratha community — a land- owning majority which had become increasingly involved in cash crop production since the mid-1980s. The remaining, approximately Lessons from Marathwada 83

50 per cent of the population, were marginal or landless farmers, or members of the service castes (TARU1993; Parasuraman 1995). Within the latter group were included construction workers — the Suthars (carpenters), the Gawandis (masons) and the Wadars (stonecutters) — who were highly skilled at producing wada-style courtyard homes, a vernacular style that accounted for approximately 80 per cent of all rural housing. In comparison, reinforced concrete accounted for only 2 per cent (TARU 1993). Typically, at the centre or front of the lot were

Figure 4.1: Section and Plans of a Typical Wada House Showing its Use and Expansion Over Time

Legend 2C 4 2B 3 2 House Spaces 1 Dhelaj (entry) 2 Choke (courtyard) 2B Kitchen area 2C Bathing area 2D Cattle Shed 3B 2B 3 Sopa (veranda) 2C 3B Kitchen 4 during monsoon 2 4 Kholi 3 (daily storage) 4B Kholi 4B 2D (longterm storage) Wall Types I I Existing Intermediate Phase Y/////M New

0 1.5 3 6 m

4B !2B [3iF 2B^ 3B 2C 2C

Downloaded by [University of Defence] at 01:15 24 May 2016 4 4 3 2 2 3 1 2D 4B 2D

Late Phase Initial Phase

0 1.5 3 6 m

Source: Drawn by Alex Salazar. 84 Salazar and Jigyasu

Figure 4.2: Typical Village Plan — Babalsur Village, 1994 Spaces and Buildings Extended Family i 1 Compounds I 1 •T Major Public 1 & 2 Spaces

T .-•, 2 Government r — n K r Owned Public r Space/Buildings l J T.

Religious T • i. Temples/Spaces T

Well

Ashram A

Caste/Community

Maratha' * un marked compounds 1 M Harijan H Mang M Lingayat L H Brahmin B N Chambar C Koli K Suthar S Approximate Scale

0 25 50 100 m

Source: Drawn by Alex Salazar.

courtyards meant for keeping cattle, bathing, cooking, storing farm equipment, and so on. Verandas, which were open to these courtyards, were used for keeping cattle, bathing and all kinds of living and working Downloaded by [University of Defence] at 01:15 24 May 2016 activities. And at the very back of the lot were rooms used for the long- term storage of grains and for items of daily use such as food and utensils (TARU 1993; Ahmedabad Study Action Group [ASAG] 1996). Despite the traditional appearance of this housing style, land and housing relationships have changed substantively over the years. During colonial times, under the barter system, artisans exchanged their services for the right to farm small plots of land or receive a portion of the crop as payment. But with progressive, nationalist land reforms in Lessons from Marathwada Ú 85

the 1950s — which turned tenant farmers into landowners — artisans lost their customary rights, and their work shifted mostly to a cash basis (Dadekar 1986: 127). Although no in-depth study has been done, there is evidence to show that socio-economic changes had, over time, contributed to a deterioration in building skills. In 1993, researchers from TARU noted that Wadar households were especially enterprising and had increasingly worked as subcontractors in the formal building industry — not only as stonecutters, but as masons as well. ‘Th eir control over stone-cutting activity’ gave them a ‘comparative advantage versus the Gawandi’ to obtain contracts (TARU 1993, Part III: 13) for the construction of government buildings and house additions. Additionally, the occupational pattern of the Gawandis changed: they got involved in agricultural production in order to be able to support their families, as younger artisans became less willing to take up stone masonry as

Figure 4.3: Temple at Banegaon Village Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Photograph by Alex Salazar. Note: Th is temple survived the earthquake despite the fact that 70 per cent of the homes in Banegaon village had been levelled to the ground. 86 Ú Salazar and Jigyasu

a profession. Th is led some Gawandis to blame the construction of low- quality walls (which used round, improperly bonded stones) not just on households trying to save money but also on the Wadar community, which had encroached upon their traditional domain (ibid.: 12). After the quake there was fi eld evidence of this marked deterioration in building standards: (i) in many villages, older buildings survived the earthquake (including temples and the homes of stone masons) while newer buildings collapsed; and (ii) in relocation projects the older Gawandi artisans understood stone masonry techniques properly, while younger artisans were not just less knowledgeable but also harder to re-train. Th us, one can point to a wide range of issues that prepared the ground, so to speak, for the disaster that followed: post-colonial land reforms, Green Revolution technologies, the integration of some communities into the formal building industry, and so on — all of which transformed and undermined the safety and security of rural households. Contrary to conventional wisdom, which tends to highlight the physical ‘cause’ of disasters (such as an earthquake causing the collapse of stone masonry walls) one can say that historical change, including social, cultural and environmental infl uences, predisposed the Marathwada villages to calamity. Th is perspective is consistent with the vulnerability studies of Kenneth Hewitt (1983 and 1997), Anthony Oliver-Smith (1986), Ann Varley (1994), Haque and Zaman (1994) and others, who demonstrate the historical complexity of disasters and their links to development processes. And it is in this sense, to borrow Hewitt’s phrase, that one can view the Marathwada disaster as a ‘crisis of modernity’ (1995: 117). Th is understanding of the local building culture prior to the quake, as well as the concept of hazard ‘vulnerability’ in the social sciences is critical to the remainder of this article. In their book At Risk: Natural Hazards, People’s Vulnerability, and Disasters, Blaikie et al. provide one of the most comprehensive and relevant defi nitions of vulnerability:

By ‘vulnerability’ we mean the characteristics of a person or group in terms Downloaded by [University of Defence] at 01:15 24 May 2016 of their capacity to anticipate, cope with, resist and recover from the impact of a natural hazard. It involves a combination of factors that determine the degree to which someone’s life and livelihood is put at risk by a discrete and identifi able event in nature or society […] Some groups in society are more prone than others to damage, loss, and suff ering in the context of diff ering hazards. Key characteristics of these variations of impact include class, caste, ethnicity, gender, disability, age or seniority […] Th e word ‘livelihood’ is important in the defi nition. We mean by this the command an individual, family, or other social group has over an income and/or bundles of Lessons from Marathwada Ú 87

resources that can be used or exchanged to satisfy its needs. Th is may involve information, cultural knowledge, social networks, legal rights as well as tools, land and other physical resources […] (1994: 9).

Th is defi nition is particularly important in that: (i) it defi nes natural disasters as the outcome of both physical determinants (such as the type and amount of damage to buildings caused by earthquakes) and sociocultural issues that precede and shape the outcome of disaster events; (ii) it defi nes ‘vulnerability’ not as a social problem (the more popular usage of the term, where people are seen as passive victims due to poverty, age, gender, etc.), but as a process wherein people are actively involved through daily life in both creating and reducing risks; and (iii) it places people’s ‘livelihood’ as the key concept in understanding the impact of disasters and extends this to mean much more than only physical resources. We will return to these defi nitions at the end of this paper in order to understand the policy impacts of MEERP more clearly.

In-Situ Reconstruction and Participatory Development: World Bank Policy in the Early 1990s

By the early 1990s the idea that development work itself could have devastating impacts on the local community was well understood by the World Bank anthropologists and policy makers who had studied relocation projects (Kreimer and Echeverria 1990; World Bank 1992, 1994b and 1998). As was widely recognised by social science scholars, the principal failure of these projects lay in the inability of displaced populations to re-establish place-bound sociocultural and economic practices. Renowned anthropologist Th ayer Scudder’s work is particularly important in this regard, as he pioneered studies that showed how relocation creates physiological stress (increased mortality); psychological stress (feelings of trauma, guilt, ‘grieving for a lost home’, and anxiety Downloaded by [University of Defence] at 01:15 24 May 2016 about an uncertain future); and sociocultural stress associated with the economic, political and cultural aftereﬀ ects of relocation. Th e most serious kind stress that follows relocation to new settlements, ‘is brought about by a major reduction in cultural inventory due to a temporary or permanent loss of behavioral patterns, economic practices, institutions and symbols’ (Scudder and Colson 1982: 271). Th e relocation of housing, one can argue, exacerbated these problems because they were designed by planners, structural engineers and 88 Ú Salazar and Jigyasu

architects who knew little about rural life and vernacular organisations of space (Aysan and Oliver 1987; Davis 1981). Anthropologist Paul Oliver’s summary critique of relocation work after the 1970 Gediz earthquake, for example, is one of the most poignant and perhaps prophetic statements in this regard:

Th e “disaster housing” was designed for occupation by nuclear families, but the peasant families were of the extended family type — living in the prefabricated dwellings either meant overcrowding to accommodate the family, or the break-up of the family system…. Th ough they were adapted to make them more convenient to live in, like other adaptations and extensions, including the adding of outbuildings that the peasant farmers required, the additions make the house more vulnerable in the event of another earthquake… (1987: 215).

Th us, by the early 1990s, after a decade of expert critiques and NGO and CBO opposition to relocation work (Cauﬁ eld 1996; Rich 1994), the World Bank began to acknowledge the problems created by relocation and shift its policies toward minimising relocation and promoting reconstruction in-situ. In the ﬁ eld of hazards, the document Rebuilding Housing in Emergency Recovery Projects (Kreimer and Echeverria 1990), for example, clearly stated the World Bank’s position at this time:

After a disaster, inappropriate policies are sometimes adopted. In housing programs, emphasis on relocating the affected population to sites far from jobs, services, and social facilities may divert capital and eff orts from implementation of progressive, on-site permanent housing programs. Proposals for relocating people after a natural disaster are motivated by a variety of concerns, ranging from environmental to technical and political. Th e motive is often to remove settlements from unsafe areas. After a fl ood, volcano, earthquake, or cyclone, families may fi nd that their homesites and means of livelihood have simply disappeared. Even so, relocation should be considered only in clear cut cases of vulnerability, to reduce signifi cant risks that cannot be otherwise addressed. Downloaded by [University of Defence] at 01:15 24 May 2016 Additionally, by the early 1990s, the Bank’s ad hoc methods of policy formation, where sociologists and anthropologists reviewed projects after construction, were rejected (Cernea 1993). To address this gap in planning, the World Bank shifted its focus towards Participatory Rural Appraisal (PRA) and Benefi ciary Assessment methods, where project benefi ciaries were consulted during policy formation (Mosely et al. 1995). Th e acceptance of PRA in policy-making practices thus became a way to maintain a social–scientifi c basis for policy decisions as well Lessons from Marathwada Ú 89

as a political device to bring NGOs and CBOs into the development process.

The Politics of Participation

After the Marathwada earthquake, the only credible PRA research was conducted by TARU, whose report (TARU 1993) was utilised by World Bank offi cials and central government agencies to outline policy options (World Bank 1994a; Government of India [GoI] 1993a). One of the key fi ndings of TARU’s technical damage assessment was that while the stone masonry techniques could use improvement, the traditional post-and-beam system had actually prevented the collapse of many buildings, thereby drastically reducing the number of casualties. For this reason, the seismic retrofi tting of vernacular structures was considered a viable option, and villages that required relocation were recommended to: (i) hire local masons, stonecutters and carpenters in the production of modifi ed, earthquake-safe vernacular homes built mostly with local materials; and (ii) base the spatial planning of new villages, and the design of houses, on vernacular forms. As stated by the TARU report, this would require a ‘longer start-up time and greater investment in institution development’, but it would also insure ‘long term continuity and integration of technologies into the local idiom of building; lower relative costs; higher accountability and greater possibility of community participation’ (1993: iii). Similar to the views of many other architects and engineers involved in post-disaster work, these conclusions were not unusual (for instance, ASAG 1994; Baker 1993; Menon and Bhaskar 1994). Indeed, such ‘bottom up’ approaches to rural development, especially housing technologies, were common in India, where a plethora of NGOs are working in the well-established environmental housing movement. Nonetheless, these views did run counter to conventional wisdom.

Downloaded by [University of Defence] at 01:15 24 May 2016 Rather than seeing the earthquake as a crisis of modernity, policy makers considered the collapse of traditional buildings as the result of the ‘backwardness’ and ‘poverty’ of rural life. Reinforcing this perspective were the views of local residents who, over the years, had become disillusioned with local construction materials such as mud, thatch and stone, and were enamoured with the ‘modern’ forms of urban settlements. Cement and steel, it was thought, were easier to maintain and longer lasting. Th is bias was reinforced through faulty scientiﬁ c data provided by Indian Space Research Organization (ISRO) and National 90 Ú Salazar and Jigyasu

Geophysical Research Institute (NGRI) who released remote sensing images that showed the close proximity of lineaments (fractures in the earth surface) to traditional settlements and detailed the depth of ‘black soil’ (GoI 1993b), a material especially good for agriculture but highly expansive and unstable. Th e GoM used this data in policy reports and before mass media to promote the idea that living on black soil was dangerous, bolstering their eff orts to relocate as many villages as was politically possible (GoM 1993a: 28–29; 1993b; 1994a: 13). However, black soil has been built on for centuries in Marathwada, and it was well documented after the quake that structural failures of building foundations did not typically occur. Additionally, in interviews with NGRI fi eld scientists in 1994 it was discovered that no ground verifi cation of the satellite images had been done. It was impossible to know, therefore, which of the lineaments were active faults and which ones were simply other kinds of formations that posed no threat to human habitation. Th e government’s pseudo-scientifi c rationale became popular within a few months of the quake, a time when villagers were completely dependent on outside aid for survival and were living in complete fear of stone masonry construction. Th e erstwhile Chief Minister of Maharashtra, Sharad Pawar, promoted relocation further during his 1993–94 campaign when he advocated ‘speedy’ reconstruction and provided relocation sites to top industrialists, such at Tata and Hindustan Petroleum, who promoted prefabricated housing technologies. Capitalising on this political opportunity, industrialists were able to break ground as early as 24 October 1993, just 24 days after the quake! (Times of India, 22 October 1993). And as news spread that the fi rst wave of ‘modern’ relocation projects were being built with lots of cement and steel, the overwhelming message to artisans and the local population was that traditional settlement sites and stone masonry techniques were no longer safe. Th e natural result of this was that many villagers simply embraced the GoM rhetoric that relocated homes would be ‘earthquake proof’ (ASAG, TARU and Youth for unity and Voluntary Action [YUVA] 1994: 71). For many Downloaded by [University of Defence] at 01:15 24 May 2016 marginalised communities this seemingly dependent relationship on the state was understandable: it was their only chance at getting a ‘free’ pucca house equipped with the necessary basic services (Tata Institute of Social Sciences) [TISS] 1994: 24–26), something that hitherto only the wealthiest castes and urban populations had access to. While the eff ort was promoted by Chief Minister Pawar, the programmes were managed by K. S. Sidhu, the MEERP Project Co-ordinator, and later by J. Joseph, the MEERP Project Director. Th eir work did not start out smoothly. Th e decision to relocate ran into various social and Lessons from Marathwada Ú 91

political pressures at the village level, often leading to the bifurcation of settlements along caste and community lines with the weakest members of society being marginalised here as well, such as at Holi, Matola and Utka (TARU and YUVA 1995). Th e push to relocate also sparked a court case. In February 1994, attorney Krishna Das fi led a petition in the Maharashtra High Court to extend emergency services and shelter to neglected communities hit hard by the quake. Th e court agreed in March 1994, and ‘inevitably, the judiciary, upset by the government’s stonewalling, appointed an independent enquiry commission, which reported that there was a complete absence of supervision over quality control and the materials used by the government and donor agencies’ (Das 2000). With additional public pressure from the Bhukampa Kriti Samiti, representing the panchayats of the seven neglected core villages, the High Court ordered in August 1994 the re-categorisation and reconstruction of the villages, and the monitoring and accountability of the rehabilitation programme. Th us, one can conclude that relocation work progressed principally due to political pressure from within governmental structures, including scientifi c agencies of the GoM and GoI, as well as the willingness of the World Bank to fund projects that contradicted its own policy guidelines. Equally important, the development of post-disaster psychological fears among local residents and advocating that additional villages be included in the MEERP was principally an outcome of the active promotion of speedy reconstruction by the GoM rather than the ‘cause’ of faulty policy decisions as argued in other policy reports (for instance, Nikolic-Brzev et al. 1999: 18). For NGOs involved in the environmental housing movement, this ground reality was devastating and ensured that their socially and ecologically conscious designs would become isolated experiments in a fl ood of GoM/World Bank-sponsored urban resettlement schemes. Figure 4.4 shows workers recycling stone and mud from old walls for

Downloaded by [University of Defence] at 01:15 24 May 2016 reuse at the new Banegaon village under the guidance of Laurie Baker in 1994. Baker worked on this site in 1993–94 using local materials and labour, and with the support of local residents. Th e developer, however, had political ties and ultimately caved in to such pressures by removing Baker from the job. Th e government, it was reported, did not want such a high proﬁ le architect to provide an alternative to standard relocation. New architects were subsequently brought in and houses were built with conventional concrete and brick materials (see Menon and Bhaskar 1994 for more details). Figure 4.4: Reconstruction Work at Banegaon Village Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Photo by Alex Salazar. Lessons from Marathwada Ú 93

Part II: An Assessment of Policy Impacts (1994–2001)

A closer look at the World Bank’s development policy reveals that the appropriate technology PRA approach to disaster mitigation was never really what it had in mind for the MEERP. Rather, there was a diff erent undercurrent of policies that the World Bank pushed in order to legit- imise post-disaster projects: namely ‘enablement’ housing policies. Adopted fi rst in Latin America, and later embraced by the international community, enablement began in the form of modest ‘participatory’ sites and services projects, whereby government agencies relocated squatters onto plots of land equipped with basic infrastructure elements, such as roads and sewers (Turner and Fichter 1972; Harms 1982). In more recent years, this strategy has evolved into programmes where NGOs, CBOs and private companies carried out participatory projects, and governments relaxed regulatory controls on housing markets (Pugh 1995; Fernandes and Varley 1998). Initially, the GoM was resistant to this model and remained confi - dent about its plan to relocate about 83 villages (GoI 1993a: 13) without any real public process and with complete disregard for the village context (GoM 1993a: 49). However, with media attention to the story and opposition from some NGOs, the GoM was forced to rewrite its proposal and reduce the number of relocation projects to 52. Rather than marginalising villagers, peoples’ participation became the essential strategy that would ‘ensure socially, culturally and economically self- sustaining communities in an environment that includes appropriate housing and civic amenities’ (GoM 1994a: 2–3). Th e MEERP was launched in May 1994, and had fi ve principal components, namely, housing, infrastructure development, economic rehabilitation, social and community rehabilitation, and technical assistance, training and equipment. We will limit our discussion to Downloaded by [University of Defence] at 01:15 24 May 2016 housing, implemented by the Project Management Unit (PMU) — a development agency created to co-ordinate contracts, material procurement, engineering and design. Housing was produced in three types of villages: Type ‘A’ villages were to be relocated, Type ‘B’ villages were to be reconstructed in-situ, and Type ‘C’ villages were to be primarily retrofi tted. By 2001, these rehabilitation processes were almost complete. Th e following sections assess some of the major outcomes of these housing initiatives. 94 Salazar and Jigyasu

Relocated Type 'A' Villages

In order to qualify for relocation, villages were to have had more than 70 per cent of their buildings severely damaged, a high number of deaths and to have been built on black cotton soil of over 2 metres depth. Initial- ly, the town planning offices in Latur and Osmanabad districts designed about 20 new village plans, which were constructed by donors (corporations and NGOs). On the basis of earlier landholdings, families were allotted houses of three types: 'A' houses had an area of 750 sq. ft. and

Figure 4.5: New Babalsur Village's Layout Plan Made by Town Planners

Plot Uses L0 Gram Panchayat (village council) 1 Anganwadi (daycare) 2 T c Maharashtra State 3 c_ :s 1.K iy *' Electricity Board c Hospital 4 ;9

Play Ground 5 • Primary School 6 Religious Center 7 7 8 ; 2 Samaj Mandir (welfare center) 8 12 Forest 9 C Water Supply It M J m 5 Agricultural Processing 11 I Garden 17 Slaughter House 13 Cattle 14 17 J Vet Hospital 15 i- S Bus Stop 16 Shopping Center 17 Crematorium IE •••i Burial Grounds IS C E 7 | Ft I m House Plots* § C Downloaded by [University of Defence] at 01:15 24 May 2016 Families with A H ii < 1 ha. of land K

Families with B 13 I I II 1:> 1-7 ha. of land N Families with C 7+ha. of land * Along entire row unless otherwise indicated. 0 50 100 200 m

Source: Drawing by Alex Salazar. Lessons from Marathwada 95

were provided to families with more than 7 hectares of land; 'B' houses had an area of 400 sq. ft. and were provided to families holding 1-7 hectares of land; and *C houses had a carpet area of 250 sq. ft. and were provided to families with up to 1 hectare of land or to those who were landless (Figure 4.5). Donors were allowed to modify these plans, which explains the wide variety of housing types in the region. Later, the PMU took control and implemented participatory plans for an additional 30-odd villages with the use of Community Participa- tion Consultants (CPC) from TISS. Taking lessons from agencies such as the Maharashtra Housing and Area Development Authority (MHADA) and the Housing and Urban Development Corporation (HUDCO), the PMU utilised CPC and methods that allowed the villagers to participate — like the squatters in urban development projects — in a few design meetings with NGOs and housing professionals (see Figure 4.6). Now that relocation is complete, it is fascinating to see how people have modified their new houses by initiating changes and additions to

Figure 4.6: Typical Housing Units Constructed by MHADA

Road Road Road

1 6 2 2 1 1 2 3 3 J i 3 4 5 7

7 Families with <1 hectares of land

Families with 1-7 hectares of land Downloaded by [University of Defence] at 01:15 24 May 2016 Legend 7 1 Veranda 2 Living 3 Kitchen Families with 7+ hectares of land 4 Room 5 Bedroom 6 Storeroom 7 WC and Bathing 0 2.5 5 10m

Source: Drawing by Alex Salazar. 96 Ú Salazar and Jigyasu

the physical structure. Changes in house form include the addition of rooms, outdoor kitchens, courtyards and access points that recreate the traditional wada-style courtyard homes. Also noteworthy is the change in the materials used for this work. Very few villagers have used reinforced concrete (except for those who are very well off ), some have used bricks, and many have used stone, corrugated tin sheets, even bamboo and twigs. Th e old village fi gures well into how people have coped over the last few years. Residents have maintained religious associations through daily visits to their old temples, they have incorporated their beautiful doorways and, in some places, have reused dressed stone masonry, and wooden beams and columns in their new homes. Despite their eff orts, however, people have also encountered many serious hardships, which are described below. Environmental Problems/Reduced Labour Output Traditionally, agricultural land surrounded the villages, and the whole rural ecology was sustained by this delicate relationship of the people to the natural resources around them. However, relocation was done on agricultural land anywhere from 1 to 7 kilometres from the village centre. As a result, many villagers either lost their land, thus becoming landless forever (even though some fi nancial compensation was off ered to them) or had been relocated too far from their fi elds. Th is increased hardships for the villagers, increasing their commute time and travel costs and in the process reducing labour output (ASAG 1998). Socio-Spatial Problems/Caste Dependency Traditional Marathwada settlements are characterised by narrow streets, a hierarchy of public and private open spaces used for religious as well as other activities, clusters of attached housing (each unit serving extending families) of distinct types characterised by traditional occupation patterns, and a social system based on neighbourhood units

Downloaded by [University of Defence] at 01:15 24 May 2016 and interdependencies that ensure mutual sustainability. However, relocated villages were designed with a ‘city-like’ plan, with wide streets forming a grid pattern and detached housing (each unit serving one family) covering a huge area manifold larger than the old village sites. Th is dramatic socio-spatial change has created many new and unnecessary social and economic disparities. According to local sources, the isolation of families and the physical distances between communities/ neighbourhoods has led to a resurgence of the caste system and greater dependency on outside resources. Lessons from Marathwada Ú 97

House Design Problems/Reduced Livelihood Opportunities In many cases people vacated their allotted houses and/or moved in with other family members or neighbours by initiating house extensions. Th ese extensions have come at a great price, especially for the poorest members of society who were allotted only 250-sq. ft. homes. At Gubal, for example, one of the families has taken loans from money lenders worth tens of thousands of additional rupees (not counting their labour) to build two additional rooms: one for sleeping and another for cooking, to create a more traditional, rectangular wada-style courtyard house. To repay the loans, two family members have been indentured to a local landlord for two to four years at half their normal income. Artisans have also suﬀ ered disproportionately. Traditionally, artisans are either landless or have minimal land holdings, and act as a support system for the village, and are not expected to cultivate land. MEERP house allotment criteria thus relegated them to the plots and houses of the smallest size, resulting in less than adequate space to be able to carry out income- generating activities. Infrastructure Problems/Wasted Public Resources Th e large size of the relocated villages meant expensive infrastructure, which was provided by the government. Th e MEERP failed to consider

Figure 4.7: Abandoned Shopping Complex at New Killari Village, 2001 Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Photograph by Alex Salazar. Note: Like the roadways, the scale of open spaces at new villages is so large that they are unusable. 98 Ú Salazar and Jigyasu

the village committees’ lack of fi nancial resources to be able to provide for maintenance costs, leading to the widespread problem of road and drainage/sewer system deterioration. Village committees responded by increasing taxes to cover the cost of maintenance — a cost that poor villagers are unable to aff ord. For example, in 2001, in Jewli village, the poorest people, who formerly had to pay Rs 135 (approximately US$ 2.7) per house per year, are now made to pay Rs 1,200 (approximately US$ 24). Due to their inability to pay, only 10–13 per cent of the projected revenue is being collected. Another common infrastructure problem has been the provision of toilets. People here are not used to having toilets in their houses (they use the fi elds) and now typically use the toilet only to store grain. Th is changed usage indicates that the tremendous cost of building such facilities to address important public health problems has been a waste. Construction Problems/Increased Hazardous Conditions In many villages the quality of new ‘earthquake resistant’ construction appears to be very poor, due principally to the GoM/PMU target-oriented approach that stressed speedy reconstruction, lacked construction oversight, and suff ered due to the lack of water, required for the curing of concrete, in the drought-prone region. At Killari, for example, 53 houses were ordered to be demolished because of their poor quality (Aubrey 2000). Leakages and damp are also a common problem, arising due to porous concrete blocks that lack proper pointing. However, the most serious concern has been the development of ‘through cracks’ in some houses due to a recent earthquake of mild intensity (4.0 on the Richter scale, in June 2000). In one village, Rebe Chincholi, out of fear, villagers have vacated some of these houses. If a moderate-intensity earthquake can cause such damage, then one can imagine the consequences of an earthquake equal to the intensity of the 1993 quake. Th erefore, in spite of people’s extraordinary capacity to adapt to the environment over time,

Downloaded by [University of Defence] at 01:15 24 May 2016 relocation into concrete bungalows, in many situations, may have actually created conditions that were physically more dangerous. Abandonment Problems As a consequence of the above issues, and many others, it is common to ﬁ nd abandoned housing units at the new villages. Typically, they appear to be partially abandoned, Killari being the most well-known and which appears to be only 50 per cent occupied (see Figure 4.8). Other lesser- known villages have been completely vacated. At Sayyed Hipparga, for Lessons from Marathwada Ú 99

Figure 4.8: Abandoned Neighbourhood at the New Killari Village

Source: Photography by Alex Salazar.

example, residents decided to vacate the new settlements and move back to their old site, reconstructing vernacular houses by employing traditional techniques in their entirety. At Sarni, villagers refused to take possession of the new site because it was too far from their ﬁ elds (over 5 km), would re-quire repayment of house loans (for the larger type ‘A’ and ‘B’ houses), and since they felt frustrated by their lack of participation in decisions (TARU and YUVA 1995). Hassalgaon was abandoned for the want of water (Aubrey 2000).

Reconstructed In-Situ Type ‘B’ Villages

Villages reconstructed in-situ were expected to have suﬀ ered more than 70 per cent damage but on ‘good’ soils — i.e., less than 2 metres depth of black cotton soil. In terms of the individual allotment of houses, the programme was more or less identical to Type ‘A’ Villages (above), but

Downloaded by [University of Defence] at 01:15 24 May 2016 it had the additional task of clearing old house plots of debris prior to construction. Initially, there were 10 villages in this category (GoM 1993a), but due to social and political pressure, and lawsuits ﬁ led by panchayats (see above), the number grew to about 22 villages and ultimately the GoM decided to relocate these ‘hybrid’ villages to new sites (Nikolic-Brzev et al. 1999). As a result, by 2001, the number of relocated villages ballooned from 52 to 74. An independent evaluation of these relocated villages is not possible at this time, however, policy documents state that the GoM agreed to provide schools, electricity, 100 Ú Salazar and Jigyasu

and water supply, but not roads or drainage systems; in three or four villages with strong leadership, owners took responsibility for managing construction activities and in the remaining villages, donor organisations (such as CARITAS), fully or partially, managed construction activities along with the PMU or homeowners (ibid.). With the demise of the formal in-situ reconstruction programme it is interesting to note two additional villages that ended up being reconstructed in-situ: Tembhe and Pardhewadi. At Tembhe, HUDCO rebuilt homes on the foundations of the old houses. Th e whole village was thus recreated as it was before; even the front facades of houses used stones salvaged from the old houses. However, there are some problems with this approach. Except for the front facades the rest of the buildings were made of cement blocks which do not allow changes to be made in the future to the house form. At Pardhewadi, a village originally slated for the RRS programme, three organisations worked with a community of about 150 households and eventually organised a collective approach to rebuilding in-situ. The lead organisation at Pardhewadi was an NGO named Manavlok (based in Beed district) that was originally involved with providing medical services and helping co-ordinate relief work in the disaster area. As time went on, two other organisations, Pardhewadi Samajik Sewa Mandal (an informal Latur-based group involving architects and civil engineers) and Marathi Vigyan Parishad (a Mumbai-based group with people skilled in house design and construction) got involved with the project and worked with residents to rebuild the village through an innovative community-led design and construction approach. Th e organisations’ approach utilised the family’s own labour as well as the skills of local craftspeople to rebuild with locally available brick and stone, much of which was recycled from walls that had collapsed in the earthquake. Th e chief diffi culty faced by the NGOs was to convince people who had been left traumatised by the earthquake that a properly Downloaded by [University of Defence] at 01:15 24 May 2016 rebuilt traditional home could be as safe as a new government-built concrete house. Th ey and other organisations promoted the idea of seismically retrofi tting traditional homes and employing good common- sense masonry construction with just a minimum amount of concrete and steel. Th e technology itself was not diffi cult but convincing villagers to use it was. After months of informal meetings and construction demonstrations through model houses, however, the residents of Pardhewadi were Lessons from Marathwada Ú 101

fi nally won over and formed a village committee to take on the responsibilities of material and labour mobilisation and resolving disputes (TARU 1995). Th ey agreed on a few basic principles: that families could reconstruct their own houses as they chose on their own lots, with their own or contracted labour; that they would contribute part of their lands for road widening, and if necessary for the location of a school or community hall; and that they would employ safe construction and retrofi tting techniques. In return, villagers and artisans received the NGOs’ co-ordination and training inputs in earthquake-safe building technologies, fi nancial assistance towards the reconstruction of their own homes (including, if they desired, a toilet), and fi nancial assistance towards upgrading village-wide utilities, roads, drainage systems, and agricultural land watershed development. Th e villagers thus became involved in all aspects of the work: dirt roads were widened, just slightly, to improve drainage and safety, foundations were laid, gravel sifted to prepare aggregate for concrete, walls plastered, wood sawn for doors and windows, bricks heaved off trucks to the waiting hands of masons perched on rooftops, and so on. Overall, the work at Pardhewadi has rightly become a positive example of reconstruction in-situ in Marathwada. Residents have avoided many of the problems experienced at relocated sites by rebuilding on old foundations, maintaining their physical proximity and the basic village form. Th ey have also been able to execute more extensive rebuilding than typical RRS villages by reducing construction costs through the recycling of material. And they have eff ectively learnt earthquake-safe building methods by helping to manage work that utilised local masons in the construction of their own homes. While Pardhewadi was not a type ‘B’ village offi cially, the project demonstrates that in-situ work, had it been carried out on a much wider scale (in lieu of relocation), could have helped villagers in the core disaster area recover better and mitigate future disasters. Downloaded by [University of Defence] at 01:15 24 May 2016 Reconstruction, Repair and Seismic Retrofi tting Type ‘C’ Villages

Villages included in the RRS programme were decided on the basis of a detailed GoM engineering survey. In these villages, few people had died and not much physical destruction had taken place. Some were beautiful, traditional settlements with long historical continuity resulting in various heritage components such as fortress houses (garhis), 102 Salazar and Jigyasu

Figure 4.9: Typical Retrofitting Techniques

Reduce mud roof thickness by inserting a polythene sheet. Reduce height of paxapet and rebuild withgfloditoriE? masonry techniques.

Construct a concreti bond beam on top oi all walls.

Insert concrete- ferrocement "through stones" every 1 sq, in.

Connect posts to beams with steel knoeb racing.

Source: Drawing by Alex Salazar. Adapted from ASAG (1994) and Arya (1994).

fortifications and water structures, in addition to vernacular housing. In concept, no matter how deteriorated the traditional houses were before the earthquake, their strengthening and retrofitting would have proved to be a much cheaper option considering the local availability of materials and craftsmen. Moreover, strengthening and retrofitting need not be a highly technical job; several indigenous methods have also been developed based on the resources and skills available (see Figure 4.9). It was recognised in the initial stages of this programmme that traditional housing types could be strengthened or retrofitted with cheap and indigenously developed techniques. Taking the lead from Dr Arya of Roorkee University, a pioneer in seismic engineering of unstabilised vernacular buildings, the GoM experimented with retrofitting work in 1994 and subsequently developed this programme. In these villages, the GoM provided two packages of financial assistance for in-situ re-

Downloaded by [University of Defence] at 01:15 24 May 2016 construction, repairs and strengthening, depending upon the severity of damage. This package included Rs 15,000 for those houses which suffered damage of categories I, II and III, as defined by the International Association of Earthquake Engineering, and Rs 30,000 for those which suffered damage of categories IV and V. Since the government was so involved with new construction, measures such as repairs, strengthening and retrofitting, which were in fact a major component of the programme, got sidelined. The critical turning point in the development of retrofitting policies was the government's Lessons from Marathwada Ú 103

decision that ‘found that the repairing and strengthening of stone and mud houses is very diffi cult’ (GoM 1994b), and prohibited the use of mud mortar even though the Bureau of Indian Standards allows it (Das, Aybrevi and ASAG 1998). Th is resolution also introduced the construction of a ‘safe and habitable new room’ as an option for families, which opened the door to the total transformation of this programme away from retrofi tting and towards the construction of new buildings in existing villages. Moreover, for each village, home to hundreds of families, the government appointed only two junior engineers to provide technical assistance. Th us, technical assistance was in short supply and villagers were simply provided the money and were expected to carry out these repairs on their own. Th is had a marked negative impact and led to several major failures, which are discussed below. Faulty Policies and Technical Guidance From the inception of the RRS programme, it was apparent that retrofi tting and strengthening was not a preferred technology package for the benefi ciaries. According to the Quality Assurance and Technical Audit consultants, only 0.1 per cent of the benefi ciaries decided to repair and strengthen their houses (Vatsa 2001). Th is was due mainly to the fact that engineers are trained in urban construction techniques and themselves perceive vernacular buildings as outdated and weak. As such, they strongly advocated that local people construct ‘modern’ housing using brick and concrete, as had been popularised through various government reports and media coverage. Building additional rooms in brick and concrete was also logistically familiar to the engineers and fi t well into the ‘target’-oriented demands of the PMU which required quick production and the completion of the RRS programme (see TARU 1995, TARU and YUVA 1995 for details). As a result, these attitudes and policies reinforced the villagers’ attitude against vernacular building methods. As they had suff ered extreme trauma, many

Downloaded by [University of Defence] at 01:15 24 May 2016 residents were too scared to risk their lives in any way, and submitted to expert views, which rarely recommended retroﬁ tting as an option. Th e opportunity to gain a new pucca room addition, which normally they would not be able to aﬀ ord and which is perceived as being superior to vernacular construction, was also too much to give up. Limited Success of Participation Community participation was an integral part of the RRS programme, introduced at the insistence of World Bank and advocated by many 104 Ú Salazar and Jigyasu

NGOs. For this purpose CPCs were hired. While being managed by the PMU, the CPC contract went to Swayam Shikshan Prayog (SSP), an NGO that started as a project of the Society for Promotion of Area Resource Centres (SPARC), a women’s development NGO based in Mumbai that focuses on squatter housing. Th e aim of this participatory eff ort was to educate local villagers and artisans in construction methods using local materials, primarily stone, while empowering women and changing power relationships between men and women at a village level. SSP had a strong presence in over 500 villages (Gopalan 2000), while the full RRS programme was operational in over 2,400 villages and included a total of 189,000 houses (Nikolic-Brzev et al. 1999). Th us, only approximately 20 per cent of the RRS villages had actively incorporated participatory eff orts into the work. In these villages ‘over 1,000 women were trained to supervise earthquake-resistant construction. Women leaders also initiated Gram Sabhas and linked with Gram Panchayat members to solve common problems such as water, transport, and availability of masons’ (Gopalan 2000). While the creation of women’s groups does appear to have been successful in terms of equity issues, SSP/GoM were slow to develop and ran into many diffi culties (see TARU 1995, TARU and YUVA 1995 for details). In the end, like the remainder of the RRS programme, the participatory component ended up promoting the construction of new room additions using brick masonry. Figure 4.10 shows a retrofi tted stone and mud home on the right, and a new brick home (with limited use of stone) on the left, at Mogarga village. Th is village received extensive technical support and RRS aid from government engineers but the technologies have not taken root. Th e homeowners use only half the retrofi tted house during the day, and sleep in their new, brick homes at night. Ongoing Hazardous Living Conditions With the construction of new concrete and brick rooms, many old Downloaded by [University of Defence] at 01:15 24 May 2016 houses have been left without any seismic retrofi tting work being done. Villagers have continued to use these buildings, which poses a tremendous risk should another earthquake occur. In some places, where villagers used the money allocated by the government to construct new houses just outside the old villages, poor quality bricks of mud mortar and corrugated tin sheets for roofi ng is a common sight. Th ese new houses are also very vulnerable to earthquakes. Lessons from Marathwada Ú 105

Figure 4.10: Mogarga Village, 2001

Source: Photograph by Alex Salazar.

Missed Opportunities In spite of an attempt by local government offi cials to involve villagers in reconstruction work, the overall direction of the World Bank and GoM housing policy towards new construction marginalised NGOs who were strong advocates for the use of local materials and artisanal labour. For example, ASAG successfully demonstrated innovative techniques in retrofi tting vernacular stone and mud buildings. Starting in 1994, it worked directly with hundreds of families and artisans to teach them how to repair and retrofi t their individual homes. In some villages this

Downloaded by [University of Defence] at 01:15 24 May 2016 boosted the people’s confi dence in the technology. At Nagarsoga and Almala villages, for example, these retrofi tted vernacular homes remain occupied and the families have become local experts for other villagers who have returned to construction using stone and mud. ASAG’s methods, successful at a small scale, however, were slow to develop and ran into many diffi culties (see TARU 1995; TARU and YUVA 1995 for details), and ultimately were not integrated into the larger RRS policies promoted by the World Bank and the GoM. 106 Ú Salazar and Jigyasu

Part III: Vulnerability and Livelihoods: Was the MEERP Successful at Mitigating Future Disasters?

From the perspective of this article, which grounds its view in a ‘vulnerability’ perspective, there is clearly little in the MEERP to be positive about. As previously noted, by vulnerability we mean ‘characteristics of a person or group in terms of their capacity to anticipate, cope with, resist and recover from the impact of a natural hazard. It involves a combination of factors that determine the degree to which someone’s life and livelihood is put at risk by a discrete and identiﬁ able event in nature or society’ (Blaikie et al. 1994). With this deﬁ nition in mind, eight years after the quake, it is important to summarise the outcome of the MEERP in terms of the impact that the production of housing has had on the capacity of villagers and local builders to recover from the 1993 disaster and mitigate future hazards through current activities.

Lost Opportunity in Participatory Reconstruction

Th e GoM’s emphasis on speedy construction and the use of large building contractors for housing work, which took place immediately after the quake, during the relief stage, and was promoted through ‘donor’ relocation housing (with infrastructure ﬁ nanced by the World Bank), created a sense of dependency in the local population and contributed to the people’s lack of faith in vernacular stone-and-mud building practices. Th is policy context and social reality, during the rstﬁ year after the quake, had a tremendous negative impact on all three MEERP housing programmes. At relocation type ‘A’ villages (52 villages; 27,919 homes), this shift in people’s attitudes basically guaranteed a passive population and an inability for participatory reconstruction to occur. At

Downloaded by [University of Defence] at 01:15 24 May 2016 neighbouring reconstruction in-situ type ‘B’ villages (22 villages; 10,628 homes), attitudes were much the same, resulting in their relocation and undermining the entire programme — with the exception of work at a few villages. And at the RRS type ‘C’ villages (2,400 villages; 212,000 homes), the lack of faith in vernacular building methods resulted in the transformation of this housing model from one of participatory reconstruction into a target-oriented programme that promoted the construction of new ‘safe’ rooms by individual households — with the opportunity to participate mainly in programme management and only Lessons from Marathwada Ú 107

in approximately 500 villages where the SSP was active. As a result, the opportunity for educating and re-training local artisans and builders in proper, earthquake-safe stone masonry techniques, was lost. Th ese major failings of the MEERP have been particularly detrimental to livelihood opportunities for the traditional local artisans.

Limited Impact of Traditional Stone Masonry Trainings

Under the training component of the MEERP, a large number of workshops were organised to train unskilled labour and masons in earthquake- safe technologies — an eﬀ ort designed to improve building technologies and prevent the migration of artisans out of the region, as was common even prior to the earthquake. For traditional artisans, two-day seminars on stone masonry construction were organised, reaching approximately 4,000 masons in Latur and Osmanabad districts; and for unskilled labour, two-week seminars were organised, focusing on masonry construction (brick, concrete block and stone), reaching approximately 6,800 individuals in Latur, Osmanabad, Satara and Solapur districts (Nikolic-Brzev et al. 1999). However, since nearly all reconstruction work focused on ‘modern’ techniques (concrete blocks and brick), traditional masons could not really make use of their improved skills for indigenous construction. Once reconstruction activity was over, many artisans migrated out to nearby cities in search of jobs, leaving mostly unskilled labourers to work as masons in both relocated and RRS villages.

Loss of Exemplary Stone Masonry Buildings

Th e failure of participatory work and masonry training turned the

Downloaded by [University of Defence] at 01:15 24 May 2016 Marathwada ‘natural’ disaster into one of the biggest man-made disasters that the country has witnessed. In the fi rst few years after the quake, some people slowly vacated and demolished their old houses, selling off their well-dressed stone blocks and wooden beams and columns at petty prices. Wood members have also been burned as fuel wood or, more commonly, sawn down for use as light-weight materials to support tin roofs, a new practice commonly seen in the region. And the once ubiquitous and strong white-mud mortar has long been sold off as fertiliser or left to wash away in the monsoons. All this has contributed 108 Ú Salazar and Jigyasu

to the people’s lack of faith in traditional building practices. So strong are these perceptions that in some places people have demolished age- old temples of the fi nest stone and wood carvings and made imitations in concrete. Th ese new buildings are often poor-quality constructions because of the expensive nature of cement and the scarcity of water. Th us, nearly a decade later, one fi nds ‘modern’ reconstructed temples and other public buildings with cracks and water seepage problems that are highly prone to collapse. Th e loss of examples of good quality vernacular homes and other buildings that were important cultural landmarks is the result of unsuccessful MEERP policies and represents another way in which vernacular building trades have been undermined: younger artisans and local builders no longer have these buildings to learn from. For these reasons, namely, limited opportunity for participation in rebuilding activities, the minimal impact of retraining programmes and the loss of exemplary stone masonry buildings, the traditional stone masonry community in Marathwada has suff ered a tremendous blow. Th eir livelihood was fi rst disrupted by the earthquake and then completely destroyed by the MEERP’s policies, which failed to incorporate their needs. As a result, as observed in 2001, wherever new construction is taking place, at both relocated and RRS villages, not much has improved

Figure 4.11: Addition of Stone Wall and Room Constructed in the Traditional Style at New Babalsur Village, 2001 Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Photograph by Alex Salazar. Lessons from Marathwada Ú 109

with regard to earthquake safety. Villagers have continued to use brick construction, and in many cases returned to stone masonry technologies (at least for boundary walls and walls up to sill level), but ‘earthquake- safe’ features, such as through stones, lintel bands and knee bracings, are rare to ﬁ nd. Th us, despite the best intentions of the government, non-governmental and corporate organisations that participated post- disaster housing work in Marathwada, long-term disaster mitigation still has not occurred.

Part IV: Conclusions and Policy Recommendations

I have also observed, at many villages, and at many places, the curing was stopped within 2 or 3 days…. Th e people said no, there is no water, what should we do, we have to bring the water from long distance…. Technical wing of World Bank was visiting at that time, but that visit was so infrequent… World Bank also could not do anything good, and nothing more than pour in the money. Professor Gowande, Civil Engineer, Latur, 2001

In Latur what happened probably the contractor came in and constructed, the government gave the money...and they went in and constructed the whole house, which people didn’t use. Now, probably, they’re using it for grain storage or cattle sheds. United Nations Development Programme (UNDP) Engineer, Bhuj, 2001

Latur people wanted house, but in Kutch they want living house. Former PMU Junior Engineer working for UNDP, Bhuj, 2001

After a decade of lessons, it is not surprising that the MEERP is widely

Downloaded by [University of Defence] at 01:15 24 May 2016 viewed as an example of what not to do after an earthquake disaster. At relocated villages people are well aware of the housing abandonment problem, the poor quality of new construction, the general lack of conﬁ dence in their earthquake safety, the lack of proper training in how to maintain new concrete structures, and a sense of ‘dependency’ created by relocation. Likewise, the reconstruction in-situ programme has had a negligible impact on improving the MEERP, for nearly all of these villages were ultimately relocated. And the RRS programme too has left much to be desired. Rather than teach households and artisans 110 Ú Salazar and Jigyasu

how to repair and retrofi t their own homes, the programme promoted the construction of new earthquake-safe rooms in brick or concrete by petty contractors, contributing to the general distrust of traditional stone masonry techniques while leaving old houses standing without any retrofi tting work being done. In conclusion, one can say in fact that the MEERP has created so many social, cultural and economic problems that many people are now living in as bad or worse conditions than prior to the quake, and that they may indeed suff er a repeat disaster if a sizable earthquake were to occur. Th is physical danger that exists in Marathwada today, as described in this article is deeply connected with the social and economic vulnerability of the traditional local artisans: the loss of material and land resources, the degradation of traditional artisanal skills, the incom- patibility of external post-disaster interventions and the continued distrust of traditional building techniques, have reinforced their low social standing and reduced livelihood opportunities. Blaikie et al., it appears, had it correct nearly a decade ago when they highlighted that ‘Relief and reconstruction can aggravate divisions and patterns of inequity within a society. Social, economic, and political vulnerability are often reconstructed after the disaster, thus reproducing the conditions for a repeat disaster’ (1994: 210). While this article has done its best to outline how this happened, clearly much more fi eldwork is required to document the present condition of artisans in Marathwada — their economic and social standing as it relates to other communities at a village level, as well as more detailed knowledge about their history and involvement in the MEERP. Such a study would shed more light on the long-term impact of the MEERP, as well as help expand our understanding of vulnerability as it relates to housing disasters. As artisanal, castes are principally involved in building activities, understanding how they ‘anticipate, cope with, resist and recover from the impact of a natural hazard’ (ibid.) is of critical importance. Downloaded by [University of Defence] at 01:15 24 May 2016 In conclusion, one can also say that the work done in the region — by international donors, the GoM and the variety of corporate and non- corporate NGOs — off ers many lessons to learn, some being rather exemplary. Reports about this work must be acknowledged here, as they represent a body of knowledge that has shaped the evolution of post-disaster housing work in India, especially because many of the individuals and organisations connected with these reports remain active in the post-disaster housing fi eld. A few of the most relevant reports are: Lessons from Marathwada Ú 111

TARU (1995), TARU and YUVA (1995), ASAG (1998), and Nikolic-Brzev et al. (1999). It is also important to note that while many of these reports agree with the perspective of this article, some do not. While a healthy debate will no doubt continue, we off er here some fi nal comments, a short list of general guidelines for policy makers that are necessary for building appropriate post-earthquake housing in rural India, with the participation of local communities. Create Policies that Encourage Retrofi tting, Reconstruction In-Situ and Participation

z Set clear guidelines for the rapid assessment of village/household damage to avoid confusion and potential political pressures to re-categorise villages. Involve all village communities in this process, to ensure a buy-in and resolve diff erences if possible. Th is allows families to plan their future better and become involved in reconstruction activities. z Structure reconstruction options using fi nancial incentives that encourage communities to opt for retrofi tting and/or rebuilding in-situ, rather than relocation. z Require village-level feasibility studies that investigate retrofi t- ting vernacular buildings and in-situ reconstruction as a pre- requisite for agencies to design and build post-disaster housing and/or receive reconstruction funds. Create Collaborative Government–NGO–CBO Rehabilitation Projects

z Establish reasonable deadlines for the start of construction activity — years, not months — and make the process of housing and community participation the focus of reconstruction activities. Avoid ‘target’-oriented policies. z Minimise the use of large corporate donors and builders in recon-

Downloaded by [University of Defence] at 01:15 24 May 2016 struction activities; provide clear guidelines that allow corporate donations to fund participatory reconstruction by NGOs and CBOs. z Carefully assess the capacity, roles and responsibilities of government–NGO–CBO collaborations. Allow NGOs and CBOs time to develop processes that encourage community participation. z Create flexible policies and funding mechanisms that allow NGOs and CBOs greater control of project implementation, with appropriate degrees of decentralised production to tap local 112 Ú Salazar and Jigyasu

labour and material markets, and decentralised decision-making that allows for empowerment and the community organising activities in the ﬁ eld. Promote Local Artisans and Builders

z Make artisans and local builders the focus of all reconstruction activities; requiring extensive re-training programmes. Relax technical standards to make typical vernacular building methods possible, with appropriate earthquake-safe technologies. z Promote the formalisation of artisanal trades and local material markets through artisan certifi cation programmes and government mechanisms that enforce building codes that are relevant to construction activities commonly used in the rural housing sector. z Create repair and retrofi tting ‘confi dence building’ education campaigns that are integral to all reconstruction activities (in core and peripheral areas). Education materials and activities should be linked to built projects in the community to be effective. Require household participation in the reconstruction of their own homes; avoid classroom exercises that are theoretical.

Acknowledgements

Portions of this article are adapted from several previously written sources: Salazar (1994, 1999, 2001 [with P. Engle] , 2002a and 2002b) and Jigyasu (2001, 2002a and 2002b); see these source documents for more details. Alex Salazar would like to thank the Graham Foundation (Chicago, USA), Ahmedabad Study Action Group (ASAG) and Society for the Promotion of Area Resource Centres (SPARC) for their support during the initial phase of ﬁ eld research (1993–94); architect Amol Gowande and intern architect Venkatesh Th ota for their support in eldﬁ research

Downloaded by [University of Defence] at 01:15 24 May 2016 in 2001; and Peter Engel for his research on Pardhewadi village. Rohit Jigyasu would like to thank the Department of Town and Regional Planning, Norwegian University of Science and Technology, where he undertook his doctoral research on ‘Reducing Disaster Vulnerability through Local Knowledge and Capacity — the case of earthquake prone rural communities in India and Nepal’. He is thankful to many resource persons, including members of the local community, who helped in providing vital information during ﬁ eldwork. Th e perspective of this article, however, does not necessarily reﬂ ect their views. Lessons from Marathwada Ú 113

References

Aysan, Y. and P. Oliver. 1987. Housing and Culture After Earthquakes: A Guide for Future Policy Making on Housing in Seismic Areas. Oxford: Oxford Polytechnic. Aria, A.S. 1994. Retrofi tting of Stone Houses in Marathwada Area of Maharashtra. New Delhi: Building Materials and Technology Promotion Council (BMTPC), Ministry of Urban Development, Government of India. Ahmedabad Study and Action Group (ASAG). 1994. Repairs and Retrofi tting: A Manual for People of Marathwada. Ahmedabad: ASAG. ———. 1996. An Attempt at the Preservation of Vernacular Architecture of Marathwada in the Post Earthquake Period, prepared by Rajendra Desai for the 2nd Congress on Traditional Sciences and Technologies (27 December 1995–1 January 1996). Anna University, Chennai. ———. 1998. Latur Earthquake-Rehabilitation and Aftermath. Ahmedabad: ASAG. ASAG, TARU and Youth for Unity and Voluntary Action (YUVA). 1994. Baseline Socio-Economic and Participatory Design Exercise in Earthquake Aff ected Areas of Maharashtra. New Delhi: BMTPC, Ministry of Urban Development, Government of India. Aubrey, J. 2000. ‘Gender Issues in the Marathwada Earthquake’, in S. Parasuraman and P.V. Unnikrishnan (eds), India Disasters Report: Towards a Policy Initiative in Hazards. New Delhi. Oxford University Press. Awotona, A. 1997. Reconstruction After Disasters, Issues and Practices. Vermont: Ashgate. Baker, L. 1993. A Report concerning the Marathwada Earthquake. Latur: Oxfam India Trust. Blaikie, P., T. Cannon, I. Davis and B. Wisner. 1994. At Risk: Natural Hazards, People’s Vulnerability, and Disasters. London: Routledge. Caufi eld, C. 1996 Masters of Illusion: Th e World Bank and the Poverty of Nations. New York: Henry Holt and Company. Cernea, M.M. 1993. ‘Anthropological and Sociological Research for Policy Development on Population Resettlement’, in M.M. Cernea and S.E. Guggenheim (eds), Anthropological Approaches to Resettlement Policy: Practice

Downloaded by [University of Defence] at 01:15 24 May 2016 and Th eory. Boulder, CO: Westview Press. Coburn, A.W., J.D.L. Leslie and A. Tabban. 1984. ‘Reconstruction and Resettlement 11 years later: A Case Study of Bingol, Eastern Turkey’, in S. Schupisser and J. Studer (eds), Earthquake Relief in Less Industrialized Areas. Rotterdam: A. A. Balkema. Coburn, A.W., J. Petrovski, D. Ristic, I. Armillas, and N. Biering. 1990. ‘Earth- quake Reconstruction Programme Formulation Mission to the Islamic Republic of Iran, Mission Report and Technical Review of the Impact of the Earthquake of 21 June 1990 in the Provinces of Gilan and Zanjan’, United Nations Development Programme (UNDP), UNDRO/90/9, GE.90-03511. 114 Ú Salazar and Jigyasu

Dadekar, H.C. 1986. Men to Bombay Women at Home: Urban Inﬂ uence on Sugao Village, Deccan Maharashtra, India, 1942–1982. Ann Arbor: Center for South and Southeast Asian Studies, University of Michigan. Das, K. 2000. ‘Socio-Legal Intervention in Latur’, in S. Parasuraman and P.V. Unnikrishnan, India Disasters Report: Towards a Policy Initiative in Hazards. New Delhi: Oxford University Press. Das, K., J. Aybrevi and ASAG. 1998. Assessing Maharashtra Earthquake Rehabili- tation Programme. New Delhi: BMTPC, Ministry of Urban Development, Government of India. Davis, I., ed. 1981. Disasters and the Small Dwelling. Oxford: Pergamon Press. Fernandes, E. and A. Varely. 1998. Illegal Cities: Law and Urban Change in Developing Countries. London and New York: Zed Books. Government of India (GoI). 1993a. Report of the Advisory Committee: On Recon- struction of Houses Damaged in Maharashtra and Karnataka During the Earth- quake on September 30th, 1993. New Delhi: GoI. GoI. 1993b. ‘Map Showing Suitable Sites for Rehabilitation of Earthquake Aﬀ ected Villages of Latur and Osmanabad Districts, Maharashtra’, prepared by the Department of Space, GoI. Government of Maharashtra (GoM). 1993a. Proposal for Maharashtra Earthquake Rehabilitation Programme. Bombay: GoM. GoM. 1993b. Government Resolution December 3, 1993. Bombay: GoM. ———. 1994a. Earthquake Rehabilitation Policy of Government of Maharashtra. Bombay: Earthquake Rehabilitation Cell. ———. 1994b. Government Resolution June 6, 1994. Bombay: GoM. Gopalan, P. 2000. ‘Community Participation in Latur’, in S. Parasuraman and P.V. Unnikrishnan, India Disasters Report: Towards a Policy Initiative in Hazards. New Delhi. Oxford University Press. Hansen, A. and A. Oliver-Smith, eds. 1982. Involuntary Migration and Re- settlement: Th e Problems and Responses of Dislocated People. Boulder, CO: Westview Press. Haque, E. and M. Zaman. 1994. ‘Vulnerability and response to reverine hazards in Bangladesh: A Critique of Flood Control and Mitigation Approaches’, in A. Varley (ed.), Disasters, Development and Environment. West Sussex: John Wiley. Downloaded by [University of Defence] at 01:15 24 May 2016 Harms, H. 1982. ‘Historical Perspectives on the Practice and Purpose of Self-Help Housing’, in P. Ward (ed.), Self Help Housing: A Critique. London: Mansell Publishing. Hewitt, K., ed. 1983. Interpretations of Calamity: From the Viewpoint of Human Ecology. Boston: Allen and Unwin. Hewitt, K. 1995. ‘Sustainable Disasters: Perspectives and Powers in the Discourse of Calamity’, in J. Crush (ed.), Power of Development. London: Routledge. ———. 1997. Regions of Risk: A Geographical Introduction to Disasters. Essex: Addison Wesley Longman. Lessons from Marathwada Ú 115

Jigyasu, R. 2001. ‘From “Natural” to “Cultural” Disaster, Consequences of Post- earthquake Rehabilitation process on Cultural Heritage in Marathwada Region, India’, Proceedings of the International Conference on the Seismic Performance of Traditional Buildings, Istanbul, Bulletin of the New Zealand Society for Earthquake Engineering 33 (3), September: 237–42. ———. 2002a. ‘From Marathwada to Gujarat — Emerging Challenges in Post- Earthquake Rehabilitation for Sustainable Eco-Development in South Asia’, Proceedings of the Conference ‘Improving Post-Disaster Reconstruction in Developing Countries’, 23–25 May, University of Montreal. ———. 2002b. ‘Reducing Disaster Vulnerability through Local Knowledge and Capacity. Th e Case of Earthquake prone Rural Communities in India and Nepal’, PhD dissertation, Norwegian University of Science and Technology, Trondheim. Kreimer, A. and E. Echeverria. 1990. ‘Rebuilding Housing In Emergency Recovery Projects’, Policy and Research Division, The World Bank Environment Department, Paper No. 1990–17, World Bank. Mosely, P., J. Harrigan and J. Toye. 1995. Aid and Power: Th e World Bank and Policy-based Lending, Vol. 1. London and New York: Routledge. Nikolic-Brzev, S., M. Green, F. Krimgold and L. Seeber. 1999. Lessons Learned Over Time, Vol. II. Oakland: Earthquake Engineering Research Institute. Oliver, P. 1987. Dwellings: Th e House Across the World. Austin: University of Texas Press. ———. 1992. ‘Rebirth of a Rajput Village’, Traditional Settlements Review 111 (11): 13–21. Oliver-Smith, A. 1986. The Martyred City: Death and Rebirth in the Andes. Albuquerque: University of New Mexico Press. ———. 1990. ‘Post-Disaster Housing Reconstruction and Social Inequality: A Challenge to Policy and Practice’, Disasters 14 (1): 7–19. ———. 1994. ‘Peru’s Five Hundred Year Earthquake: Vulnerability in Historical Context’, in A. Varley, Disasters, Development and Environment. New York: John Wiley. Parasuraman, S. 1995. ‘Th e Impact of the 1993 Latur-Osmanabad (Maharashtra) Earthquake on Lives, Livelihoods and Property’, Disasters 19 (2): 156–69. Menon, P. and B. Bhaskar. 1994. ‘Building without Baker’, Frontline, 1 June Downloaded by [University of Defence] at 01:15 24 May 2016 1994. Parsa, A.R.G. 1985. ‘An Appraisal of Earthquake Resistant Housing Programmes in Iran’, MPhil dissertation, University of Newcastle, UK. Pugh, C. 1995. ‘Th e Role of the World Bank in Housing’, in B. Aldrich and R.S. Sandhu, Housing the Urban Poor: Policy and Practice in Developing Countries. New Delhi: Sage Publications. Razani, R. 1984. ‘Earthquake Disaster Area Reconstruction Experience in Iran’, in S. Schupisser and J. Studer (eds), Earthquake Relief in Less Industrialized Areas. Rotterdam and Boston: A.A. Balkema. 116 Ú Salazar and Jigyasu

Rich, B. 1994. Mortgaging the Earth: Th e World Bank, Environmental Impoverishment, and the Crisis of Development. Boston: Beacon Press. Salazar, A. 1994. ‘Tradition and Invention: A Study of the Types of Construction in the Earthquake Aﬀ ected Area of Maharashtra, India’. Self-published report for NGOs in Marathwada. ———. 1999. ‘Disasters, the World Bank and Participation: Relocation Housing after the 1993 Earthquake in Marathwada, India’, Th ird World Planning Review 21 (1): 83–105. ———. 2002a. ‘Normal Life After Disasters? 8 Years of Housing Lessons, From Marathwada to Gujarat’, Architecture + Design, January–February. ———. 2002b. ‘Th e Crisis and Modernity of Housing Disasters in Developing Countries: Participatory housing and technology after the Marathwada (1993) Earthquake’. Proceedings of the International I-Rec Conference ‘Improving Post Disaster Reconstruction in Developing Countries’, Universite de Montreal, http://www.grif.umontreal.ca/pages/irecconference.html (accessed 23 May 2009). Salazar, A. and P. Engle. 2001. ‘Lessons from Latur’, Architecture + Design, May–June. Scudder, T. and E. Colson. 1982. ‘From Welfare to Development: A Conceptual Framework for the Analysis of Dislocated People’, in A. Hanson and A. Oliver-Smith (eds), Involuntary Migration and Resettlement: Th e Problems and Responses of Dislocated People. Boulder, CO: Westview Press. TARU. 1993. Action Plan for Reconstruction in Earthquake Affected Region of Maharashtra. New Delhi: BMTPC, Ministry of Urban Development, Government of India. ———. 1995. Government and Non-Government Cooperation in the Field of Human Settlements: A Case Study of the Maharashtra Emergency Earthquake Rehabilitation Programme. New Delhi: TARU. TARU and YUVA. 1995. Tragedy Followed by Disaster: Th e Marathwada Earthquake and Rehabilitation Programme. New Delhi: TARU. Tata Institute of Social Sciences (TISS). 1994. Survey of People Aﬀ ected by Earth- quake in Latur and Osmanabad Districts. Bombay: TISS. Times of India. 1993. ‘7 projects to begin on Sunday’, 22 October. Turner, F.C. and R. Fichter, eds. 1972. Freedom to Build. New York: Macmillan. Downloaded by [University of Defence] at 01:15 24 May 2016 Varley, A., ed. 1994. Disasters, Development and Environment. West Sussex: John Wiley. Vatsa, K.S. 2001. ‘Rhetoric and Reality of Post-disaster Rehabilitation after the Latur Earthquake of 1993: A Rejoinder’. Unpublished paper. World Bank. 1992. Participatory Development and the World Bank. Washington, DC: Th e International Bank for Reconstruction and Development/World Bank. ———. 1994a. Memorandum and Recommendation of the President of the Inter- national Development Association to the Executive Directors on a Proposed Credit Lessons from Marathwada Ú 117

of SDR million to India for a Maharashtra Emergency Earthquake Rehabilitation Project. Report No. P-6262 IN. Washington, DC: Th e World Bank. World Bank. 1994b. Resettlement and Development: Th e Bankwide Review of Projects Involving Involuntary Resettlement 1986–1993. Washington, DC: Th e World Bank Environment Department. ———. 1998. ‘World Bank to Strengthen Implementation of Resettlement in Light of Study: Bank’s independent Operations Evaluation Department (OED) report shows widely diﬀ ering outcomes in sample of eight dam projects’. Press Release. Washington, DC: Th e World Bank Group. Downloaded by [University of Defence] at 01:15 24 May 2016 5 Women Take the Lead Turning a Crisis into an Opportunity for Development Prema Gopalan

Life, after disasters, is what happens to people, while the state is busy making other plans.

Disasters are devastating. An earthquake tears apart the social, political and economic fabric of communities and rents lives asunder. Th e rehabilitation of earthquake-aff ected communities is a challenging task. It calls for vision, innovation, resources and, above all, long-term commitment. Rebuilding lives and livelihoods, together with vital life- support systems such as shelter, healthcare, schools, ration shops, water supply, and sanitation may take years, depending on the scale of the disaster. Disasters aff ect vulnerable people the most, and during a disaster women are more vulnerable than men because they have to take on the added responsibility of looking after their families, the injured and the sick. Under extremely diffi cult and trying circumstances they care for orphaned children, the elderly and the disabled. However, men and women respond to a crisis in diff erent ways. Most women do not control the resources needed to tackle a crisis because of gender hierarchies and the division of labour. In fact, their workload increases tremendously due to the damaged infrastructure and housing, and poor sanitation and water facilities. Th ey have to use all their ingenuity and manage everything using dwindling fi nancial resources, which they do with Downloaded by [University of Defence] at 01:15 24 May 2016 astounding commitment and resilience. Th us, although women are often the fi rst to respond to a disaster, they are the last to be consulted in the planning of disaster response initiatives. And yet, wherever the state has provided for the involvement of communities, especially women, the latter have been at the centre of all important interventions. For this women need to be organised into collectives and enabled to participate in the decision-making processes within their own communities. Th e experience of the Swayam Shikshan Prayog Women Take the Lead Ú 119

(SSP)1 shows that right from the recovery stage, women need to be involved in rebuilding shelter and restoring basic services. Involving women who are already part of an informal network goes a long way in quicken- ing the speed and eﬃ ciency of reconstruction. SSP has helped unleash this positive force both in Latur and in Gujarat, after the earthquakes of 1994 and 2001 respectively.

The Yawning Chasm

The Latur Earthquake

Maharashtra, 30 September 1993: Th e Latur–Osmanabad region was struck by an earthquake which aﬀ ected an area of approximately 150 sq. km. and as a result of which 52 villages had to be relocated completely and nearly 1,300 villages had severely damaged houses. Th e disaster killed over 7,500 people and injured at least 16,000 others, and 200,000 houses required repair and strengthening. Th e Repair and Strengthening (R&S) programme to rebuild houses was initiated by the Maharashtra government in 1993 and lasted until 1998. On 30 September 1993, SSP was appointed Community Par- ticipation and Monitoring Consultant by the government for the R&S programme. Covering 1,300 villages and 200,000 households, the project was supported by the World Bank, UNDP, DFID, and other international agencies. At the local level, Mahila Mandals were recruited by SSP into the reconstruction eﬀ ort. After the earthquake rehabilitation project was completed in 1998, SSP looked to steer these women’s groups and communities involved in reconstruction towards a more broad-based community development strategy. Sunita Madhole lived in Chincholi village with her husband, two- month-old daughter and extended family. When their house collapsed that September night, her husband was killed and she was trapped in the

Downloaded by [University of Defence] at 01:15 24 May 2016 debris. Th eir baby was unconscious and seemed to have died. While her in-laws did not attempt to rescue her, some neighbours saved Madhole’s

1 SSP is a development organisation that partners with 5,000 savings and credit groups, comprising 60,000 women and local governments across 1,000 villages. SSP mobilises women’s groups to participate in the transition from supply-led to demand-driven service delivery in collaboration with Gram Panchayats (local governments). Gram Panchayats and women’s groups design plan and implement community-driven water and sanitation projects in 220 villages. 120 Ú Prema Gopalan

life by pulling her out of the rubble. She then discovered that her child had survived too. Sunita rose from the rubble and helped others rise too and they all rebuilt their lives and livelihoods. Sunita’s eff orts at rehabilitation were guided by SSP and she became an active member of her local self-help group. SSP (literally ‘self-education for empowerment’) redesigned the State’s project, taking care to ensure that grass-roots women’s collectives would be centrally involved in the rebuilding of homes and communities. At the core of SSP’s strategy is the concept of learning from peers. SSP believes that peer learning is the most eff ective means to transfer innovations across groups and communities. Th e idea was to disseminate relevant information and to motivate women to take charge of their own lives and help re-establish and regenerate their villages. As a fi rst step, the programme helped build awareness among women about their rights and entitlements, and provided them with basic training in simple, low-cost earthquake-resistant building technology. Soon women in Latur began to demand that the fi nancial aid provided to families by the government be deposited in joint bank accounts so that they too could have a say in how the money was to be spent. Th ey also began to supervise construction so as to ensure quality as well as convenience and, in the process, reduced such corrupt building practices that could compromise their safety.

The Story of Kachchh

Gujarat, 26 January 2001: An earthquake of 8.0 magnitude on the Richter scale with its epicentre at Bhuj damaged over 620,000 households in villages, towns and cities spread across several districts. Th e response was unprecedented. Th e media was full of reports of valour, sacrifi ce, tenacity, compassion and generosity from across the country and elsewhere. Re- lief work started and the State’s rehabilitation strategy was formulated. Downloaded by [University of Defence] at 01:15 24 May 2016 Lessons from previous rehabilitation eff orts, in particular the Latur earthquake, were at the core of designing sustainable policy initiatives and eff ective mechanisms that would work for the people. Th e Gujarat government fi nalised a reconstruction package for 7,904 villages and towns. It was a four-layer package that would cover reconstruction of collapsed buildings, as well as severely and partially damaged buildings. Women Take the Lead Ú 121

What does Post-Disaster Intervention Involve?

Th e three R’s of post-disaster intervention are:

(i) Relief (ii) Rehabilitation (iii) Reconstruction

Each of these phases oﬀ ers possibilities for building local capacities and reshaping the trajectory of development in the area. SSP’s experience after the Latur quake in 1993, the Turkey earthquake of 1999, and then again Gujarat in 2001, showed the importance of developing learning tools so that mechanisms of information and knowledge transfer allow inter-community exchanges and peer learning across established boundaries and cultures. Grass-roots women’s groups and organisations such as SSP and the Society for the Promotion of Area Resource Centres (SPARC) in India, together with international networks such as Grassroots Organisations Operating Together in Sisterhood (GROOTS) and coalitions such as the Huairou Commission are striving to facilitate strategies that open up spaces for the participation of women and communities in post-disaster settlements.

Hands that Soothe: Women and Relief

Th e earthquakes in Latur and Kachchh have shown that the eﬀ orts of women in the immediate post-disaster scenario include restoration of livelihoods using new survival strategies and rebuilding homes from whatever was left in the rubble. Organised initially by non-governmental organisations (NGOs) working in the earthquake-hit villages, women came out of their homes to engage themselves actively not only in

Downloaded by [University of Defence] at 01:15 24 May 2016 livelihood activities but also in local governance. Once they entered the public arena, they began not only to reconstruct their lives but also to address their vulnerabilities — the ones that always existed, as well as those created or made worse by the earthquake. Th e relief stage is characterised by a looming benefactor, the government and aid agencies, and hapless victims, and the atmosphere is full of misplaced sympathy and a charity approach. Th is phase may last from two months to two years. It calls for the laying of a foundation for a community-driven rehabilitation strategy before embarking on a 122 Ú Prema Gopalan

permanent rehabilitation plan. Special eff orts are needed to set up decentralised units, with personnel trained in disaster management in order to handle the fl ow of resources such that they can bridge the gap between the government and the aff ected people. With increased political consciousness, women are no longer hesi- tant in demanding their rights. Being the actual engines of recovery in most disasters, they display resilience and resourcefulness in extremely diffi cult conditions, and are self-reliant in proactive community work. To illustrate, there is the case of women in Jamsar. Th ey recall their early apprehensions regarding SSP workers. For instance, 70-year-old Laxmiben exclaimed:

Initially when these people came to our village we were terriﬁ ed. Th ey asked us to get united and save money. I was so scared. I used to hide behind the window and watch! We wondered why they wanted to come here to our village, what they wanted from us. But gradually we have understood the importance of what they were saying, the importance of savings, demanding our rights, speaking up. Today there are three Bachat Mandals in Jamsar. We have gained tremendous self-conﬁ dence (SSP 2002b).

Hands that Help: Women and Rehabilitation

Women-initiated activities, undertaken as an immediate response to a disaster situation (such as the distribution of supplies, shelter provision and community-support services), are often invisible and but rarely acknowledged. Th e end result is that women are excluded from information, decision-making and from designing aid programmes. But women have now broken these barriers. ‘We want to come out and do something on our own. We want a better life for our children’ (SSP 2002e). Th ese words belong to the women of Kachchh. Initially, they would intervene in local issues and enter mainstream decision-making processes with the help of NGOs and

Downloaded by [University of Defence] at 01:15 24 May 2016 community-based organisations (CBOs), but now the situation has changed. Th ey speak for themselves and are involved in day-to-day local governance, such as the institutions of Panchayati Raj, the Gram Sabha, monitoring basic services, getting training and visiting places from where they could learn and replicate methods of community development. Th is change is important because Kachchh is dominated by high caste communities and there has typically been male dominance in the public arena. Today, women in both Latur and Kachchh have a kind of power and spirit to be able to raise their voices within and without — inside Women Take the Lead Ú 123

the family and in village gatherings — and share in ownership through savings and credit schemes. People-to-People Exchanges During times of crisis, people need to interact with others who may have coped with similar disasters. When exchanges are supported between aff ected communities and those that have undergone similar trauma, the benefi ts are two-fold: the exchanges help local groups to fi rst formulate and strengthen self-help initiatives for their own rehabilitation, and later, they help initiate development in larger areas. Sharing Experiences Women from Latur travelled to Gujarat in January 2001 and shared their experiences on how they rebuilt their lives. Motivated by fellow earthquake survivors from Maharashtra, women in two districts of Gujarat also began to get involved in reconstruction, livelihoods, water supply, education and other essential aspects of the rehabilitation process. Bhanuben of Utbeth Champer in Jodia narrates her lessons:

Th e women of Latur have made a lot of progress after the 1993 earthquake. Th ey have mobilized and organized themselves and worked together. Th e more women work together in unity, the more they can move ahead in society (SSP 2002a).

An interesting illustration of the participatory process is the construction of a Dhobi Ghat in Jamsar. Anish, the engineer from SSP commented:

If we were to make the Dhobi Ghat on our own we would have made a raised platform since women in Maharashtra stand and wash clothes. But that would have been a blunder since the women in Gujarat prefer to sit and wash clothes (SSP 2002e). Downloaded by [University of Defence] at 01:15 24 May 2016 Building Communities SSP believes that the building of social infrastructure in the form of community resource centres can act as a powerful tool for organising women’s groups. Th ese centres provide a space for self-empowerment through a range of capacity-building activities. For instance, a situational analysis of housing and basic services can be conducted to understand local needs and lobby for the effective re-allocation of resources. Participatory resource assessment can be undertaken to understand 124 Ú Prema Gopalan

women and community issues and evolve community initiatives. Th ese centres can design community spaces for women and for children’s activities, keeping in mind concerns of income generation, health, savings and credit, etc. Th ey can enable community organisation of housing, co- ordinating the procurement of materials, selecting housing sites with the benefi ciaries, organising transport, masons, water and services. In doing all this, these resource centres would be facilitating and realising the mobilisation of women. As they become aware house owners, they will decide the design and features that they want in their houses and will also contribute their labour, skills and materials towards the reconstruction of their homes. And to that end, the centres can conduct training workshops to equip women with leadership and support skills, thus facilitating the mobilisation of communities. The Initial Approach Th e R&S programme was fi nally announced in June 1994, after an initial damage survey had been completed and a list of benefi ciaries drawn up. When the SSP team stepped into the Repair, Reconstruction and Strengthening (RRS) villages in September 1994, ‘they were confronted with a situation fraught with confl ict’ (SSP 1998: 15). Th ey conducted a fi eld survey of 400 house owners and discovered that at that point everyone was fi ghting just to get on the list of benefi ciaries. Th e survey also revealed that 80 per cent of the benefi ciaries preferred reconstruction (total rebuilding of their house) to repair and strengthening. Eff orts (in the form of discussions, pamphlets and street theatre) were made to try and convince homeowners that retrofi tting would be a safe option, but these were only marginally successful. Th e vast majority of homeowners who participated in the RRS ultimately elected to build one ‘safe’ room — a new room built with seismically resistant features — that would be attached to the original house. Th e original house, however, retained its vulnerability.

Downloaded by [University of Defence] at 01:15 24 May 2016 Change in Strategy By 1996 it had become clear that there were major problems with the community participation strategy. Communicating with individual house owners presented a signiﬁ cant challenge to the State government, particularly because most of the programme designers and administrators were located in Mumbai. Villagers were reluctant to participate in the programme. ‘Within villages, home owners needed advice on a daily basis to resolve problems — common walls, availability of land for Women Take the Lead Ú 125

construction, etc. Th e government-appointed village-level committee proved inadequate to generate enthusiasm among people, nor were they adequate to address problems on a daily basis’. In April 1996, a rapid appraisal of 10,000 households was conducted by SSP to determine why there had been little or no progress in construction. Th e results of the survey were used to redesign the programme. For example, the survey revealed that the people had in fact bought the reconstruction material already, but that no masons had been available. Th is led to the introduction of mason training programmes as well as the idea of contracting labour for 10 to 15 houses at a time. Th e new community programme was designed to capitalise on the strength of the village as a unit and was implemented in April 1996 as part of the Project Management Unit (PMU) action plan. It included:

z Disseminating information. z Education and skill-building for earthquake-resistant construction. z Motivating house owners to build earthquake-resistant houses. z Motivating house owners to utilise government grants for the reconstruction of houses within the given time schedule. z Forming collectives for construction management. z Settlement planning exercises and resource mapping. z House design and planning with women. z Solving problems of individual beneﬁ ciaries. z Setting up grievance redressal and feedback mechanisms. z Monitoring the progress of construction with the beneﬁ ciaries. z Producing plans and estimates after discussions with house owners. z Training masons in earthquake-resistant technology. z Preparing documents for the release of material coupons and the remaining instalments of aid money. z Study tours for exposure to collective construction processes. z Assisting Mahila Mandals with co-ordination and management. Downloaded by [University of Defence] at 01:15 24 May 2016

Hands that Cure and Rejuvenate: Reconstruction and the Empowerment of Grass-Roots Women’s Groups

Rebuilding public infrastructure provides an opportunity for the creation of a community with active participation from women. Crucial to implementing a scaled-up reconstruction programme in a settlement is 126 Ú Prema Gopalan

a community self-monitoring system that can be led by women’s groups. Monitoring ‘progress’ ensures the quality of construction and provides on-the-spot technical guidance to house owners, and SSP’s core principle behind organising women’s groups is that the agenda of reconstruction will also serve as the focal point for organising women and to address the practical and strategic concerns in their development. Th e mobilisation of women and being able to recognise and understand their role in the reconstruction process are probably amongst the most important achievements of any pilot intervention. Th e presence of women is much needed in the planning and implementation of reconstruction and other recovery processes, in decision-making structures and transitional governments. Women have to deal with local resource issues such as water, sanitation and health, and given a chance, they make it a point to raise these issues at forums where they have a collective voice. By providing such platforms to women’s collectives, Savings and Credit Groups (SCGs) have made the right move. Th is will encourage women to come forward, get together, and raise issues close to their lives. Th ese women’s groups address priority issues like credit, food security, water and sanitation, health, education, and social infrastructure. Th ey initiate demonstration projects, get the community involved in planning, acquiring skills through training, and increasing their participation in local governance. Th e latter being particularly important because it will enable them to have social responsibility and develop leadership skills. Th is social participation will help create a women-enabling economic climate, a suitable labour market, as well as a pro-women social and legal environment. Creating small enterprises will also enable them to increase awareness and understand social dynamics. In this context, several villages in Kachchh today are witnessing a collective but rather slow movement towards women’s empowerment. In the case of certain villages (in both Latur and Kachchh), with active NGO or aid agency involvement, women have been able enter

Downloaded by [University of Defence] at 01:15 24 May 2016 mainstream decision-making processes. Central and State laws that promote women’s participation in local governance aided this entry. Also, often, disaster response initiatives — such as training in earthquake- safe building — followed by livelihood activities that were supported by their own savings and credit schemes gave women a voice in their family, the neighbourhood as well as in formal village gatherings. It is for the ﬁ rst time that most of these women are coming out in the open, lifting their veils, one could say, to speak up for themselves and their communities. Women Take the Lead Ú 127

Th us, Muriben of Nagpur says:

[I]t has not been easy. We come from such a conservative background, our husbands are not happy that we go to Rapar (another village) to attend Taluka level meetings… Many of the men would build the house walls so high that we would never be able to look out, and nobody would be ever able to see us. Th us we would be completely cut oﬀ from the world outside our homes. We have to be able to break out of this situation. We cannot aﬀ ord to be shy or scared and sit quiet any more… (SSP 2002e).

The Latur Earthquake With the Repair and Strengthening Programme in progress in 1,300 villages simultaneously, project managers and government offi cials could not communicate with each benefi ciary directly. It was imperative, therefore, to ensure people’s participation and to strengthen their ability to handle reconstruction work with limited supervision. A need was thus felt to fi nd a village-based agency that could liaise between the administration and the communities. SSP discovered, through its fi eld surveys, that every village had at least one women’s group, but, usually, these groups were inactive. SSP therefore decided to take on the challenge of reactivating these groups and initiated the community facilitator programme. Members of the programme belonged to the women’s groups, who acted as an interface between the government and the communities. Community facilitators had the responsibility of supporting house owners in the more technical aspects of construction, including optimal accessing and utilisation of their entitlements, understanding earthquake-resistant construction and using appropriate technology and resources. Crucial to the community facilitator programme was the government’s decision to ‘empower’ women’s groups as dialogue assistants, to monitor village-wise reconstruction of houses. Th e women were thus defying long-held gender stereotypes and there was a signifi cant shift from their being restricted to the house and to

Downloaded by [University of Defence] at 01:15 24 May 2016 fi eldwork, to entering construction and assuming public roles. At the village level, local leaders also supported the participation of women fi rst in reconstruction and later in long-term development. SSP’s experience of working with a ‘critical mass’ of women’s groups from these 200,000 households revealed that it pays to listen to women. Which is why strengthening the participation of women’s groups and communities in reconstruction, rehabilitation and in their subsequent development in the quake-aff ected areas in Kachchh is at the core of SSP’s and its partners’ strategies. 128 Ú Prema Gopalan

The Kachchh Earthquake After the earthquake struck Gujarat, SSP, along with local Women’s Federations, shared its insights and experiences from Maharashtra with aﬀ ected communities in Gujarat. SSP’s resource pool for transferring lessons consisted of community and technical teams, and 300 leaders of federations who had actively worked for post-earthquake rehabilitation in Maharashtra. Th e organisation and the grass-roots women leaders brought in a range of strategies that created the right conditions for women and poor communities to participate in post- disaster rehabilitation, leading to people-centred development. Th e planning and demonstration of temporary shelters was in progress already in certain villages. Th e presence of women leaders from Latur allowed the Gujarati women the space to come and talk in small groups. Th ey learned how to clear and mark sites, and erect bamboos, they started fetching water and mixing cement and sand in an eﬀ ort to begin reconstruction. In the mapping of houses, women were quick to get involved with designing the interiors — where should the stove be kept, the churner for the curd, the shelves for the gods, and the kitchen, etc.

How the Revised Approach Helped

One important element of the revised approach was the reliance on Samvad Sahayaks, village communication assistants, to facilitate communication between the villages and the Government of Maharashtra (GoM) and to motivate villagers to participate in the repair and strengthening programme. Th e Samvad Sahayaks are women leaders from the various villages, appointed by the Mahila Mandals or other village- based organisations. Between April 1996 and March 1998, 300 Samvad Sahayaks were appointed in as many villages. Until this point the Mahila Mandals had not been active in the reconstruction programme. Th ey had

Downloaded by [University of Defence] at 01:15 24 May 2016 been involved in social and cultural activities, such as adult literacy and health programmes, as well as income generation schemes. Th is change in strategy represents something important, which is also supported by experience also from other earthquake rebuilding eff orts — existing CBOs are more successful at understanding community dynamics and culture, and are more eff ective, therefore, as organisations involved in recovery at the local level than structures like the village-level committees which are created specifi cally for earthquake recovery. Women Take the Lead Ú 129

As noted by the Additional Collector (district administrator) in Latur:

You might say that…it will be of little use to involve Mahila Mandals, it will be an academic exercise. Once the oﬃ cials saw the work, they were convinced. Th e Government has understood the importance of women in village development. Women were there monitoring the Gram Panchayat [village council]. Mahila Mandals can give a boost to all the government programs (SSP 1998: 36).

He added, ‘Th e Mahila Mandals facilitated co-ordination between offi cials’ (ibid.: 14). Initially, it took many meetings and village visits to convince the fi rst batch of 25 Mahila Mandals to appoint women as Samvad Sahayaks. Women were in fact ridiculed, ‘Oh! You have now become an engineer’. However, once women got involved, they were quick to learn how to negotiate with materials, designs and matching resources to the type of construction. Th e Sahayaks did not waste any time. Immediately after their appointment they conducted a house-to-house survey, and soon enough they knew all the benefi ciaries. Th ey met the house owners, spoke of construction techniques, negotiated support from the village committees and Gram Panchayats, and also demanded accountability from the government’s junior engineers. Th ey attended the meetings with taluka offi cials to draw attention to procedural problems (ibid.: 33). A wide range of techniques were used by the Samvad Sahayaks and Mahila Mandals to encourage village participation in the programme. Th ey thought of themselves as community resource teams, working together and also sharing ideas across villages. Some of their techniques and methods included:

z Setting up cluster units and collective purchasing committees, to make the purchase of building materials less expensive, and to share information easily. People at the cluster level could volunteer to supervise construction. Downloaded by [University of Defence] at 01:15 24 May 2016 z Peer learning through melavas, which are traditional village get- togethers, used for information dissemination, that helped enhance the visibility of Mahila Mandals as agents of village development. Government oﬃ cials were also invited to these melavas and through this interaction both were able to understand each others’ strengths and limitations. z Widespread information dissemination campaigns using techniques like melas (village cultural fairs), exhibitions, posters, workshops with village contact persons; the training of Gram Sevaks, 130 Ú Prema Gopalan

Sarpanches and all elected members by holding exhibitions on local market days; undertaking jatras or yatras (expeditions on foot, covering long distances to temples and other places of religious interest) to convey information on the programme; educating people regarding earthquake-resistant construction, as well as a 15-minute video motivating house owners, particularly women, to participate in house reconstruction. z Cluster-level dialogue workshops, held at a central location, with 5 to 10 villages, to assist benefi ciaries in discussing problems that they were not able to solve within their villages and to give feedback regarding their progress. Th ese meetings increased accountability between villagers through a comparison of progress, problems and solutions. At the village level, the police Patil, Sarpanch, head- master and other leaders were present. Th e Collector adopted this idea, called it a circle workshop and used it for grievance redressal. Banks were also included in these meetings. z Construction supervision workshops for Mahila Mandals and village volunteers, who could in turn educate house owners on the technical aspects of construction. According to SSP’s evaluation, people were aware of earthquake-resistant technology but they needed assistance in ensuring its practice. Th e junior engineers and masons were too few, and because of time constraints and the pressure to meet targets, they were unable to transfer knowledge effi ciently. Th e Samvad Sahayaks played a key role in bridging this information gap. For example, in March 1997, 25 Mahila Mandal teams in Latur district assisted the SSP at a week-long exhibition that was attended by over 30,000 people. Posters on earthquake- resistant construction, model houses, slide shows on construction techniques and video fi lms to motivate owners were among the techniques used. z For those benefi ciaries who were having trouble in mobilising Downloaded by [University of Defence] at 01:15 24 May 2016 labour. Th e Samvad Sahayaks advised house owners to contract labour for 10 to 15 houses at a time. z Th e Samvad Sahayaks, through training workshops, had learned about cost-saving design and construction and were able to suggest these to house owners. z Th e Samvad Sahayaks would sometimes initiate a project in one village and use it as a pilot or demonstration project for other villages to consider. Women Take the Lead Ú 131

z Th e Sahayaks took house owners to ‘best practice’ villages. Th ese ‘success stories’ were then used to motivate other villagers. Th ese exchange visits and study tours proved to be useful tools for sharing information and contributing to the general improvement of the quality of construction.

Th e strategy of involving women’s groups and CBOs proved to be very eﬀ ective in reaching the villagers and involving them in reconstruction. With the involvement of the Mahila Mandals the utilisation of the payment instalments also went up to 90 per cent in Latur and even higher in Osmanabad district.

As entire communities and leadership were mobilised to participate, individual house owners, especially women, were motivated. Contributions by individual house owners dramatically increased in terms of cash, labour, recycled materials and participation in ensuring earthquake resistant features (SSP 1998: 40).

Greater recognition and visibility was given to women’s participation through the involvement of the Mahila Mandals and Samvad Sahayaks. Th e programme clearly established the need for involving women’s groups in the planning, monitoring and implementation of the rehabilitation and reconstruction programme. According to one GoM oﬃ cial:

Th e Samvad Sahayak was useful because she was from the same village and everyone knew her. Her credibility was already established. People had negative perceptions of village government functionaries and they were unable to elicit participation of house owners and community groups. Th e beneﬁ ciaries try to give excuses but in front of the Samvad Sahayak they cannot do this (ibid.: 33).

Downloaded by [University of Defence] at 01:15 24 May 2016 Within the rehabilitation project, the Mahila Mandals also showed a great capacity to work not only on reconstruction but also on other development-related issues. As community organisations involved in cultural and social issues prior to the earthquake rebuilding programme, they were well positioned to advocate for broader development goals. Activities were taken up on many fronts — health, education, savings and credit, self-employment, water resource management, etc. Th e SSP report describes the community participation eﬀ ort in two villages, Wadala and Masobavadi: 132 Ú Prema Gopalan

More than a year after the project’s inception, Wadala and Masobavadi still stand as examples of community participation in the post-earthquake reconstruction process. People in these villages changed regarding issues of village development. Th ey no longer wait for the government to deliver but take the ﬁ rst step. Social relations have been altered between diﬀ erent social groups and between men and women. People now understand that by working as a group for the development of their villages they are stronger than individuals. In addition, the important role that women’s collectives can play in the development of the village is now widely recognized, and their work is not discharged as ‘women’s business’ anymore (ibid.: 45).

Th e crescendo of women’s involvement started with the mapping of their old houses and the planning of their to-be-reconstructed houses based on a functional use of space. Very soon mapping emerged as a powerful participatory tool capable of activating women’s groups. It became clear that the women should be given basic training in earthquake-resistant construction technology. Th e next step for women’s participation was to supervise the building site and make sure that masons were actually building what they had planned. Since women had attended meetings and seen the training of masons, they knew about Earthquake Resistant (EQR) features. Th erefore their supervision also ensured good standards for the quality of construction. A number of them contributed self-help labour and management of the building site, making sure that labour, material, water and money were always available when needed. Women used their practical sense and creativity and their knowledge of what is needed in the house to start introducing interesting innovations in their newly reconstructed houses. Marghaben, for instance, used her knowledge of the mason’s training programme to rebuild her own house. She maintained, ‘We used the beam–column technique but in place of the column we are using four iron rods tied together. Th is is simpler and cheaper’ (SSP 2002e).

Downloaded by [University of Defence] at 01:15 24 May 2016 Initially, reconstruction provided women a common concern and a subject for discussion during the meetings of their collectives. Th eir conﬁ dence thereafter increased and they became full-ﬂ edged active members.

Model Buildings

In the initial phase of post-earthquake housing rehabilitation, over 500 model houses were constructed, scattered all over the aﬀ ected area, to Women Take the Lead Ú 133

demonstrate cost-eff ective building techniques, the use of local materials and the incorporation of earthquake-resistant construction features. Th e objective of this exercise was to demonstrate modern-day improvements in traditional building practices, and, especially, to generate confi dence among residents about the use of stone and its by-products for house construction. Th e majority of the buildings, 400 out of 528 in total, were constructed in the two worst aff ected districts, Latur and Osmanabad, and the remaining were constructed in Solapur, Satara and Sangli. Th e model houses had diff erent types of plans and material options, e.g., stone masonry, stone-crete block masonry, hollow-block masonry and clay-brick masonry. Th is was a particularly eff ective ‘confi dence-building’ tool for the benefi ciaries, who were themselves afraid to reside in stone houses. Needless to say, the model houses were constructed with the seismic features incorporated and highlighted by painting the bond beams in bright colours. By and large, the model buildings were constructed at exposed locations, and in front of each building a board was put up which summarised in Marathi (the local language) the salient seismic features of each building.

The Pilot Demonstration Project

In the first year of post-earthquake rehabilitation, the PMU field staff had found it very hard to launch the RRS programme. In spite of considerable eff ort, the GoM fi eld engineers were not able to initiate the physical implementation of this community-managed and GoM-assisted programme. Th e benefi ciaries too had found it diffi cult to realise the time- bound nature of the rehabilitation, and to mobilise the materials and labour required for construction. Th e scope of the project was to assist the benefi ciaries in the construction/retrofi tting of at least one house in each village. Th e PMU engineering staff , particularly the junior engineers,

Downloaded by [University of Defence] at 01:15 24 May 2016 assisted the benefi ciaries in their eff orts to procure the materials and artisans needed for construction; some engineers even accompanied benefi ciaries to the material markets and assisted them in the selection of building materials, e.g., cement, steel, sand and bricks. Further, the junior engineers mobilised construction labour (masons) and oriented them towards earthquake-resistant stone/brick masonry construction. Th e benefi ciaries who constructed the ‘demonstration houses’ received full fi nancial assistance in cash and the project represented a major breakthrough in the RRSP implementation. 134 Ú Prema Gopalan

The Training Component

Training of Artisans

In order to ensure the successful transfer of knowledge regarding earthquake-resistant construction technology to the rural communities, the government relentlessly pursued training programmes for artisans in the earthquake-aﬀ ected areas. Early in the project preparation phase, the government realised that the number of traditional artisans (especially masons) working in the area before the earthquake (approximately 2 per cent of the population) was not adequate to meet the demands of in-situ rehabilitation of over 200,000 damaged houses. Moreover, the programme was to be managed by the local communities and, by and large, it was expected that the construction was going to be carried out by local artisans. Th e government therefore launched the following three mason training initiatives:

(i) Training unskilled labour: in June 1994, the GoM’s Directorate of Vocational Education and Training (DVET), launched training programmes for unskilled labour in the earthquake-aﬀ ected areas, sponsored by the Maharashtra Emergency Earthquake Rehabilitation Programme (MEERP). An existing network of 32 vocational training centres in the worst aﬀ ected districts of Latur, Osmanabad, Satara and Solapur was used for this purpose. Th e duration of the training was two months, and it covered four trades: masonry, carpentry, electrical works, and welding. Th e curriculum for training in masonry skills covered basic elements of stone and brick masonry construction, and elements of earthquake-resistant masonry construction. Th e duration of the training was two weeks per trade. In the period from May 1995 to May 1997, over 6,800 individuals were trained Downloaded by [University of Defence] at 01:15 24 May 2016 under this scheme. (ii) Training the traditional masons of Latur and Osmanabad: the PMU made an attempt to improve the skills of the traditional masons by imparting practical, hands-on training in earthquake-resistant construction technology for masonry buildings. Th e idea for this kind of training emerged during the implementation of RRS. Th e programme was proposed by the SSP’s community participation consultant in May 1995, and Women Take the Lead Ú 135

it started in November 1995. Th is two-day training covered the basic principles of stone masonry construction, the use of cement-based mortar, and the construction of key seismic features using masonry. In the period from November 1995 to February 1997, approximately 4,000 traditional masons were trained in Latur and Osmanabad under this scheme. Most of them participated in the RRS implementation. It is very important to note that in the villages where trained masons were doing construction an obvious improvement in the quality of masonry construction was observed; the masons were very proud of the training and liked to show their certifi cates to visitors. (iii) Hands-on masons training in the villages by the PMU engineering fi eldstaff : one of the most important initiatives was undoubtedly the training provided to traditional artisans by the PMU engineering staff working in the villages. Most importantly, junior engineers of the PMU played a very important role in educating the local artisans regarding improved construction practices. Th ey were the ones who taught the local artisans how to construct a seismic band — an important seismic feature rarely found in the area before the earthquake.

Training Women Masons

Till a few years ago, not a single woman from this region worked in the construction sector. But after the earthquake, women got acquainted with the construction sector by getting involved in the reconstruction of their own houses. Since there was an insuﬃ cient supply of labour for the reconstruction activity, women were seen as an alternative source of cheap unskilled labour. With the construction of the Mahila Mahiti Kendra (MMK) in Usturi, Downloaded by [University of Defence] at 01:15 24 May 2016 it was also decided that a major experiment had to be undertaken in training women as professional masons and putting together standardised training package developed to replicate the experience in other projects. Th e bulk of the construction of the MMK is the result of a seven-day training programme for women masons. Ten batches of eight women masons, selected from ten diﬀ erent villages (included Usturi), for a total of 80 women trainees, took shifts on the building site. 136 Ú Prema Gopalan

Training in Theory

Th e basic idea that shaped the training of women masons was that it had to be hands-on training. Right from day one, women had to be constantly on site, and theoretical sessions were limited to orientation meetings at the end of specifi c training sessions, or used to introduce any new technology. For the learning of new technology, such as the production of funicular shells and stone-crete blocks, a core group of women masons were trained. Th en the knowledge was transferred to other trainees through the resource team, and with the educational support of photos and stories from study tours. Th is foray into training women masons also had an important demonstration value, so that more such training ventures could be supported by public fi nances, and the concept of mobile building centres could be understood by both the people and the government. Finally, the lessons learnt from this experience could provide a corpus of information to be transferred to other Mahila Mandals and NGOs. Manjuben (35), for instance, refl ects on the process she was involved in:

I was on the supervision committee and ensured that the masons were using appropriate construction techniques. I have learned so much by being involved in the construction process. Today, I am able to build my house due to the knowledge that I have gained (SSP 2002a).

Training in Practice

Maintaining the basic idea of hands-on training, a number of methods, strategies, and tools were developed while the training was going on, in accordance with observations and requirements on the ground. Th e ﬁ rst day of training was organised in such a way that all the women could gain familiarity both with the building site and with all the ‘general theory’

Downloaded by [University of Defence] at 01:15 24 May 2016 behind the training. All 10 batches spent the fi rst day on site, making ‘human chains’ to carry material on site, preparing and clearing the site, mixing the mortar, carrying water, giving a fi rst try to the construction of walls, and above all chatting and exchanging information with the women supervisors from Usturi and the SSP team. Every evening, batches of 10–15 women volunteers from the Mahila Mandal would join the trainees in the human chain made to clear the site. Th is was certainly great fun, but also a very eff ective way to create a friendly atmosphere and allow women to exchange information. Women Take the Lead Ú 137

Antuba, the lead trainer for Versara village says:

Today, I am working on restoring the livelihood of the women in my village. I gather the women every day for the training programme. Our Darbar community is very conservative and would never allow us to simply sit and talk. But the training programme gives us a chance to sit and talk about our daily lives and help each other. We never used to come together like this before (ibid.).

Handing It to Them: The Role of SSP

Th e role of SSP continued to evolve from the beginning to the very end. To begin with, the organisation had an important role in initiating processes, facilitating grass-roots initiatives and activities, and mobilising women. It was involved in almost all the activities, but always had a participatory approach. So, the more women learned and became active, the less SSP was involved in the daily management of activities. By the end of the earthquake rehabilitation work, SSP’s role lay mainly in overall supervision, monitoring, documentation, the organisation of exchanges, exhibitions and melavas, in supporting women’s groups in their negotiation with the higher levels of administration, and in creating a forum for them to negotiate at the local level. Eventually, the organisation’s withdrawal was possible because of the increased self-conﬁ dence of women, their learning about the management of construction processes, and their realisation that they could take initiatives without waiting for SSP’s approval. Th ey realised that they could relate to the organisation as a partner. When a partnership is instituted, roles become clearer, balanced and shared. Th e highlights of SSP’s strategy were:

z Organising inter-community exchanges between Sakhi Federations in Latur–Osmanabad (with previous experience of disasters) to Downloaded by [University of Defence] at 01:15 24 May 2016 directly mobilise and conduct a dialogue with women in the aﬀ ected communities. z Participatory needs assessment regarding shelter and basic services. z Rebuilding people’s conﬁ dence through community awareness camps and meetings. z Building temporary shelters and providing employment through relief works. z Information campaigns on entitlements and the diffusion of earthquake-safe technologies. 138 Ú Prema Gopalan

z Organising women’s self-help groups to address social issues, including credit, health and education. z Training and empowering women’s groups to facilitate community- led management of reconstruction at the village level. z Creating dialogue forums between people and the government. z Demonstration of low-cost and region-sensitive building options through community-owned resource centres. z Building community resource centres to provide services in health, education, income-generation and vocational training. z Interaction with NGOs and the government to innovate and scale up eﬀ orts for community-based reconstruction (instead of a beneﬁ ciary approach).

Th is process can be seen as the beginning of a movement for change in a place where, often, high caste males exclusively dominate the public space. Without romanticising this slow, complex process of change with gender, economic, sociological and ultimately political overtones, it can safely be said that women in these villages of Maharashtra are on the path of progress.

Disaster Preparedness

Th ese valuable experiences with post-earthquake rehabilitation and reconstruction in Latur and Bhuj hold many valuable lessons for the management of and responding to future disasters. One such lesson is the importance of enabling NGOs with a track record of working to empower people at the grass-roots level to develop reconstruction and rehabilitation strategies that build on local knowledge, strengthen local skills and capacities, respond to local needs and contribute to local development. Th is process should involve women at the centre of the design and

Downloaded by [University of Defence] at 01:15 24 May 2016 implementation of post-disaster plans and programmes, as well as in long-term eﬀ orts towards appropriate and holistic development.

Lessons Learnt

Th ere has been much to learn from the experience of the Latur and Bhuj earthquakes, and the post-disaster relief and rehabilitation work that took place there. It is now understood that: Women Take the Lead Ú 139

z Reducing the economic vulnerability of women, and hence, their families, is a key mitigation measure that reduces potential losses from future disasters. z A long-term development perspective is critical, starting at the relief stage, in the allocation and use of resources in order to build local capacity rather than creating dependency. z It is also important to capture the momentum in the region early on to integrate women’s participation in post-disaster eff orts. z Grass-roots women’s groups can assume a range of roles effi ciently with some capacity building and support, and with much lower administrative costs than outside experts. Creating formal spaces for women’s groups to participate in post-disaster eff orts is necessary to both achieve scale and to sustain their initiatives.

Signposts

Th e Latur and Bhuj experiences inspire both hope and confi dence. Th ey suggest that it is necessary to do research and to exploit past experience to enrich one’s knowledge on disasters and their gender dimensions. Advocacy and lobbying to empower the capacity of governments, international agencies, NGOs, and society on the role of women in disasters and how they can be involved in rebuilding communities is another crucial aspect of the rehabilitation process. Networking with media and pressure groups to promote awareness among the community is needed to popularise these strategies of post-disaster management. Th e women of Latur and Kachchh know their rights and also know how to avail of them. Th ey are confi dent and they work assiduously for the betterment of their community. Post-earthquake Latur and Kachchh have thus shown that women today are a force to reckon with and that they are key players in all communities that are on the path of change. Th is transformation is refl ected in Jasuben’s voice from Khirsara, Downloaded by [University of Defence] at 01:15 24 May 2016 Rajkot:

Experience of SSP after the Latur earthquake (1993) shows that an eﬀ ective strategy is to support initiatives by empowering local communities. It showed that listening to women pays. Involving grassroots women’s groups centrally in planning of housing and infrastructure, restoring livelihoods and essential services resulted in maximizing beneﬁ ts for the poor communities (SSP 2002d). 140 Ú Prema Gopalan

References

Th e Asian Age. 2003. ‘Latur Lessons need to be implemented’, 8 October. Akçar, Sengul. 2001. Grassroots Women’s Collectives – Roles in Post-Disaster Eff ort: Potential for Sustainable Partnership and Good Governance (Lessons learned from the Marmara Earthquake in Turkey). New York: United Nations Division for the Advancement of Women. Ariyabandu, M.M. and Maithree Wickremasinghe. 2003. Gender Dimensions in Disaster Management. Colombo: ITDG South Asia Publications. Chaudhuri, Drimi. 2003. ‘Women Break Patriarchal Mould’, Th e Asian Age, 7 October. Enarson, Elaine. 2001. We Want Work: Rural Women in the Gujarat Drought and Earthquake. Quick Response Research Report No. 135. University of Colorado, Boulder: Natural Hazards Research and Applications Information Center. Fordham, Maureen, 2000. The Place of Gender in Earthquake Vulnerability and Mitigation. Anglia Polytechnic University, Chelmsford and Cambridge: Disaster Studies Project.. Gopalan, Prema. 2001. Responding to Earthquakes: People’s Participation in Recon- struction and Rehabilitation. Prepared for UN Division for the Advancement of Women (DAW) and International Strategy for Disaster Reduction (ISDR) Expert Group Meeting on ‘Environmental Management and the Mitigation of Natural Disasters: A Gender Perspective’, 6–9 November 2001, Ankara, Turkey. ———. 2007. ‘Creating Social Markets: Building Social Networks’, Microfi nance Insights, 4, September/October, http://www.microfi nanceinsights.com/ articles-new.asp?member=&id=194 Grassroots Organisation Operating Together in Sisterhood (GROOTS). 2009. Crossing Borders: To Learn, To Teach. Report of GROOTS Africa–Learning Exchange. Joseph, Ammu. 2003. ‘Learning from Latur’, Th e Hindu, Sunday, 26 October. Swayam Shikshan Prayog (SSP). 1998. Report on Community Participation Con- sultant in the Maharashtra Emergency Earthquake Rehabilitation Programme. Mumbai: SSP.

Downloaded by [University of Defence] at 01:15 24 May 2016 ———. 1999. Cementing a Future: Women’s Leadership in a Reconstruction Program. October. Mumbai: SSP. ———. 2000. Investing in Communities: Experiences of Swayam Shikshan Prayog. Mumbai: SSP. ———. 2001a. Beyond Reconstruction: Future Disaster Management in Gujarat: Conceptual Framework. Mumbai: SSP. ———. 2001b. From Crisis to Community Development: Community–State Partnership in a Post Earthquake Project in Maharashtra. Mumbai: SSP. ———. 2001c. Lessons From the Epicentre: Mainstreaming Women’s Initiatives in Disaster and Development. April. Mumbai: SSP. Women Take the Lead Ú 141

SSP. 2001d. Putting Safety in Peoples’ Hands: Promoting Safer Building Practices. Mumbai: SSP. ———. 2002a. Rebuilding Communities in Gujarat: One Year After the Earthquake. January. Mumbai: SSP. ———. 2002b. From Margin to Mainstream: Rebuilding Communities in Gujarat. January. ———. 2002c. Reality Check: A Community-Oriented Rapid Appraisal of Th e Gujarat Earthquake Emergency Rehabilitation Programme. April. Mumbai: SSP. ———. 2002d. Responding to Earthquakes: People’s Participation in Reconstruction and Rehabilitation. August. Mumbai: SSP. ———. 2002e. From Margin to Mainstream. October. ISDR. 2002. Gender Mainstreaming in Disaster Reduction. Panel presentation by Sálvano Briceño, Director UN/ISDR, to the Commission on the Status of Women, 6 March, New York. Weist, Raymond E., with Jane Mocellin and Dodo Motsisi 1992. Th e Needs of Women and Children in Disasters and Emergencies. Report prepared for the Disaster Management Training Programme of UNDP, New York, and the United Nations Disaster Relief Coordination. Downloaded by [University of Defence] at 01:15 24 May 2016 6 Compounding Disasters — First Natural, then Man-made Failed Interventions We Can Learn From Sandeep Virmani

Th ose who cannot remember the past are condemned to repeat it. — George Santayana

T here are three broad areas of rehabilitation and recovery after a disaster. Physical recovery of damaged housing and infrastructure, economic recovery, and psychological recovery, which includes both recovery from personal trauma and the re-establishment of social links. Most rehabilitation programmes focus much more on the physical recovery of housing and infrastructure, and not on livelihood recovery or psychosocial recovery. Th e reasons for this are obvious — it is both easier and more visible. But most rehabilitation programmes make a mess of this physical recovery as well. On the whole, the Gujarat government is seen to have come out of the Bhuj disaster with fl ying colours — particularly when compared with what happened after the Latur earthquake in Maharashtra. We need to give all the credit due to the Gujarat government in the matter of housing rehabilitation. Th ey took some bold decisions, executed them with confi dence, sustained the eff ort and followed through. We will fi rst look at some of the reasons for this success, and then at various areas

Downloaded by [University of Defence] at 01:15 24 May 2016 of failure.

The Roots of Success

Perhaps one important reason for the Gujarat government’s success was its decision to work with non-governmental organisations (NGOs). A signiﬁ cant policy choice was the recognition of NGOs as a large and important section of society that will implement housing. So a Failed Interventions and Compounding Disasters Ú 143

‘public–private’ partnership was designed with NGOs. Th is provided for the government to give compensation up to Rs 45,000 and the NGOs contributed any amount above this limit. Th is resulted in a working relationship between the government and the NGOs, and between NGOs themselves. Th e latter, besides partnering and infl uencing several policies, used the framework to bring a range of experiences to rehabilitation, and these in turn infl uenced its course and gave it maturity. We should, however, note one fl aw, and this is that while the idea of partnership was good, the houses ended up being built using only NGO money, while government funds were used only for other-than-housing needs. Much of the success also lies in the fact that after some initial hesitation, the government decided that housing rehabilitation should be ‘owner-driven’. Th e policy was also executed with a well developed implementation mechanism of cash-less transactions through individual bank accounts, material banks, technical guidelines, third-party technical monitoring, a campaign for seismic safety with NGOs, and so on. Th e third important initiative was the preparation and implementation of ‘development plans’ for the rehabilitation of towns. Th is required a lot of political will and vision to be able to carry it through. But perhaps the most important initiative was the setting up of the Gujarat State Disaster Management Authority (GSDMA) that gave continuity to the decisions taken in Bhuj, in spite of several subsequent disasters, including the riots in 2003. Another key reason for success, and one of the major achievements of the shelter reconstruction programme, has been the limited relocation of villages in rural Kachchh. Out of the 245 villages adopted under the Public–Private Partnership (PPP) programme only 18 have been relocated fully, and another fi ve have been partly relocated. Some of the relocation projects have been very successful. Others have been dismal failures. But 222 out of 245 villages have chosen reconstruction in-situ, and this is certainly a signifi cant reason for the success of the Gujarat

Downloaded by [University of Defence] at 01:15 24 May 2016 government’s housing rehabilitation programme.

Failures: Damage Assessment

Post-disaster, during the relief phase, the most diffi cult and important activity that has to be planned and executed is damage assessment. Im- portant because this becomes the basis of all fi nancial relationships, compensation packages and policies for the rest of the rehabilitation phase, which often lasts several years. Diffi cult because the administration 144 Ú Sandeep Virmani

is already overwhelmed and is grappling with the ineffi ciencies of relief management, and is at the same time under tremendous pressure from the world and politicians who want to quantify the damage for impact and be able to raise funds. Th e inexperience of doing such assessments more often than not makes the administration do a shoddy job, that immediately thereafter needs to be revisited, and provisions for a re- survey, in part or whole, have to be made. Th at is what happened in the Gujarat earthquake as well. However, the accuracy of a fi rst-time survey can never be achieved in re-surveys. In Gujarat, immediately after the earthquake, 600 engineers from various departments were given a quick training in categorising damage to buildings on a G1 to G5 scale.1 Th ey were also asked to measure the built-up area of each building. Compensation was calculated for a fi xed maximum area, or on the actual measurement, whichever was less. Th is area was multiplied by a fi xed rate for each category of damage. After the survey results were declared, and the people took some time to fi gure out the complicated categorisation and compensation calculation, the administration was fl ooded with complaints as people realised that for no perceptible diff erence in the size and damage, their neighbour was to receive much more compensation than themselves. As re-surveys were announced, this opened a Pandora’s box for another fl ood of complaints, not just for change of category but also claiming that their house had not been surveyed at all. It was soon possible to bribe the revenue offi cer for a fi xed sum of Rs 5,000 for a change of category, or Rs 10,000 for addition of names. Th erefore, houses that belonged to their sons, or even fi ctitious names could be added for full compensation. A broad calculation at one stage showed that more houses had been declared broken than the total families listed in the census in many areas. Two important lessons from this experience were that it was penny- wise and pound-foolish to have a complicated categorisation and varied

Downloaded by [University of Defence] at 01:15 24 May 2016 compensations calculated for each house separately. Two broad categories of either repair or new construction with only two compensation amounts would have made it easy for the engineers to assess and to avoid unhealthy comparisons in the village. It was also unfortunate because the ex-Collector of Latur, Praveen Pardesi, then heading United Nations Development Programme (UNDP) in Bhuj, kept proposing this simple categorisation to the Gujarat government, but his advice was not heeded.

1 For an explanation of the G1 to G5 categories, see n. 5 on p. 190. Failed Interventions and Compounding Disasters Ú 145

A quick assessment technique should be to outsource it. Or now, the newly-created State Disaster Management Authorities (SDMA) should have trained personnel, so that the politicians, fi nancial institutions and the world-at-large, which wants to donate to disasters, have some gross fi gures to go by. However, the detailed survey should take time and be done accurately so that this sacred process and the resultant document never comes into question. In any case, the results of this survey are only required after a few months, once the people are accommodated into their temporary shelters and the reconstruction or resettlement policies have to be designed. Th e lack of experience or improper advice actually compelled the Chief Minister, Keshubhai Patel, to announce that every family will have a single earthquake-safe room constructed before the monsoon, barely fi ve months after the earthquake! He showed not only a bad understanding of disasters and their management but of construction as well. And this dream forced the damage survey to be expedited unnecessarily. Here are some experiences of the damage assessment that was done in Gujarat:

z Th e fi rst government assessment survey put the number of totally damaged houses as 289,959, which in the re-survey increased to 317,181. z As per the fi rst survey the number of G4 category houses were 49,341 which decreased to 43,875 (less 5,516) in the re-survey, whereas the number of G5 category houses increased by 33,409 from 77,451 to 119,858 in the re-survey. Th e additional 33,409 G5 houses have been paid Rs 90,000 per family as compensation, which comes to more than Rs 300 crores! z Th e less-aff ected Nakhatrana and Abdasa talukas are where the maximum cases of corruption have been reported. In Nakhatrana the number of G5 houses registered were 328, which increased Downloaded by [University of Defence] at 01:15 24 May 2016 to 820 in the re-survey, whereas in Abdasa the number of houses increased from 1,090 to 1,570 in the re-survey. z A case of group corruption was reported where the money made by the entire village through the re-survey scam was partly spent for buying a second-hand Armada car for the re-survey team. z A 16-member family in Nani Khedoi village of Anjar has received 16 house-compensation cheques. In another case, a joint family of 15 members applied on separate names providing fake ownership documents claiming a total sum of Rs 718,760. 146 Ú Sandeep Virmani

z A group of Kachchhis, living in Mumbai for many decades, organised fl ight tickets for engineers appointed by the government for the re-survey, and fl ew them to Mumbai to explain the idea of making money by applying for a re-survey. About 50 Kachchhi families came to Kachchh and pulled down their partially damaged houses and got their house categories converted from G1 to G5. z Th e government issued recovery notices in Nakhatrana, Abdasa, Bhuj, Mandvi, Rapar, and Gandhidham worth Rs 11.25 crores in the re-survey scam. In Bhachau taluka a total of 82 notices were issued against 11 Sarpanches, 10 deputy Sarpanches, 57 Gram Panchayat representatives and four Taluka Panchayat members. z A Talati was charged for tampering with the aakarni register and inserting fi ve false names in the list. Th ese were the names of Kachchhis residing in Mumbai who were helped to get more than their entitlement. A criminal case was fi led against all of them. z In Taga village of Rapar, a total of 43 applications for housing compensation were sanctioned. But with the support of the Taluka Development Officer (TDO) and the government supervisor, 10 fraudulent applications were sanctioned. In this scam, nine residents of Taga village were arrested for claiming house compensations on fake names. Th e TDO and the supervisor were suspended. z In one meeting of the District Co-ordination Committee held under the Chairmanship of the then Collector, Mr Chibbar, he stated that compared to the census data, the number of applications received for compensation are much higher, which showed that there has been malpractice.

Failures: All NGOs are Not Equal

Let us begin with some survey ﬁ ndings, that give us feedback on how

Downloaded by [University of Defence] at 01:15 24 May 2016 NGOs performed. Th ree years after the earthquake, the rural reconstruction in Kachchh was more or less complete. Most homeowners had been staying in their new houses for six months to a year at the least. At this stage ‘Coming Together’ (a quarterly documentation of earthquake rehabilitation brought out jointly by Abhiyan, UNDP and GSDMA) carried out a satisfaction survey. One of the questions asked was whether the respondent was satisﬁ ed with his or her house. Of those who opted for the owner-driven policy, 90 per cent said they were satisﬁ ed. Of the owners who partnered Failed Interventions and Compounding Disasters Ú 147

with NGOs for their house construction 83 per cent said they were satisﬁ ed. Th e survey went on to ask another question to these satisﬁ ed respondents: ‘If you had a second chance to rebuild your house would you choose the same policy option (NGO or owner-driven)?’ and the answer was revealing: 61 per cent of the NGO-partnered respondents said they would reconsider partnering with an NGO. Th ey would opt for the owner-driven approach. Only 10 per cent of the owner-driven respondents wanted to reverse their decisions. Amongst the many reasons given for not working with NGOs were:

(i) Th e owners would have made much larger houses with the same amount of money. (ii) NGOs were not sensitive to details or to individual needs. (iii) Th e quality of construction was not as good. (iv) Th e NGOs insisted on relocation. (v) Th e NGOs used contractors.

Th e reasons given by 10 per cent of the respondents for wanting to work with NGOs was primarily that their government compensation was too little or because the contractors had cheated them. I still remember in the second month after the earthquake, the NGO co-ordination subgroup on housing brought out a recommendation that was given to the government, insisting that contractors should not be given reconstruction jobs of villages (which the Maharashtra Latur rehabilitation process had shown was a failure). Instead, the paper delineated the fi rst steps of what is today recognised as an ‘owner-driven’ policy which has been extremely successful. What is ironical is that the government’s owner-driven approach has been far more successful on all parameters that NGOs have raised in development. It had been far more democratic, allowing owners to be able to choose their materials, style and detailing; cost-eff ective because people did better management (largely dispensing with contractors), Downloaded by [University of Defence] at 01:15 24 May 2016 contributed more in terms of material (even recycling old materials) and labour to get larger homes; and empowering as they felt they had built their own homes, and communities came together to share work and rebuild their village. In the fi nal analysis there have been three types of NGOs:

(i) NGOs with vested interests (ii) NGOs that are ‘good contractors’ (iii) Participation- or empowerment-based NGOs. 148 Ú Sandeep Virmani

Th e fi rst category is interested in using the opportunity to make money or to fl aunt their work and gain acclaim, or may have religious objectives of emerging as powerful religious groups. Many such NGOs chose villages on the 80 km stretch between Bhachau and Bhuj. As all who enter Kachchh must travel on this road, to be able to showcase their ‘follies’ they even insisted on relocating the villages on the road itself, even if the village originally was only half-a-kilometre inside. Even before the owner-driven policy was announced Bhachau taluka became the NGO bazaar, where NGOs tried to convince villages and put up huge hoardings claiming they had ‘adopted’ the village. As a result of this naked exhibitionism, when the people realised that they had another option in the form of an owner-driven policy, the taluka rejected NGOs in huge numbers. As we enter the district the fi rst village we see is Vondh. Th e fl yover that brings you to the village aff ords you a panoramic view of about 1,000 houses lying waste. Rows of boxes meticulously laid out with only 10 houses occupied. Th e GoM adopted this village and involved a politician’s NGO in the construction of houses. Th e disputes began when the NGO insisted that the entire village be built on new land across the road. Th ey wanted to showcase a new township and it would be easier to construct, as the allotments would be made only after the construction. Th e village was divided in its opinion and fi nally, even as the new township was being built, people decided to forego their compensations and built their houses on their own old plinths. Th e complaints were that the new land was low-lying, the NGO had favoured a village leader by purchasing his land, the quality of construction was poor and allocation by lottery had broken their community. Contrast this with Manfara village of Bhachau taluka that was completely damaged in the earthquake. Th e village has 600 families, engaged mainly in small businesses. Not willing to join any NGO for the reconstruction of houses, the village people formed a committee and decided to take up house construction on their own. A trust was registered as Downloaded by [University of Defence] at 01:15 24 May 2016 Shree Manfara Nav Rachna Samiti, tenders were invited from contractors, engineers were consulted, house plans were prepared and land was purchased by the committee Trust. Th rough a government-subsidised bank loan 500 houses are being planned for 300 families, and for 200 families through self-fi nance along with housing compensation. Th e construction has been undertaken on a 70-acre plot of land purchased near the original village, and fi ve types of designs have been implemented depending on the needs of each family. Th ere is a provision for 100 more families for those who might wish to join later. Th e Trust has also invited Failed Interventions and Compounding Disasters Ú 149

donations for the reconstruction of a community hall, a hospital, a library and parks, a Gram Panchayat offi ce and Taluka offi ce. Another example of vested interests has been the township developed by Lions Club just outside Bhuj city. Th ey built 282 houses at a cost of Rs 105,000 each. As the township was built far from the city they found it diffi cult to fi nd benefi ciaries to occupy the houses. Finally, they got their Lions Club members to choose two benefi ciaries each. Th is selection was biased, and many who occupied the houses were not earthquake-aff ected at all, or had already taken their full compensation from the government. Th e main reason for this failure was because some infl uential members who owned land in this remote place and whose land value was low, wanted to sell some of their land and make some investments, so that the value of their remaining land would go up and this could come into the market. Th e contractor and architects were also infl uential Lions Club members. Th us the entire project was motivated to satisfy the members themselves. Th ere were many such cases in and around cities where the elite, who are landed brokers, took decisions, and projects ended up increasing their personal wealth. In comparison, the government developed relocation sites for the Road Transport Offi ce. Mundra Road and Rawalwadi have been extremely successful. Th e reason for this is that the people of the inner city of Bhuj were given a choice to relocate or stay within the city. All those who stayed within the city lost some of their original land and all those who opted for relocation got subsidised larger plots. Th is scheme was developed because the inner-city density had to be reduced. Th is choice allowed people to decide and largely those who had commercial interests within the city did not mind losing some of their land. Others chose to be in a well-laid out township with more space. Th e relocation sites were on government land chosen within the city. Th e government concentrated on providing good infrastructure like roads, sewerage, water supply, electricity, streetlights and shopping

Downloaded by [University of Defence] at 01:15 24 May 2016 arcades, giving conﬁ dence to the owners that the site would come up successfully. Th e owners therefore took the compensation and built their own houses. Many religious sects, like Mata Amritanand Mayi, Bochasanvasi Akshar Purushottam Swaminarayan (BAPS), Rashtriya Swayamsevak Sangh (RSS), World Vision, Hare Krishna, some Jamait groups, etc., have used the disaster and the money raised thereafter to further their religious ideologies. Not necessarily through direct conversion, as that would have attracted censure, but by promoting their organisational 150 Ú Sandeep Virmani

identities as ‘do-gooders’. More often than not they changed the names of the villages and put large boards proclaiming the name of the village after their religious saints, most of whom the villagers have barely heard of. BAPS even used the donation money to give special benefi ts to their followers, like whole new villages even if all of them had their own houses that were not even damaged! Examples of this are villages like Ukharmora and Gunatitpur. Th ese were home to rich Patels from Nakhatrana, Mundra and Mandvi who owned farmland in Bhuj taluka. Ukharmora and Gunatitpur were built by BAPS near their farms even though they had homes in Nakhatrana and could easily aff ord to make their own farm houses if they chose to do so. BAPS got into another embarrassing situation when they were questioned about the houses they had built, only for the Hindus of Khavda, who in turn had taken these houses and rented them out to others. Th e RSS also played a partisan role by going into Muslim-dominated areas and off ering to build houses only for Harijan Hindus. Many Harijans refused to take help from the RSS realising that this would create a rift between the two communities where otherwise they had been living peacefully. Th e second category of NGOs are the ‘good contractors’. A large number of international NGOs (INGOs) and Indian NGOs take contracts from donors and INGOs respectively to deliver housing. Th ese are all further subcontracted. Th is is the sector that attracts the new disaster professionals. INGOs have huge overhead costs, and give short contracts to professionals and consultants who fl y in and out of the disaster zones. Th e professionals charge handsomely, in the range of Rs 30,000 as starters, up to Rs 1–2 lakhs a month if they have experience and have delivered in earlier disasters. Th e costs of their construction has been in the range of Rs 400 to Rs 600 per sq ft, two or three times more than what people have built for themselves with comparable specifi cations.

Downloaded by [University of Defence] at 01:15 24 May 2016 And this does not include the cost of their expensive professionals and consultants. I was recently witness to one organisation (an INGO) that was in Indonesia after the tsunami of 2004. Th eir country co-ordinator was calling their head oﬃ ce in England asking them to identify and send an expert on tropical housing to Indonesia! Assuming that they did not think the local people competent to decide on the design of their own houses, Indonesia is also home to world-class architects who have given tropical architecture to the rest of the world. Failed Interventions and Compounding Disasters Ú 151

It is a minority of NGOs whose work is strongly based on the principles of participation and empowerment. Th ese have not only created excellent examples in rehabilitation work, but also act as the conscience- keepers of the government. Their contributions have been in the realms of policy, technology, cultural and environment sensitivity, village and neighbourhood planning, building peoples’ organisations, and infrastructure. It was in the NGO co-ordination meetings held in Abhiyan (a network of local NGOs) that the subgroup on housing drafted the policies with the then Collector, Mr Anil Mukim. Earth technologies were approved by Dr A.S. Arya, technical advisor to Government of Gujarat, to produce one of the world’s largest mud-based constructions. Artisans supported by NGOs built over 100 villages using earth. And this in spite of the fact that about 30 to 40 HUDCO building centres (for promoting appropriate building materials) were sanctioned, crores spent and most of the equipment was not used at all.

Failures: Rubble Removal

Every disaster results in rubble that needs to be cleared. It is also one of the ﬁ rst activities to be taken up after relief. Th e questions that surround this issue are:

z Where should the rubble be dumped? If unplanned the rubble is normally dumped into the nearest depressions on government land. Th ese are invariably natural drains or ponds. z How to quantify the rubble? If contracts for its removal are to be given on what basis should payments be made? z Can the rubble be reused? As embankments, building blocks, etc.

Removing debris by machine was in any case a big task. As it happened, 2001 was also a drought year and people needed relief. Abhiyan suggested

Downloaded by [University of Defence] at 01:15 24 May 2016 a debris removal programme by which people would remove their own debris and get paid for it. Th e programme did not work because direct relief work was still continuing. People were still being given grain and food, so no one was interested in working to earn enough to live on. Another factor was that people did not want to remove the debris until the damage assessment was redone — the debris was the only proof they had of the damage their houses had suﬀ ered. So ﬁ nally the government awarded contracts for debris removal. Th e system designed was simple. Th e Talati and Sarpanch of each village had to sign a document saying so 152 Ú Sandeep Virmani

many tractors of debris had been removed from the village, and against these signatures, payments for debris removal would be made. Th is became a huge racket. In fact, the corruption chain went all the way up to the Deputy Collector (who was also acting Collector for a while), who was eventually jailed for several months. Th e whole programme was a scam worth about Rs 600 crores. Finally, the government had to institute an enquiry, and people were chargesheeted and arrested. By the time the rubble in the city of Bhuj was cleared the government was able to place their own offi cers and put a better monitoring system in place. Th ere was no corruption, however, the government ordered the rubble to be dumped into one of the oldest lakes of Bhuj, Pragsar! Let me explain the ramifi cations of this folly. For arid Kachchh, the erstwhile kings (of the Jadeja dynasty of the seventeenth century) had developed an intricate water management system for the city. In the early seventeenth century, during the reign of Pragmalji II, three rivulets were connected by tunnels, gates and causeways to bring water into three lakes in the city of Bhuj. Th ese lakes were strategically located on the sandstone aquifer that was periodically recharged and ensured a perennial supply of water to the entire city through wells. On the one hand the Prime Minister’s offi ce was developing an elaborate plan with Abhiyan to revive this water harvesting system, while the other arm of government fi lled this lake with dumped rubble. A cost estimate for emptying the lake of rubble once again, came to a whop- ping Rs 5 crores, which could not be justifi ed, and so the city lost a 450-year-old water source. Most countries today, have their environment departments carry out ‘Environment Impact Assessments (EIA)’, based on which recommendations are made. Th e implementing authorities, however, do not take these very seriously. In Gujarat too, the Gujarat Ecology Commission (GEC) prepared an EIA, but they too overlooked the Pragsar dumping. Th ere was also extensive corruption with regard to rubble removal. Th e rubble removal scheme was meant for villages which had 50 or more Downloaded by [University of Defence] at 01:15 24 May 2016 damaged houses, and for debris removal from every house Rs 1,500 was to be paid. Gram Samitis were authorised to certify the removal of debris in their respective villages. Th e Samitis included the Sarpanch, Talati, the village teacher, Engineer (Bhukamp), and a village leader. Th is notifi cation was made in November 2001.

z Th e debris removal scam has been reported to be the ﬁ rst ever instance of corruption cases being registered on such a huge scale in the history of Gujarat, wherein complaints have been openly Failed Interventions and Compounding Disasters Ú 153

launched against a whole range of offi cials, from lower class employees to offi cer-level staff to village-level workers to Sarpanches and Jilla Panchayat members. In Bhuj taluka alone Rs 1.36 crores have been allegedly misused (Kutch Mitra Daily, 28 April 2003). For this scam, 203 criminal cases were fi led against government offi cers, employees and other stakeholders in the Anti-Corruption Bureau (Ahmedabad). The list also included 64 supervisors, 61 teachers, 79 Sarpanches and other leaders (see Table 6.1).

Table 6.1: Number of Cases of Corruption in Debris Removal Registered in Gujarat

Place Particulars Date of article reporting Bhuj 203 criminal cases fi led in 112 villages 28 April 2003 Rapar 183 criminal cases fi led in 73 villages 1 January 2004 Bhachau 1,888 criminal cases fi led in 70 villages 4 September 2003 Anjar 141 criminal cases fi led in 43 villages 20 January 2004 Source: Kutch Mitra Daily, 28 April 2003.

Failures: Economic Recovery

Populations from arid areas have complex and diffi cult ways of making a living. In Kachchh, where droughts force people to migrate most of the time, rain-fed agriculture is not a thriving economic activity. Like in many arid areas of the world, traditional crafts and skills are important sources of livelihood. Several eff orts in the past few decades have tried to focus on making this non-farm activity into a full-fl edged enterprise. Embroidery, block printing, weaving, leatherwork and many such crafts have a complex relationship with the market however, and supporting only production does not translate into business. Also, the cities of Kachchh are primarily based on trade and small business. Th ey work with small margins and their viability, as a result, is strongly linked with

Downloaded by [University of Defence] at 01:15 24 May 2016 traditional social linkages and locations in the market. If the production process and the markets are well established and a disaster takes away the means of production, the simplest rehabilitation strategy is to provide fi nance for the entrepreneur to re-establish his or her business, or give them the tools they need to restart production. Th is was what the government tried to do; they made grants-cum-loans and made toolkits available for the aff ected families. But the list below will show how the simplest of tasks become so diffi cult, when the basic intention to help rehabilitate the victims is not really there. 154 Ú Sandeep Virmani

Toolkits Distributed to Masons

Th e District Rural Development Agency (DRDA, most government schemes are implemented through this society chaired by the District Development Officer) invited NGOs to conduct mason training programmes for seismic-safe construction and the incentive for attending the course was a ‘toolkit’. No one asked if masons had lost their tools in the quake, and no one even asked if the person attending the course was a mason at all! Further, the toolkit for the mason contained tools for labourers as well, many of which found their way back into the market.

Weavers Given Looms to Restart their Business

A visit to Bhujodi, the largest weaving village of Kachchh, shows that many families display a weaving loom in their drawing rooms. On enquiring you are told that they do not use the loom because:

(a) Th ey have their old looms which they did not lose in the quake. (b) Th ey use pit looms but they were given standing looms. (c) Th ey are not weavers at all! Th ey had given up weaving long before the earthquake.

Farming Kits Distributed

Kits containing seeds, fertilisers, pesticides, spray pumps, etc., worth Rs 10,000–15,000 were distributed to small and medium farmers. Most of these kits were never used simply because hardly any farmers lost any of their equipment and the seeds, pesticides and fertilisers distributed were not co-ordinated with the farmers’ existing farming practices. Downloaded by [University of Defence] at 01:15 24 May 2016 All the departments, without proper damage assessment, got their share of the ﬁ nancial booty to squander. Th e list of toolkits gets more and more ridiculous. One department is known to have distributed a kit containing a bicycle, ladder and an umbrella to each artisan to go and do business and protect himself from the unbearable heat in Kachchh! Th e chain of corruption in buying toolkits still involves purchasing the tools, but an even simpler route to corruption lay in providing grants and loans. For people to restart their business grant-cum-loan schemes were Failed Interventions and Compounding Disasters Ú 155

developed between Rs 1 lakh to Rs 1 crore, with a 40 per cent loan and 60 per cent grant component. A committee involving the Sarpanch, Talati and schoolteacher along with others had to recommend genuine cases and the District Industries Corporation, in co-ordination with banks, was to distribute the money. A CBI inquiry had to be instituted to ﬁ nally stop the large-scale corruption that all, including the banks, became party to. Some of the newspaper reports of that time will give an idea of the scale of corruption and the way it was taking place. Th e impunity with which the false cases were prepared can be judged from the fact that 20 applications suggest that they are running successful beauty parlours in Banni. Incidentally, Banni is a remote region of Kachchh where traditional nomads reside! But the tragedy of this situation was that thousands of small businesses did close down permanently and the families were reduced to menial labour. Many families who were not coming forward to collect their housing compensations, when contacted by Abhiyan, said they did not have the time to be able to build homes, they had to earn and be able to eat every day. Th eir request was that instead of a grant of Rs 150,000 for a house, if they could get a loan of Rs 20,000 to Rs 25,000 they could restart their small shops that gave them an earning of Rs 7,000–8,000 a month and could stop doing menial labour. Th ey would, in time, be able to make their houses and bring back their families, who had been left with relatives and friends in other villages. Th e irony of the situation was that the government was pestering him to take his compensation of Rs 150,000 for his house but could not give him Rs 20,000 for his business!

Failures: Planning

Th e planning of Bhuj city has also had a strong housing bias, without understanding the economy that drove the city. In the reconstituting of

Downloaded by [University of Defence] at 01:15 24 May 2016 plots, commercial areas and businesses have been disturbed. Th e larger businessmen were able to re-establish themselves by buying or renting commercial space in prime locations like Hospital Road, but the small businesses collapsed. Th e government too built over 1,000 new shops but is seeing this as a commercial proposition where a developer and the Bhuj Area Development Authority (BHADA) will sell these shops at market prices. Obviously the original small shop owners will only add to the slum stock of the city. Somewhere down the line the new city forgot that this too was rehabilitation. 156 Ú Sandeep Virmani

Some of the other scams are listed below: (i) 60–40 Loan Package: As reported in the Kutch Mitra Daily (19 September 2002), 41 fake applications for loan support under the Trade and Commerce package were rejected. In an inquiry undertaken by the district-level authorities, 150 cases of fraud applications were recorded in the entire district. (ii) Criminal case fi led against Bhuj Commercial Co-operative Bank for a scam of Rs 3.25 crores in the 60–40 loan package declared by Cottage Industries–Government of Gujarat. As reported, the Bhuj Commercial Co-operative Bank granted loans to 539 fake applicants under this package, thus taking undue advantage of the 60–40 package. Th e package provided a fi nancial package in which 60 per cent of the funds were given as a grant and 40 per cent was a loan. Th e package was implemented by the District Industries Corporation (DIC), who routed the money through banks. For this, the Bhuj Commercial Co-operative Bank had submitted fake utilisation certifi cates to DIC and had opened fake loan accounts. Th e bank put the 40 per cent amount as fi xed deposits to show recovery of loan, and pocketed the 60 per cent grant as they were fake applications. In a few genuine cases, the applicants previous due amounts before the earthquake were settled with this scheme. Eight members including the Chairman, Directors and Managers of the bank were charged in this case. (iii) Criminal case fi led against Gandhidham-based business fi rms for providing fake documents to try and prove more damage to infrastructure than had actually been caused. As reported, Satyam Engineering Works had applied to avail of a loan worth Rs 200,000 under the government package. Th e company’s Mumbai-based chartered engineer presented a fake certifi cate for the purchase of new machines, although the old machines were repairable. In another case, Vision Opticals applied for a loan to compensate the damage suff ered by four machines as against the actual damage of three machines. Th e fake certifi cate was prepared by the company’s Ahmedabad-based chartered engineer. Downloaded by [University of Defence] at 01:15 24 May 2016 Failures: Temporary Houses or Permanent Slums?

It is very important to recognise that temporary facilities eventually become permanent. As part of the relief programme in Bhuj 4,500 prefabricated temporary houses were set up. Th e richer people took possession of these houses, and then rented them out to others. Th ey Failed Interventions and Compounding Disasters Ú 157

themselves either moved in with relatives, or left the city temporarily. So it was the poor, who had no other option, who moved into this temporary housing. Th ose who lived in rented housing before the quake also had to move in here, because they would get no compensation. A large percentage of the houses were given on rent to outside people who came in to build the city. Now, however, everyone wants to live in these houses permanently. Th ere are demands for a proper sewerage system. Th e whole area has turned into a slum. As the people here have come from all over the city there is no community feeling either. It would have been much better to let people decide where they want to re-house themselves, and do that in cohesive groups, rather than resettling them in barrack-like regimented housing where families of diff erent cultural backgrounds and from diff erent income groups are mixed up and share no social connections with each other. We need to begin by accepting, for instance, that a town will take time to rebuild. It is best to give people temporary housing options: let the communities themselves decide where to locate themselves. For example, in Aceh, Indonesia, after the tsunami, in 20 villages temporary houses were set up on Masjid land. Th ese are sure to be vacated when the occupants are permanently housed because the group functions as a community. In Bhuj, by contrast, people have to travel from their temporary housing 5 kilometres into the inner city to their workplaces. Th e important point to understand is that even interim processes need to be handled carefully, otherwise you can end up with people being dissatisfi ed and eventually breathing down your neck wanting the situation, somehow or the other, to be rectifi ed.

Conclusion

What has been set out above are some of the numerous failures that occurred in the Bhuj reconstruction work. Perhaps there is something to Downloaded by [University of Defence] at 01:15 24 May 2016 learn from each of these. Yet, by and large, the resettlement programme of Kachchh is perceived as a success, particularly when it is contrasted with what happened in Latur. Th e key to this success is perhaps the fact that much of the programme was owner-driven. And essential to managing such a process was the presence of committed NGOs, who were present and active in the area even before the disaster happened. Th at in itself would not have been enough though. Th e achievement in Bhuj was that these NGOs were able to work together, independent 158 Ú Sandeep Virmani

of but in co-operation with government, through a separate voluntary group that co-ordinated everyone’s work and brought it together. Th e key elements thus were: a reconstruction process driven by the victims themselves; NGOs that brought in external expertise, but who had been in the area earlier and were known to the victims; a willing and responsive government; and a neutral organisation trusted by all that co-ordinated everyone’s work. Downloaded by [University of Defence] at 01:15 24 May 2016 7 Planning the Reconstruction of Bhuj B.R. Balachandran

Nearly a decade has passed since the devastating earthquake that hit Bhuj on 26 January 2001, and while much has been achieved, many problems still persist. Th e reconstruction of Bhuj has been one of the most important learning grounds for post-disaster reconstruction in recent history. Th is article is an attempt to capture both the process adopted in Bhuj and the valuable lessons learned. Post-disaster reconstruction eff orts are almost universally characterised by certain generic problems. Th e immediate aftermath of the disaster being dominated by fear, leads to ambitious and often irrational proposals for relocating entire settlements and for introducing sweeping and stringent laws and regulations. As time passes, this fear subsides and the focus shifts to more immediate problems. Political expediency, rational thinking, public sentiments, professional advice, and bureaucratic procedures all start pulling in different directions and, eventually, the entire reconstruction process degenerates into a marketplace of ad hoc and shortsighted decision-making based on a largely political give and take. Th e reconstruction process in Bhuj was refreshingly diff erent in its conceptualisation and implementation. However, it was not without problems. In fact, the generic post-disaster problem of the trade-off between speed and accuracy was very much in evidence in Bhuj also. Th e most redeeming aspect of the reconstruction process in Bhuj was

Downloaded by [University of Defence] at 01:15 24 May 2016 the lofty vision and sheer commitment of the State government to use the city’s reconstruction to set an example. A demanding yet very co-operative citizenry made the government’s eﬀ orts tougher but also fruitful. Th e most diﬃ cult challenge of all was posed the abysmal state of property records and maps, and the systems needed for updating them. 160 Ú B.R. Balachandran

Th e absence of eﬀ ective mechanisms to create an interface between the government and the people compounded the problems. In our experience, the political dynamics of the reconstruction process unfold in a fairly sequential manner. Th erefore, the contents of this article are organised in a chronological sequence under heads corresponding to the stages in the reconstruction process in Bhuj, namely:

(i) Putting a process in place. (ii) Preparing a development plan for the entire city. (iii) Preparing detailed plans for the Walled City. (iv) Implementing the plans.

Putting a Process in Place (February 2001–May 2001)

Th e extent of devastation that happened in Bhuj, particularly in the old city, was unprecedented in the history of urban India. With no past experiences to learn from, putting in place a reconstruction strategy required careful thought. In comparison, for rural areas, the experience of earthquakes at Latur and Uttarkashi was available for reference. In fact, oﬃ cials and experts who had worked in those areas were brought in immediately to advise the government in structuring the rural reconstruction process in Kachchh. Despite the enormity of the challenge, the complexity of problems and the complete lack of previous experiences in India, the government succeeded in conceiving and operationalising a comprehensive strategy in a short period of three months.

Total Relocation vs Reconstruction

Downloaded by [University of Defence] at 01:15 24 May 2016 In the immediate aftermath of the earthquake, the discussion, both among citizens and in the government, revolved around two drastic alternatives — total relocation of the city (‘New Bhuj’) and in-situ reconstruction. Th ere were vocal proponents of both approaches amongst the public and amongst government oﬃ cials. (Studies conducted in post-disaster situations indicate that public opinion varies steadily over a period of time following a disaster — while initial reactions are motivated primarily by fear, gradually other practical considerations take over.) Th e government decided to go about this decision in a rational Reconstruction of Bhuj Ú 161

and scientifi c manner, and at the same time take public sentiment into consideration. One of the proposals was to turn the old city into an earthquake museum and build a new township for all inhabitants and businesses. Apparently, offi cials responsible for managing the situation in Bhuj at that time had drawn up a formal proposal of this nature. Th e urban design faculty at the School of Architecture, Centre for Environmental Planning and Technology (CEPT), Ahmedabad studied the situation and recommended, in a presentation to the government, that the old city be rebuilt, preserving as many of its historic buildings as possible. Environmental Planning Collaborative (EPC), a non-profit professional organisation engaged in the USAID-funded ‘Initiative for Planned and Participatory Reconstruction’, reported that public sentiment was in favour of rebuilding the old city as it was the economic as well as sociocultural hub of Bhuj. In fact, there were public protests in Bhuj led by business groups and professionals demanding that the old city be rebuilt. Across the world one can fi nd examples of entire cities being rebuilt from rubble (such as European cities destroyed in World War II or the recent example of the earthquake-devastated city of Kobe in Japan). However, rarely if ever would one fi nd a successful example of a city, particularly one with historic signifi cance, being wilfully abandoned and a new one being built elsewhere, forcibly relocating an entire population and its activities (even in disaster-prone areas). Having carefully considered all the options, in April 2001, the government formulated a reconstruction package for the aff ected urban areas of Gujarat, with separate sections for Bhuj, Bhachau, Anjar, and Rapar. Problems Faced and Lessons Learned Th e debate on whether to relocate or to rebuild was redundant, as it should have been obvious that there would have to be a combination

Downloaded by [University of Defence] at 01:15 24 May 2016 of measures such as relocation, redevelopment, regulated expansion of the city, etc. To this end, the task of assessing alternatives and proposing an approach should have been handed over immediately to a group of experts and the scope of work framed for planning the reconstruction. Th e latter could have been worked out in two stages: the rstﬁ being the overall strategy and the second, the physical planning work for which the scope of work could be detailed on the basis of the overall strategy. Th is would have saved at least three valuable months in the overall process. 162 Ú B.R. Balachandran

The Urban Reconstruction Package

Th e urban reconstruction package announced in April 2001 thus favoured partial reconstruction and partial relocation. It envisaged a reduction of development intensity in the urban areas by restricting both the height of the building and the permissible fl oor space index, implying a horizontal expansion of the city, both during reconstruction and in the years to come. In order to guide and regulate the reconstruction and growth of the city, the government package announced that town planning would be carried out and that the existing development control regulations would be revised. Th e package off ered plots at relocation sites both to homes and businesses that wished to relocate from the Walled City and to those that would be aff ected by the town planning proposals. Th ough the sites were not specifi ed, possible locations were indicated based on the availability of vacant government land. In the case of Bhuj, the package also specifi cally mentioned that government properties inside the Walled City would be made available for the redevelopment of the Walled City. Th ere were special provisions for tenants and unauthorised/illegal settlements. Problems Faced and Lessons Learned Th e provisions for tenants were vague; had they been clearer, the resistance to town planning proposals later on would have been less. Also, unlike the rural reconstruction package, there were no provisions in urban areas for a partnership with NGOs. Th ough this was introduced later on, it was not structured as well as in the case of rural reconstruction. Finally, the policy for valuation of plots off ered at the relocation sites and plots to be surrendered in the Walled City was also somewhat confusing.

Downloaded by [University of Defence] at 01:15 24 May 2016 The Institutional Framework

It was quite clear by this time that the task that lay ahead was of monumental proportions and that none of the local and government agencies had the technical or ﬁ nancial resources to manage the reconstruction. With the assistance of experts from the US Agency for International Development (USAID) Financial Institutions Reform and Expansion- Debt (FIRE-D) project and in consultation with various local institutions and international agencies, the government, through its Urban Reconstruction of Bhuj 163

Development Department led by Dr Manjula Subramaniam (IAS), created the following institutional framework for undertaking reconstruction (Figure 7.1):

(z) In May 2001, the government created Area Development Authorities (ADAs) in Bhuj, Bhachau, Anjar, and Rapar under the provisions of the Gujarat Town Planning and Urban De- velopment Act, 1976. The ADAs were made responsible for implementing town planning proposals and ensuring adherence to improved regulations. (zz) The government negotiated a Rs 500-crore loan from the Asian Development Bank to fund urban reconstruction. (zzz) Since the post-earthquake urban reconstruction project demanded special attention, a dedicated staff and special skills, the government decided to designate the Gujarat Urban De- velopment Company Limited (GUDC) as the implementing agency for the project. The GUDC is a Special Purpose Vehicle established by the government (before the earthquake) for conceptualising and implementing urban development projects. (zv) To support the ADAs and GUDC, the government decided to hire consultants to carry out town and infrastructure planning, and to scrutinise applications for building permissions.

Figure 7.1: Institutional Framework of Reconstruction Process

PLANNED RECONSTRUCTION PROCESS TECHNICAL Area Development TOWN PLANNING ADVISORY CELL Authority CONSULTANT • Earthquake Gujarat Urban • Conduct stakeholder engineering expertise Development consultations • Retrofitting expertise Company • Prepare Development • Construction Plans

Downloaded by [University of Defence] at 01:15 24 May 2016 certification

INFRASTRUCTURE DESIGN & SUPERVISION CONSULTANT • Infrastructure design • Supervision of implementation

Source: Prepared by the author. 164 Ú B.R. Balachandran

Problems Faced and Lessons Learned Th e most important element of the institutional framework, which made the entire urban reconstruction process successful, was the role played by the Gujarat Urban Development Company (GUDC). Th e fact that they had considerable autonomy in functioning and had a committed team of highly qualifi ed and competent professionals made a huge diff erence to the process. To give examples, the author worked very closely with Ashutosh Dhar, an urban planner, and Mr U.P. Dharmadhikari, an engineer who managed all the projects. Both stayed the course of the reconstruction process and brought to the process the sort of effi ciency that is unusual in the public sector. Th e government included two experts on the Board of the Bhuj Area Development Authority (BHADA), but it was not able to utilise their expertise in any signifi cant manner. Also, there was absolutely no local representation on the Board. Perhaps the intention was to avoid unnecessary interference, but in retrospect such fears seem to have misplaced. Local representation would have helped both in improving the plan and in legitimising it. Th e municipality, being an elected body, has a larger stake in the overall development of the city. However, this body had been completely sidelined. While the ADA played a larger role in the reconstruction process, by preparing the development plan, the town planning schemes and issuing building permissions, in the post-reconstruction period, capital assets, and their operation and management, have to be handed over to the municipality. Th erefore, there should have been a systematic process of involving the municipality and building their capacity inorder to over all the work to them eventually. Th e municipal limits should also have been extended and made co-terminus with BHADA limits to enable a smooth transition. And lastly, the agency hired to provide engineering expertise and scrutinise building permissions eventually got removed for various reasons, mainly based on the perception that an external Downloaded by [University of Defence] at 01:15 24 May 2016 agency should not be involved in this task.

Legal Framework and Tools for Planning and Implementation

To guide and regulate the reconstruction and future growth of the city, the government decided to utilise statutory town planning mechanisms like development plans, since ad hoc project formulation would not Reconstruction of Bhuj Ú 165

suﬃ ce for ensuring mitigation of future disasters. Moreover, the government viewed the reconstruction process as an opportunity to develop these cities as models of urban development and to demonstrate the use of modern planning techniques.1 Th e government reviewed the conventional procedures adopted for the preparation of development plans and identiﬁ ed ways in which to shorten the duration of the process. Development plan preparation is a three-stage process:

(i) Preparation and publication of the draft development plan under Section 13 of the Town Planning Act for objections and suggestions from the public. (ii) Review of objections and suggestions, revision of the plan and publication under Section 15 for a second round of objections and suggestions. (iii) Review of objections and suggestions, revision of the plan and submission to the State government for sanction.

To cut short the time required for the above process, in mid-May 2001, the Town Planning Department prepared a draft development plan on the basis of readily available data and maps, and published the same under Section 13 for objections and suggestions. Meanwhile, the government appointed consultants through an open competitive bid to undertake a more detailed and systematic exercise. Problems Faced and Lessons Learned Th e plan published under Section 13 in mid-May 2001 created confusion. Th e low level of detail in the base map and the apparent insensi- tivity of some proposals disturbed the people. For example, the plan proposed that all streets in the Walled City be widened. Th is meant that all buildings on the market streets — most of which had survived the earthquake — would have to be pulled down, thereby ruining the Downloaded by [University of Defence] at 01:15 24 May 2016 local economy. Moreover, apparently, visitors to the BHADA oﬃ ce were told that this plan was considered ﬁ nal at the same time that the town planning consultants were holding public consultations for a new plan.

1 Th is was not an easy decision at that time and Dr Manjula Subramaniam, the then Principal Secretary, Urban Development, Government of Gujarat, had to ﬁ ght for support on this approach. 166 B.R. Balachandran

Figure 7.2: Overall Process Flowchart

Development Plan City Level

Infrastructure Plan Town Planning Schemes City Level Development Control Micro Level Regulations

Construction of City Construction of Infrastructure Plan Level Infrastructure Buildings Micro Level

Construction of Micro Level Infrastructure

Source: Prepared by the author.

In retrospect, the confusion could have been avoided if one of the following strategies had been adopted. The Town Planning Act could have been amended to introduce special provisions for disaster-afflicted areas, changing laws, rules, procedures, and conventions to increase speed and flexibility. Such modifications were later introduced for town planning schemes. Following the Hanshin earthquake (1995) in Japan, the Japanese government enacted a Special Act for Disaster Afflicted Urban Areas, with special provisions for urban planning including the creation of neighbourhood committees for land readjustment projects. The other option was that since the consultants were anyway expected to prepare a conceptual plan by mid-July, this plan could have been officially published under Section 13. Downloaded by [University of Defence] at 01:15 24 May 2016 Preparation and Sanction of the City-Level Development Plan (June 2001-December 2001)

The preparation, publication, revision, and sanction of a city-level development plan normally takes two years at the very least for a city the size of Bhuj. However, in the case of Bhuj, Bhachau, Anjar, and Rapar, despite the complexities involved, the government achieved this stupendous task in just six months (see Table 7.1), and that too with much greater Reconstruction of Bhuj Ú 167

Table 7.1: Timeline for Development Plan Preparation

18 May 2001 Start of Project z Base map preparation z Data collection z Stakeholder consultations 15 July 2001 Completion of Conceptual Development Plan z Stakeholder consultations z Detailed studies z Final base map 15 September 2001 Completion of Draft Development Plan z Formal objections and suggestions from citizens z Plan revision and ﬁ nalisation z Submission to State government 15 December 2001 Sanction of Draft Development Plan Source: Prepared by the author.

detail, accuracy and public participation than ever before, at least in Gujarat if not in India.

Mapping, Studies and Public Consultations

Th e planning area delineated for Bhuj covered 56 sq. km. to accommodate future growth of the city and its suburbs that presently house a population of 125,000. Th e last development plan for Bhuj was made in 1976 — 25 years before the earthquake (Figure 7.3). Th e city had not been resurveyed in the last 30 years. Th e base maps were hopelessly outdated and devoid of detail, and there were no contour maps available, which are needed to plan infrastructure. Th e entire planning area was re-surveyed in great detail with the latest computerised equipment in just two months (Figure 7.4). Th e reconciliation of the survey outputs with the 30-year-old oﬃ cial maps and land records of the City Survey Department and the District Inspector of Land Records — a very tedious and painstaking task — took Downloaded by [University of Defence] at 01:15 24 May 2016 an additional four months. Th is procedure of reconciliation is known in Gujarati as melavni, which literally means ‘matching’. A series of studies was carried out in record time to assess earthquake risk in diﬀ erent parts of the city and surrounding areas as well as to provide inputs to the development plan. Th ese studies included land suitability analyses, demographic studies, survey of the land market, a study of development regulations, assessment of infrastructure status and needs, sizing up the intensity of damage and future seismic vulnerability, and understanding rehabilitation needs (Figure 7.5). Figure 7.3: Bhuj Development Plan, 1976

Source: Environmental Planning Collaborative, Ahmedabad.

Figure 7.4: Base Map Preparation (a) Extract from old base map (b) Fresh cadastral and topographical survey output Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Environmental Planning Collaborative, Ahmedabad. Figure 7.5: Examples of Studies Carried Out for Preparation of Development Plan (a) Existing land use

LECENO H ASAtMM^

| 1 WWWMI, • OOHOMY

I . i.ii*a«ie

(WNTMI

|^B wunuu

UOCHRAJ AAKHA!

MILITARY AREA PMUCVTUTT

•• 1TUITMB

(b) Intensity of damage

A B AIRPORT AREA

RA1TA

11 HE5TOF BHUJ VLL.ViL-i i

12 10 REST OF BHUJ VILLAGE-?

MACtAPAR

MOCHISA. RAKHAL MILITARY

AREA

Downloaded by [University of Defence] at 01:15 24 May 2016 fle$T OF LECEHO SHIM MJA IWHIVT VILLAGER • •• YTLJUM *>UH&»R«

VWCH. BCHJM1W.T

a 'J <• ARE4 :•

•WBMMMDWT

:• «

Source: Environmental Planning Collaborative, Ahmedabad. 170 B.R. Balachandran

In addition, a household-level questionnaire survey was carried out, interviewing 2,500 families and several hundred trade and industrial establishments. A series of public consultation exercises were carried out in two rounds (see Table 7.2). The first round of meetings was with opinion leaders and specific stakeholder groups (Figure 7.6), leading up to the preparation of a conceptual development plan. In the second round of consultations the conceptual development plan was presented at a series of public meetings and group discussions (Figure 7.7), including a city- level workshop with invitees from a cross-section of Bhuj society. In all, over 150 consultation meetings were held and documented. During the

Table 7.2: Public Consultation Exercises

Stakeholder groups consulted No. of meetings Local and State government officials 56 Community groups 10 Representatives of trade and industry 13 Real estate developers 6 Opinion leaders 12 Professional groups 6 Women's groups 1 Artisans' groups 1 Defense officials 3 Source: Bhuj Development Plan, Environmental Planning Collaborative, Ahmedabad.

Figure 7.6: Stakeholders in the Planning Process

Cultural Area Development Municipality institutions Authority

NGOs

Financial Developers

Downloaded by [University of Defence] at 01:15 24 May 2016 institutions Stakeholders

District Collector Business community

Citizens

Source: Prepared by the author. Figure 7.7: Stakeholder Consultations (a) Focus group discussion

(b) Open oﬃ ce for public interaction Downloaded by [University of Defence] at 01:15 24 May 2016

(Figure 7.7 Continued) (Figure 7.7 Continued)

(d) Group discussion in city-level workshop Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Environmental Planning Collaborative, Ahmedabad. Reconstruction of Bhuj Ú 173

consultations it emerged that public opinion on the various issues concerning reconstruction was quite fragmented. Moreover, there was no strong local leadership. Problems Faced and Lessons Learned Th e planning team faced tremendous problems in matching the survey outputs with the offi cial maps and land records. Bhuj is no exception in this matter of defi cient records. Th e case of Bhuj only illustrates the importance of creating accurate base maps for all our cities and regularly updating all records. Th e GUDC commissioned a study by the Geological Survey of India to assess soil conditions in various parts of the city from the point of view of suitability for the foundations of various kinds of buildings. While the scientists had broadly categorised the city in terms of good, fair and poor soil conditions, their study could not give concrete micro-level information that could be incorporated in the regulations, nor did the study conclusively rule out development in any part of the city. ‘Micro-zonation’ was not carried out for Bhuj. It is an exercise that takes a long time and has to be carried out meticulously. Th e results of a micro-zonation study would probably provide inputs for structural design of buildings in specifi c locations rather than land-use zoning in the development plan. Also, in retrospect, perhaps the government could have spent more time on proactive consensus building than on opinion seeking. And fi nally, the town planning consultants had, in mid-July, proposed the establishment of a Community Resource Centre as a joint eff ort of the local administration and the planning team or an NGO, meant to provide a single window facility for providing information and assistance and help in consensus building. Th is should have been taken up.

Vision Statement

Downloaded by [University of Defence] at 01:15 24 May 2016 Th e vision statement for Bhuj, developed through public consultations and studies, attempted to visualise what citizens would like to see when they look at Bhuj a decade after the earthquake. Th ey would like to see that:

z Bhuj is a vibrant centre for trade and commerce in Kachchh, specialising in handicrafts, mining-based industries and building- material manufacturing and trading. z Bhuj is a major tourist destination and the main entry point for tourism in Kachchh. (Box continued) 174 Ú B.R. Balachandran

(Box continued)

z All earthquake-aff ected citizens of Bhuj have been rehabilitated physically, economically and socially. z Bhuj is equipped to withstand and manage disasters with minimum loss of life and property. z Th e urban form of Bhuj, particularly the Walled City refl ects its traditional cultural identity. z Bhuj has an effi cient water management system that conserves precious water resources, reuses water and recharges the aquifers. z Buildings, public spaces and public services and amenities in Bhuj are designed to cater to the needs of vulnerable social groups. Source: Bhuj Development Plan, Environmental Planning Collaborative, Ahmedabad.

While the above vision statement was not taken suﬃ ciently seriously at the administrative level, having involved stakeholders in a consultative process ensured that many of the ideas contained in the development plan would develop a life of their own. For example, the vision of managing water was pursued by Sahjeevan, Hunnar Shaala and Abhiyan — NGOs in Bhuj — through their own programmes.

Development Approaches for the City

Typically, statutory development plans in our country focus only on two aspects, the ﬁ rst being land-use planning and development controls — the regulatory aspect. Th e second aspect is the demarcation of land for public infrastructure such as roads, public open spaces, and other public facilities like bus terminals, etc. Th e development plan for Bhuj had both these aspects and in addition, it also looked at strategic planning for the city’s development. Th e overall approach of the development plan consisted of the development of relocation sites for those who opted to move out of the

Downloaded by [University of Defence] at 01:15 24 May 2016 Walled City; development of a skeletal framework of wide roads to accommodate the horizontal growth of the city; comprehensive redevelopment of the Walled City; and initiatives to stimulate economic development. Th e speciﬁ c sections of the development plan report were as follows:

(i) Relocation and Rehabilitation (ii) Economic Development (iii) Land Development (iv) Road Network and Transportation Reconstruction of Bhuj Ú 175

(v) Physical Infrastructure (vi) Social Infrastructure (vii) Open Spaces, Water Bodies and Environment ( viii) Heritage Conservation and Tourism (ix) Solid Waste Management (x) Informal Sector (xi) Implementation Strategy

Th e approach adopted for relocation and rehabilitation is dealt with later in this chapter. In the section on Economic Development, there were several proposals, short-term and long-term, to help revive the economy. Th ese included initiatives like setting up temporary markets, to establishing a new freight complex, an industrial zone and tourism promotion. Th e section on Road Network and Transportation conceptualised the future urban structure of Bhuj with ring and radial roads, building on the existing pattern (see Figure 7.8). Th e proposed land-use zoning

Figure 7.8: Proposed Road Network for Bhuj Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Bhuj Development Plan, Environmental Planning Collaborative, Ahmedabad. 176 B.R. Balachandran

is integrated with this road network in a synergistic manner, such that, for instance, the commercial corridors follow the arterials. The plan facilitates the redistribution of the intensity of land-use — both residential and commercial — over a much larger area than pre- earthquake. Earlier, a quarter of the population of Bhuj lived in the Walled City — 35,000 people in 1 sq. km. and the remaining 90,000 spread over 20 sq. km. Post-earthquake, the government took a decision that in the entire Kachchh region, building height should be restricted to two floors by default, with only rare exceptions to be made. The intention was to strengthen building controls and enhance the quality of design and construction before relaxing height limits. In any case, the latter, along with the provisions for relocation and rehabilitation ensured that during the reconstruction process, the city would spread outwards. The proposed land-use plan (Figure 7.9) is de-signed to accommodate not only this immediate requirement, but also the demand for land generated by the anticipated growth of the city in the post-disaster economic boom.

Figure 7.9: Proposed Land-Use Plan for Bhuj Downloaded by [University of Defence] at 01:15 24 May 2016

'•[={ VILLAGE BOUHQAUr ^_ IHOU5TRUH IQNI

I I BoyHOJuJJ"™ ^^B INSTmiTlQN*L ZONE rj——| METRE GWJOE KAILVW | | RECREATIONS ZONE

^_ anatDOAUoc RAILWAY ^H RECRCATION*L ZONE •ren 2!m«u«*« "—> "'«'•»»*> a -: a 5?2K« . 1,^3^ WATER BOO* I DEFENCE MIA

Source: Bhuj Development Plan, Environmental Planning Collaborative, Ahmedabad. Reconstruction of Bhuj 177

The planning of physical infrastructure is dealt with separately later in the article. Assessments of the social infrastructure revealed a concentration of health and educational infrastructure in the south and west of Bhuj, while the north and east — where the poorer sections lived — were deprived. Proposals dealt with enabling a balanced distribution of facilities, but have not really been taken up for implementation. Bhuj is blessed with a bowl-shaped topography. Rainwater drains into the Hamirsar lake in the middle of the city through a series of water bodies interconnected by water channels both natural and man-made (Figure 7.10). This water management system had deteriorated over the last five decades — many lakes were filled up and many channels were blocked. The key proposal, formulated through stakeholder consultations, was to revive this system through specific interventions. It was

Figure 7.10: Water Bodies in Bhuj

7 Pr.srjK.ir T;il t\ Catchment

WAULED

^frWee Lakes /€•-.'. M.incjLrii Area j f%c Mochrrai Rikhal j 5 Bn chimi Chlie 13 ^aten merit

1b Haripi Canal 6b Miriapar Area \ y,

Downloaded by [University of Defence] at 01:15 24 May 2016 Catch menti

j*a LakklX fcatchnne]« 2 H.i.Vu.:!... ) 1a TapJa CatEhkont kh NiLndra fRoatfTalay 3 Dhunaraja Catchment

Source: Hamirsar Project Presentation by Sahjeevan, Bhuj and EPC, Ahmedabad. 178 Ú B.R. Balachandran

also proposed to develop the edges of all water bodies and water channels as publicly accessible green spaces. Th ough these proposals were not taken up as part of the reconstruction project, there was a public initiative led by an NGO. Th e project is far from achieving its objectives, it is one of the fi rst eff orts in the country to undertake comprehensive urban watershed management. Th e development plan proposals for heritage conservation dealt with both the need for restoring specifi c heritage assets as well as realising the economic potential of cultural tourism. While between 2003 and 2006, much has been achieved in term of restoration of specifi c heritage structures through the eff orts of Mr Pradeep Sharma, Collector of Kachchh (described later in the article), a concerted eff ort to establish conservation practices is still pending. Th rough a separate initiative, Azhar Tyabji, an art historian at Environmental Planning Collabora- tive (EPC), Ahmedabad, authored a book on the cultural history of Bhuj. Another independent initiative was led by Debashish Nayak of Foundation for Conservation and Research of Urban Traditional

Figure 7.11: Heritage Walk Brochure Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Debashish Nayak, Foundation for Conservation and Research of Urban Traditional Architecture (CRUTA Foundation), Ahmedabad. Reconstruction of Bhuj Ú 179

Architecture (CRUTA Foundation) Foundation to establish a Heritage Society and to launch a heritage walk similar to the one he created in Ahmedabad. Proposals for solid waste management, dealt with conceptually in the development plan, are currently being detailed and implemented. Th e entire section on the Informal Sector, which dealt with both informal settlements (slums) and informal sector economic activity is probably the most neglected part of the Plan. However, recently there have been many initiatives for providing access to water supply and sewerage to slums. Th e biggest hurdle in all such initiatives is the issue of granting tenure. Much better work on this issue has been carried out in Bhachau as compared to Bhuj. Th e Implementation Strategy section provided a broad estimate of the cost of implementation and proposed land management mechanisms to recover at least a part of the cost. However, the entire focus of the reconstruction process was on speedy implementation of public infrastructure and therefore cost recovery took a back seat.

Publication and Sanction of the Development Plan

Based on responses received during the presentation of the conceptual development plan in July–August 2001, modifi cations were made, proposals were detailed and the draft development plan was published in September 2001 (under Section 15 of the Act). Th e plan was open for objections and suggestions from the public for two months, that is, till mid-November 2001. After incorporating changes based on public responses and corrections in the base map, the plan was fi nalised in December 2001 and immediately sanctioned by the State government. Problems Faced and Lessons Learned When the contract for preparing the development plan was awarded, Downloaded by [University of Defence] at 01:15 24 May 2016 an accurate survey of the entire area was included to ensure that time was not wasted in demarcation of road alignments, etc. Th is was being done for the fi rst time in Gujarat (and perhaps in India). At that time nobody anticipated the magnitude of problems that existed in the available base map. Th erefore the need for a meticulous melavni — the Gujarati term used to refer to the reconciliation of survey output with offi cial records — was not envisaged. Despite it not being part of the contract, the consultants carried out the melavni exercise. Th is was not suffi ciently accurate at the Section 15 180 Ú B.R. Balachandran

stage and some confusion resulted. Subsequently, an intensive exercise was carried out with the District Inspector of Land Records (DILR) and City Survey departments and at the time of sanction about 90 per cent accuracy was achieved. However, some problems remained, which came out only during the demarcation of roads (which, incidentally, started very late). Th e importance of accurate base maps for cities cannot be over- emphasised. In normal development plans, no survey is carried out during the planning stage and, therefore, during implementation, considerable changes in alignment are required. In Bhuj, such changes have been minimal. A systematic eff ort has to be mounted to create accurate base maps for all Indian cities without waiting for disasters to strike each one of them. Th is requires close collaboration between the Urban Development Department and the offi ce of the Settlement Commissioner, under whom the City Survey Department and the DILR work. It is also worthwhile to consider amendments to law-making urban local bodies the offi cial custodians of maps and land records in urban areas and making it mandatory for them to maintain accurate and updated maps and land records. At the Section 15 stage, while adopting the general framework commonly used all over Gujarat, the consultants had customised Development Control Regulations, in particular the height and bulk standards, specifi cally for local conditions of Bhuj. A simplifi ed format developed for Gandhinagar was used in Bhuj. Heritage Regulations were incorporated along with a preliminary list of heritage buildings/ structures. Most of the embarrassing language errors that we fi nd in regulations all over Gujarat were corrected in this version. However, at the Section 16 stage, the State government dictated that all four cities in Kachchh should have identical regulations. A town planning offi cial was made responsible for creating the new version, but was given only four days to carry out the task. Th e Bhuj version was abandoned. Th e soft copy of the AUDA regulation was taken Downloaded by [University of Defence] at 01:15 24 May 2016 and ‘Ahmedabad’ was replaced with ‘Bhuj’, ‘Bhachau’, etc. Following the sanction of the development plan, when local professionals protested that many rules like margins, etc. were unviable in Bhuj, BHADA and the government had to hurriedly make modifi cations to the rules again. Also, at the time of sanction, the heritage list was removed. It has not been reintroduced till date. Meanwhile the listed buildings are disappearing very fast. Th ere is a provision in the regulations for creating a Heritage Committee. Th ere is considerable awareness among people about heritage now and the time has come to start working on this issue again. Reconstruction of Bhuj Ú 181

Stafﬁ ng and Infrastructure of BHADA

Soon after the development plan was published under Section 15, BHADA was moved to a larger offi ce. Furniture and equipment were procured. Initially BHADA was staff ed by people drawn temporarily from other parts of the State, but later on dedicated staff was provided. To scrutinise building permissions applications, the government had appointed consultants who set up a Technical Advisory Cell in BHADA. It was probably the fi rst time in Gujarat that an ADA was set up and operationalised at such speed. Problems Faced and Lessons Learned Till October 2001, BHADA had no offi ce to speak of and no technical staff , not to mention computers and printers. Th e CEA2 had other signifi cant responsibilities at the district level and therefore did not get time to devote to BHADA which got a full-time CEA only in April 2002. During the public consultations conducted in the early stages, BHADA hardly participated, in fact the top offi cials consciously shied away from public contact. Th e then Deputy Town Planner had limited experience and hardly participated in any of the planning work. It was only just before the publication under Section 13 that a competent Town Planner, Mr P.V. Prasad was deputed to BHADA and that too ‘on tour’. He was fi nally posted there in September–October 2001 and stayed involved right through the reconstruction process, giving it much needed continuity in decision-making, which was lacking at all other positions as key offi cials changed every few months. In retrospect, it seems that the ‘face of authority’ should have been visible from the beginning in the entire public consultation process.

Planning of Relocation Sites

Th e draft development plan published in September 2001 proposed Downloaded by [University of Defence] at 01:15 24 May 2016 seven possible relocation sites based on the availability of government land. Th is ensured that no cumbersome land acquisition process had to be undertaken. By September 2001 the relocation sites were designated, by October they were transferred to BHADA and demarcated on the ground, and in November 2001 work on the layout was initiated.

2 CEA is the Chief Executive Authority — the administrative heads of all Development Authorities in Gujarat are called CEA. 182 Ú B.R. Balachandran

In December 2001, the preliminary layouts were ready (Figure 7.12). Based on the announced government policy, a procedure was formulated to determine what relocation applicants were entitled to receive their ‘entitlement’. Forms were printed, inviting applications for relocation plots and offi cials were appointed to scrutinise them. Work on the relocation sites and the entitlement process really took off after Mr Bipin Bhatt (a Gujarat Administrative Services offi cer) was appointed as the full-time CEA of BHADA.3 Problems Faced and Lessons Learned In the planning framework for reconstruction, it was extremely critical to designate, design and develop the relocation sites quickly. In mid-July 2001, the consultants had proposed a detailed step-by-step process for the designation, design, development, entitlement, and allocation of plots. However, avoidable delays occurred. Further, although the broad locations were identifi ed in July 2001 and conveyed to the District Collector and BHADA, and all the paper work had been done by August 2001, the actual transfer of lands to BHADA by the Collector was delayed till October 2001. After the Collector transferred the lands to BHADA, there was a delay in getting the Joint Measurement Survey done for demarcation. After demarcation, it was suddenly realised that the design was not there in any consultant’s scope of work. Instead of hiring a consultant quickly, the task was entrusted to the already overworked town planner of BHADA. He prepared the design with informal assistance from various consultants, which took several months. Th e GUDC proposed that the entitlement process be handed over to an NGO. Eventually this too was done in-house by BHADA, but took a long time. Th e policy for entitlement too was rather unclear till recently. These delays in the development of relocation plots and in the entitlement and plot allocation process also had a detrimental eff ect on the town planning schemes. If the process had been faster, more Downloaded by [University of Defence] at 01:15 24 May 2016 people would have opted to move out, thereby reducing the pressure on the Walled City.

3 Soon after taking over as CEA, Mr Bhatt cleared one of the relocation sites of a squatter settlement. Instead of merely evicting the squatters, he almost simultaneously provided them with housing on land with tenure, made them pay a portion of the cost and helped them get support for the remaining cost. Th is swift and positive action had a profound eﬀ ect in boosting the morale of the city which was beginning to lose faith in the administration. Reconstruction of Bhuj 183

Figure 7.12: Relocation Sites

WALLED or, rGIOC,

BHUJIVO 1 1 FORT

MHJTART , AREA 7

B5f MBA [ \ BHADA BOUNDARY R VILLAGE BOUNDARY F3^ CITY SURVEY BOUNDARY I 1 METRE GAUGE RLWY LINE ^M BROAD GAUGE RLWY LIME I =3 | GAIVtTAL BOUNDARY G5] LAKE fcj*^d WATER BODY ] DEFENCE AREA ] EXISTING ROAD | | EXISTING STRUCTURE ^_ SITE IDENTIFIED FOR ^™ RELOCATION

Source: Bhuj Development Plan, Environmental Planning Collaborative, Ahmedabad.

The Walled City - A Complex Problem

There was and still is a clear consensus that the Walled City needed to Downloaded by [University of Defence] at 01:15 24 May 2016 be drastically improved. The main requirement was to enhance safety and enable effective disaster management. For this it was necessary to make the street network more efficient and create more open spaces. The existing street network was full of bottlenecks. At the risk of oversimplifying the issues, it can be said that there were two clear options — to widen the existing main streets or create new streets by using the space created by collapsed buildings (see Figure 7.13). The first option would have meant the demolition of large numbers of standing buildings, particularly the thriving market streets that 184 B.R. Balachandran

had survived the earthquake. This would have resulted in a massive displacement of people and businesses (which are the heart of Bhuj's economy). The second option was clearly better, not only from the point of view of economy but also from considerations of practical implementation. The development plan published in September 2001 proposed that a set of new, wide loop roads be created utilising open patches of land, giving access to the markets and to the entire Walled City while converting the market streets into pedestrian areas. This approach was endorsed at all levels of decision-making from the general public in Bhuj right up to the Chief Minister. However, there was a small group of people who felt that the markets should be demolished and widened, that not doing so amounts to 'being partisan to commercial interests'. There were three options again for realising the proposed plan:

Figure 7.13: Conceptual Layout of Streets Inserted in the Post-Earthquake Fabric Downloaded by [University of Defence] at 01:15 24 May 2016

I—?l EXBT NO HJILWKiS Q| EXtSTWC MAHKF Tf.THEFTS IK PROPOSED I? Uh WIDE ROAD C3 fXBTHO OTYlfYEl PKCWCTS Q PKOfOStD PARK»K MLft BS UtOTOStO » Ml WOt HOW

Source: Town Planning Schemes — Bhuj, Environmental Planning Collaborative, Ahmedabad. Reconstruction of Bhuj Ú 185

(i) Put road lines in the development plan and hope that the roads will be created over many years when the buildings are rebuilt. (ii) Acquire the land and buildings coming in the way of major road alignments through the land acquisition process (leaving the rest of the Walled City as it is). (iii) Reorganise all the open plots using the town planning scheme process, improving their layout and simultaneously creating new streets.

Having evaluated all the options, it was decided the town planning scheme option was preferable as it would result in an overall improvement, and at the same time spread the burden of land/property loss evenly across all aff ected properties. Displacement would be minimised. Since urban renewal in such a complex situation had never been attempted before in India (and perhaps only rarely in the world), it was a diffi cult decision for the government to take. In October there were signifi cant changes in the government and work on the town planning schemes could start only by the end of 2001.4 It is worth noting that the Government of Japan had adopted a very similar approach to the post-earthquake reconstruction of Kobe. Th ey fi rst prepared a city-level plan for the major road networks and other city-level infrastructure. As part of this plan, they delineated ‘land readjustment areas’ and formed neighbourhood committees. These committees were made responsible for taking ownership of micro-level planning decisions and they were provided the services of town-planners to carry out the technical work of physical planning. As in the case of Bhuj, they also ran into serious problems resulting from inaccurate base maps and outdated land records. Th e reconstruction process in Kobe is still on, nearly a decade after the quake. Downloaded by [University of Defence] at 01:15 24 May 2016

4 Th e new head of the Urban Development Department, Ms Sudha Anchalia (IAS) prepared a very detailed analytical note on the pros and cons of preparing town planning schemes for the Walled City and not doing so, and presented the same to the new chief minister and other top oﬃ cials and ministers before proceeding. It was quite clear to all the key decision-makers at that time, that what was about to be attempted was pathbreaking, but fraught with procedural complexities. 186 Ú B.R. Balachandran

Problems Faced and Lessons Learned Th ere were some avoidable delays in decision-making related to the town planning schemes. Th ough the proposal was rstfi fl oated in mid-July 2001, it took till September 2001 for the government to decide to opt in their favour. All of September and part of October were spent debating which agency should hire the consultant required for the town planning schemes — GUDC or BHADA? When the chief minister changed in October, the government decided to review whether indeed town planning schemes were required and viable in the Walled City. Th e fi nal decision to prepare the schemes was taken at the chief minister’s level at the end of November 2002. Even after the decision was taken, it took all of two-and-a-half months to commission the work — the contract was fi nally awarded in mid- February 2002. An interesting fact is that in August 2001, the big debate was how long the town planning scheme’s preparation would take. Both consultants and knowledgeable people in the town planning department said nine months, while the decision makers insisted that it should be done in two months. Eventually, the decision makers wasted eight months to arrive at a decision and commission the work, and the work itself, despite a mammoth eff ort, took eight months to complete. If the work had been started in August 2001, it would have been completed by April 2002 and implementation could have started immediately after.

Preparation and Sanction of Town Planning Schemes for the Walled City (January 2002–February 2003)

Th e planning process carried out for the redevelopment of the Walled City of Bhuj is perhaps the most complex physical planning exercise ever

Downloaded by [University of Defence] at 01:15 24 May 2016 attempted in India. It consumed over 200,000 hours of work by highly qualiﬁ ed professionals using the best available technology, co-ordinating the work of nearly a dozen organisations and dealing with over 12,000 plots for which over 30,000 persons have ownership claims. What has been achieved is almost unbelievable given the conditions in which the work was carried out. Town planning schemes are essentially land readjustment schemes where the existing plots of land, known as Original Plots are reconstituted into Final Plots, taking away a small portion of land from each Reconstruction of Bhuj Ú 187

plot to create new streets and open spaces. In the Walled City of Bhuj, the objectives of the town planning scheme were:

(i) To enable eff ective disaster management, and reduce loss of life and property in future disasters by creating a more effi cient street network. (ii) To improve the shape of plots and the pattern in which they are laid out, making for more effi cient buildings. (iii) Creating more open spaces and space for other utilities (Figure 7.14).

Th e town planning scheme process consists of three distinct stages. The first, known as the draft town planning scheme is under the command of the local body — in this case the Bhuj Area Development Authority. At this stage a draft plan is prepared and published for public viewing. After incorporating the objections and suggestions of the owners, the plan is submitted to the State government for sanction. Once the draft plan is sanctioned, the local body can take possession of the roads and other land allocated for public use. Meanwhile the State government appoints a town planning offi cer (TPO). Th e TPO, a quasi- judicial authority, arbitrates between the owners and the government. Th ree rounds of individual hearings are given to the owners, after which the physical plan — the layout — is fi nalised. Th e last stage involves the implementation of the physical plan on the ground and the resolution of all outstanding issues including fi nancial matters, known as the fi nal town planning scheme. Table 7.3 presents the timeline for the town planning scheme process as it took place in Bhuj.

Draft Town Planning Schemes

The EPC team started working on the town planning schemes for

Downloaded by [University of Defence] at 01:15 24 May 2016 the Walled City in December 2001 in anticipation of the contract being signed (which happened two-and-a-half months later). Th eir work was constrained by two problems. First, in September 2001 itself, the consultants had pointed out that the base map and property records of the Walled City were in very bad shape. However, when they started the actual work, the planning team ran into many more problems than they had anticipated. Second, the total station survey of the Walled City had been carried out in May 2001, when the debris clearance had just started. After debris removal, the situation changed 188 Ú B.R. Balachandran

Figure 7.14: Plot Reorganisation (a) Neighbourhood before plot reorganisation

(b) Neighbourhood after plot reorganisation Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Town Planning Schemes — Bhuj, Environmental Planning Collaborative, Ahmedabad. Reconstruction of Bhuj Ú 189

Table 7.3: Timeline for Preparation of Town Planning Scheme

13 February 2002 Start of project z Base map preparation z Data collection z Stakeholder consultations 24 April 2002 Completion of Draft Town Planning Scheme z Reconstitution of plots z Preparation of ‘F Form’, report z Owners’ meetings z Publication of Draft Scheme, inviting objections 12 August 2002 Sanction of Draft Town Planning Scheme z Formal objections and suggestions from citizens z Plan revision and ﬁ nalisation z Submission to State government, sanction February 2003 Completion of Preliminary Scheme z Individual hearings with owners z Finalisation of plan and ﬁ nancial details Source: Compiled by the author.

considerably. Th e planning team had to carry out a partial resurvey and work on the base map till early March. Given the near impossible deadline, it was decided to freeze the base map and work on the reconstitution of open plots so that at least the draft town planning scheme could be published. Remaining problems were to be solved before submission of the draft schemes to the State government and during the preparation of the preliminary schemes. Th e following policy was adopted for deduction of land from open plots:

Plots less than 30 sq. m. No deduction 30 sq m to 100 sq m 10% 100 sq m to 200 sq m 20% 200 sq m to 500 sq m 30% More than 500 sq m 35% Downloaded by [University of Defence] at 01:15 24 May 2016

It was decided that standing buildings would be spared from deduction unless they were aﬀ ected by the proposed road alignments. In April 2002, eight draft town planning schemes were presented to the public at a series of owners’ meetings and then published to receive more objections and suggestions. Following publication, the consultants and BHADA, in collaboration with the Bhuj Development Council, an NGO led by Nalin Upadhyay, a respected columnist with a 190 Ú B.R. Balachandran

construction background, conducted over 20 meetings to present and discuss the schemes with the people. Th ese were well attended and conducted by BHADA with support from the planning team. Th e Bhuj Development Council established ward offi ces in each scheme area, displayed maps there and assisted people in fi ling written responses. Th ey also facilitated the creation of neighbourhood committees in some areas, but Mr Upadhyay’s eff orts to create a city-level citizens’ committee met with little success. Th e people who checked the published maps pointed out both mistakes in the maps as well as what they felt were errors in judgement. Th e land deduction policy was received well by the majority of the people. Th e responses received from people were reviewed case by case, and the plans were modifi ed to correct genuine mistakes and to improve the layout. Th e government appointed TPOs one month in advance of their actual duty to check the schemes before submission of the schemes to the government for sanction so that mistakes could be minimised. Th e draft schemes were submitted to the government at the end of July and were sanctioned early in August (see Figure 7.15).

Problems Faced and Lessons Learned

Th e debris removal contracts awarded in May 2001 were time-bound contracts. Th e work remained incomplete at the end of the time limit in July 2001. When the town planning scheme work started, there were still a large number of G55 buildings and thousands of tonnes of debris lying around. Th e planning team informed the government about this in February 2001 and brought this up regularly in meetings. However, the demolition of G5 buildings and the removal of debris remained incomplete till late in the process. It was important to mark all buildings offi cially designated as G5 on the base map. However, the offi cial records identifi ed G5 buildings

Downloaded by [University of Defence] at 01:15 24 May 2016 by municipal house number, while the base map had only city survey numbers. Th e only register that matched city survey numbers with municipal house numbers (the Inquiry Register of the City Survey Department) had old house numbers. Th e municipality had re-numbered the houses, but could not produce any register that had both old and new

5 After the disaster, damaged buildings were classifi ed into fi ve diff erent categories, G1 to G5, in increasing order of the extent of damage sustained. G5 buildings were those which needed to be demolished. Reconstruction of Bhuj 191

Figure 7.15: Town Planning Scheme No. 2 (a) Town Planning Scheme No. 2 before planning

(b) Town Planning Scheme No. 2 after planning

Source: Town Planning Schemes — Bhuj, Environmental Planning Collaborative, Downloaded by [University of Defence] at 01:15 24 May 2016 Ahmedabad.

house numbers. Therefore, the planning team had to carry out their own survey of building conditions to be able to take planning decisions. A team of government surveyors was to be deputed to assist in base map preparation and verification. However, this did not happen. If the Municipality and City Survey staff had been deputed to work with the planning team on resurveying the entire area and creating an accurate base map and property record database of the entire Walled City, 192 Ú B.R. Balachandran

the actual planning could have been carried out with much greater ease and fewer mistakes. Th e planning team was to work under the direct and close supervision of the BHADA Town Planner. However, during the period December 2001 to March 2002, he was forced to spend over 70 per cent of his time on the design of relocation sites and in the scrutiny of applications for relocation plots, in addition to his regular duties related to building permissions. Had he been allowed to focus on the town planning schemes, many problems could have been sorted out earlier. When the draft town planning schemes were being fi nalised for publication, the then decision makers insisted — contrary to the policy decided earlier — that the main market streets should be widened to at least 6 metres. Th is was a diffi cult decision to implement and a deviation from declared policy. Th e policy announced in the government’s urban reconstruction package that government properties in the Walled City would be utilised for the redevelopment of the Walled City, was not followed. Only small portions of large government properties such as the police headquarters and the jail have been deducted, thus requiring more deduction from private properties and resulting in less open spaces. Th e government properties should have been transferred to BHADA immediately after the announcement of the government package for urban areas. Initially, the government had selected a set of experienced, senior- level offi cers as TPOs. However, very soon after, they were changed and the present TPOs were appointed, who are junior-level offi cials with insuffi cient experience. Th ey were most reluctant to take up the work, as the obstacles seemed insurmountable, working conditions were tough, the highly sophisticated technology being used was alien to them and the departures from convention were diffi cult to stomach. Th ey badly needed the guidance, reassurance and support of senior, experienced offi cials, but this was not available. Th e delay in developing relocation sites continued to be a problem for Downloaded by [University of Defence] at 01:15 24 May 2016 the town planning schemes. Th e list of plots going in for relocation could not be fi nalised even at the time of submission of Draft Schemes to the government. Moreover, the development of the sites was continuously getting delayed. As late as May–June 2002, changes were made in the shape of plots, service lanes introduced and then removed; all these changes resulting in avoidable delays. At the time of awarding the contract for town planning scheme preparation, it was decided that the demarcation of the schemes would be handled in-house by BHADA directly, with staff from DILR/City Reconstruction of Bhuj Ú 193

Survey. Towards the end of the scheme’s preparation it was decided that the demarcation is best outsourced. However, the procurement process was delayed considerably. Th e scheme of deductions with abrupt transitions at particular plot sizes meant that plots which were near the bottom of a block might suﬀ er a larger deduction than slightly smaller plots which fell just short of that block. To overcome this, the minimum after deduction was made equal to the maximum of the next lower block. A scheme with no discontinuities and smoother transitions — similar to income tax brackets — would have been better. A major demand put forward by Nalin Upadhyay and his colleagues at the Bhuj Development Council was that the traditional neighbourhood culture (falia sanskriti6) should as far as possible be preserved in the reconstruction. To facilitate this, Mr Upadhyay suggested the formation and oﬃ cial legitimisation of neighbourhood committees (falia samitis). Th ough he fought relentlessly for it, he received very little support.

Preliminary Town Planning Schemes

Th e major tasks at this stage were: fi nalisation of the town planning schemes by the TPOs, and demarcation of major roads and later of Final Plots as fi nalised by the TPOs. Th anks to reforms recently introduced in the Gujarat Town Planning and Urban Development Act, it is now possible to take possession of road alignments immediately after sanction of Draft Schemes by the Government. Utilising this provision, BHADA started taking possession of the proposed major roads in the Walled City, issuing notices to the concerned people. Th e State government assigned two senior offi cials, Mr M.M. Bhaumick and Mr K.M. Panchal, to assist BHADA in the process. Th e major roads were demarcated by mid-October 2002. At the instance of the above offi cials, it was decided to implement a stretch of highly visible street to send a clear message

Downloaded by [University of Defence] at 01:15 24 May 2016 that the town planning process is fi rmly underway. Several buildings aff ected by the proposed road alignments needed to be demolished. Th is activity was initiated. Meanwhile, the TPOs for the eight schemes established offi ce in Bhuj. Th e TPOs act in a quasi-judicial capacity. Th eir work involves the following stages:

6 Falia is the Gujarati/Kachchhi word for the traditional neighbourhood in old settlements. 194 Ú B.R. Balachandran

(i) Issue notices to individual owners, hear them individually and make appropriate modifi cations to the plan. (ii) Issue a second notice to the aff ected parties, hear them again and make further modifi cations to the plan. (iii) Issue a third and fi nal notice to the aff ected parties, hear them again and fi nalise the plan. (iv) Declare what is known as the ‘Preliminary Scheme’, wherein the physical plan is fi nalised; issue notices to all owners informing them of the fi nal decision regarding their individual plots. (v) Submit the Preliminary Scheme to the government and get it sanctioned. (vi) Prepare and submit the ‘Final Scheme’, which deals with all the fi nancial issues such as compensation, incremental contribution and cost of works.

Th e TPOs issued notice for the fi rst hearing to individual owners in early September 2002 and then began the hearings. Despite collaborating with the municipality and local groups, they faced a lot of diffi culty in contacting all the owners as proper addresses were not available and people had moved out of the Walled City. Th e second round of hearings was completed by the end of November 2002. Based on the requests made by owners, the TPOs started revising the fi nal plot layout in December 2002. As the layouts got fi nalised, simultaneously, the fi nal plots were demarcated on the ground. By the last week of January, the layouts were fi nalised and presented to the BHADA Board for its approval (see Figure 7.16). Th e demarcation and the fi nal round of hearing were completed by mid-February 2003. Th e process of handing over fi nal plots started in the last week of February and continued till the end of March 2003. Th e completion of one of the town planning schemes was considerably delayed due to litigation.

Downloaded by [University of Defence] at 01:15 24 May 2016 Problems Faced and Lessons Learned Th e TPOs were not given proper offi ce facilities, computers, support staff , etc., till the end of September, a month after they joined. When demarcation was started, the planning team felt that building outlines need to be resurveyed for precise co-ordinates and to update the base map for changes in the ground situation. Some G5 buildings had been removed, some new unauthorised constructions had come up, and the errors and omissions in the Draft Scheme survey had to be rectifi ed to Reconstruction of Bhuj Ú 195

Figure 7.16: Walled City — Final Layout

Source: Town Planning Schemes — Bhuj, Environmental Planning Collaborative, Ahmedabad.

ensure that the demarcation was precise. Simultaneously, a detailed survey and accurate plot to plot melavni of internal divisions in clusters of standing buildings was also started. As this process progressed, simultaneously the TPOs were conducting their hearings. An assembly line was set up for the resurvey, to carry out melavni and to provide revised base map drawings to the TPOs. Th e latter were expected to revise the reconstitution of plot layouts based on their hearings. However, they were unable to proceed as they did not have Downloaded by [University of Defence] at 01:15 24 May 2016 AutoCAD operators to carry out reconstitution (except Town Planning Scheme No. 8). In mid-December, on the government’s request, the Environmental Planning Collaborative provided additional AutoCAD operators to the TPOs. Tremendous pressure was exerted by the government on the entire system to complete the Preliminary Scheme at the earliest. As a result, the entire ‘hearing’ process became somewhat a formality and got wrapped up in the shortest possible time. When the ﬁ nal plots were 196 Ú B.R. Balachandran

demarcated on the ground, all hell broke loose. During the revised reconstitution, grievous errors had occurred in the fi nal plot allocation. New plots blocked the entrances of existing buildings in several locations. In many cases, shared portions of house clusters had been swal- lowed up in the reconstitution, leaving the projections of standing houses sticking into the new fi nal plots. Th ere emerged a whole set of problems arising out of the abysmal state of property records and the apathy of owners in updating the records. In a large numbers of cases there were issues relating to joint ownership. In other cases, the original properties had been amalgamated or sub- divided, but the records did not refl ect these transformations. As a result, the fi nal plot allocations had been made without considering these issues. As a result of all these problems the owners of hundreds of properties suff ered hardships and delays in receiving building permissions for their houses. Th e Town Planner of BHADA, Mr P.V. Prasad, spent many months in a highly engaged eff ort with the property owners to

Figure 7.17: Walled City — Destruction Caused by the Earthquake (a) Soniwad, Walled City in February 2001 Downloaded by [University of Defence] at 01:15 24 May 2016

(Figure 7.17 Continued) (Figure 7.17 Continued) (b) Soniwad, Walled City in January 2004

Source: Author. 198 Ú B.R. Balachandran

resolve the last lot of the problems that emerged from the implementation of the town planning schemes. It is with immeasurable gratitude and respect that I put on record the incredible commitment that Mr Prasad brought to his work in planning the reconstruction of Bhuj. He is still engaged in resolving the last lot of the problems that emerged from the implementation of the town planning schemes.

Infrastructure Planning

The development plan at the city level included conceptual plans for infrastructure. The sanction of the development plan was followed immediately by the preparation of detailed plans for city-level infrastructure by infrastructure design and supervision consultants. Th e scope of their work included engineering design, packaging and construction supervision of projects for roads, water supply, sewerage, storm water design and public buildings. Over a period of two years, the GUDC, with local facilitation from BHADA, implemented the key

Figure 7.18: Infrastructure Implementation in the Walled City (a) Infrastructure Implementation in Saraf Bazar Downloaded by [University of Defence] at 01:15 24 May 2016

(Figure 7.18 Continued) Reconstruction of Bhuj Ú 199

(Figure 7.18 Continued)

(b) Saraf Bazar after Infrastructure Implementation

Source: Author.

improvements proposed in the development plan, such as ring and radial roads, trunk mains for water supply and sewerage, etc. One of the interesting aspects of urban infrastructure design in Bhuj and in the three other towns in Kachchh is that the networks are designed with features like ﬂ exible joints so as to withstand the eﬀ ect of ground movements. Th e infrastructure networks in the Walled City took much longer to design and implement. Th e design process started in mid-2002 when

Downloaded by [University of Defence] at 01:15 24 May 2016 the draft plans were sanctioned and the road network was almost ﬁ nal. Th e implementation started in early 2003 and was completed in late 2004. Th e design of the system in the Walled City was highly complex. A completely new network was being inserted into a densely built, inner-city area. Some parts of the Walled City were functional, with homes lived in and businesses working, being served by the old, damaged system. Implementing the new system while people were still living there and shops were open was quite a nightmare for the administration. Th e implementation of the infrastructure could not have been achieved 200 Ú B.R. Balachandran

without the untiring eﬀ orts of a very capable team of professionals in the GUDC, including people like Mr U.P. Dharmadhikari, who managed the engineering projects and Mr Ashutosh Dhar, an urban planner.

Bhuj: Rising from the Rubble

Th e reconstruction process picked up pace and started showing results after the reins of administration were taken over by a dynamic Collector named Pradeep Sharma in mid-2003. In the period that followed, Bhuj saw not only the implementation of the proposals of the development plan and the town planning schemes at a faster pace, but they were supplemented by a large number of other initiatives that helped revive the sociocultural life and economy of Bhuj. Th e prominent streets in the city were quickly developed and provided with streetlights. Many public gardens were renovated and a large new public garden was built on a hill at Rawalwadi, near one of the largest relocation sites. Known as the ‘Hill Garden’, this is now one of the most visited public spaces in Bhuj. Several heritage structures were restored

Figure 7.19: Bhuj — Rising from the Rubble (a) A Walled City Street coming back to life Downloaded by [University of Defence] at 01:15 24 May 2016

(Figure 7.19 Continued) (Figure 7.19 Continued) (b) Sarpat Gate after Restoration

Source: Author. 202 Ú B.R. Balachandran

including three Walled City gates, the Alfred High School, the vegetable market at Sharaf Bazar, etc. Today, the city of Bhuj presents an optimistic picture. It has, almost overnight, transformed from a dusty little district headquarters into a modern urban centre. With its brand new infrastructure, Bhuj can now support much faster growth in population and economic activities. A hundred times more investment has gone into physical and social infrastructure in Bhuj following the earthquake than was invested in the 10 years preceding the earthquake. Most importantly, the citizens of Bhuj are now acquainted with the process of regulated building construction, which we hope will ensure safer construction and mitigate the impacts of a future earthquake.

Conclusions

In conclusion some of the key lessons learned are summarised here:

z Continuity in decision-making is extremely critical in any process of this kind. Between January 2001 and January 2004, Kachchh saw fi ve Collectors; BHADA saw four CEAs; GUDC saw four Man- aging Directors and even the Urban Development Department of the Gujarat government saw fi ve Principal Secretaries. Th ere were two factors of continuity in the entire programme — the technical team of planners, project managers and engineers in GUDC and the Town Planner in BHADA, Mr Prasad. Many uncertainties and delays that occurred during the process were in good measure due to these frequent changes. z A trade-off between the time required, the quality of work and the level of public participation is inevitable in a post-disaster reconstruction process. The true challenge before political leaders and bureaucrats is the negotiation of this trade-off in a

Downloaded by [University of Defence] at 01:15 24 May 2016 transparent manner, taking the people along. z Structuring the reconstruction process and putting in place the appropriate institutional arrangements is extremely crucial. At this stage, the involvement of experts who have reconstruction experience is most beneﬁ cial. z As far as possible, the reconstruction process should be undertaken using existing legal and institutional mechanisms. Th e disaster should be made an opportunity to strengthen and streamline these systems. Reconstruction of Bhuj Ú 203

z Th e importance of maintaining accurate maps in all our cities cannot be over-emphasised. Th ere is a Bhuj waiting to happen in every Indian city — if not an earthquake, it will be a fi re or a fl ood. Th e institutional architecture for mapping and maintaining land records in cities needs to be drastically altered and a large campaign must be launched to catch up on lost time. z In the post-disaster situation, it is important to show results fast and continuously. Many of the proposals in the development plan could have been taken up for implementation while the planning for the Walled City was going on, particularly those that had economic benefi ts for the city.

To reiterate, despite all the problems that have been listed here, the reconstruction of Bhuj can truly be considered one of the landmark post- disaster projects in the world. In no small measure is the credit for this due to the absolute commitment shown by the government machinery in making Bhuj a model for other cities to follow. Downloaded by [University of Defence] at 01:15 24 May 2016 8 Participative Rehabilitation A Dilemma M.N. Joglekar

There can be no doubt that the victims of a disaster should be engaged as active participants in the process of reconstruction. But such eng- agement is fraught with complexities, and by no means guarantees a successful outcome. Th is chapter is an attempt to understand these complexities. It concludes with a case study of a successfully reconstructed slum following the Chamoli earthquake in Uttaranchal in 1999.

Participative Rehabilitation

Th e post-disaster rehabilitation process is beyond the means of the affected individuals and therefore usually involves several actors. ‘Participative rehabilitation’ is in fact a cherished ideal. However, its success depends on the complexity and scale of the disaster whose consequences are to be mitigated. Th e larger the scale, the greater the eff ort needed for mitigation — putting it that much further beyond the capacity and control of the disaster-aff ected people. Not only is there an increase in the number of participants involved in the mitigation eff ort but each group also brings with it its own idiosyncrasies, egos and diff ering objectives. Th e participants may be institutions and not individuals, but they are fi nally represented by real people with all their human weaknesses. Moreover, other than the aff ected people no

Downloaded by [University of Defence] at 01:15 24 May 2016 one else has a major or real stake in the outcome. With limited resources at their disposal, they become the ultimate suﬀ erers. In order to be successful, therefore, the rehabilitation process must evolve a common ground of attitudes between the various actors involved. It is here that the principal dilemma of the process surfaces: who should participate, to what extent? Should people’s perceptions override the perspective visions of qualitative development — both rural and urban? Participative Rehabilitation 205

Being a complex process, rehabilitation depends on co-ordination between the various participants/actors, each having different objectives and attitudes (see Figure 8.1 and Table 8.1). Put in a lighter vein, the post-disaster rehabilitation scenario consists of hapless victims (beneficiaries), an overburdened and unwilling bureaucracy, over- enthusiastic technocrats with less than adequate sensitivity (particularly those from the establishment) to community preferences, opportun- istic non-governmental organisations (NGOs) with a self-imposed moral obligation to help the victims, donors looking for tax exemption and publicity, politicians looking for vote bank expansion, domestic funding agencies generally ignored in preference to lucrative dollars and foreign visits, international funding agencies ready to step in with their consultants and a desire to discipline the erring developing world through a reform agenda, construction agencies and producers/importers of foreign, disaster-preventive systems already obsolete in those countries, larger construction companies out to promote their business at the cost of people-oriented and decentralized delivery systems, and last but not least the media — a two-edged sword that helps as well as complicates the process. With so many actors involved in the system, a massive and complex set of relationships emerges. The success of the project then essentially

Figure 8.1: Actors in a Rehabilitation Gam<

Construction Media Bldg. material Industry Politician

PEOPLE BENEFICIARIES Donors REST Bureaucrats Downloaded by [University of Defence] at 01:15 24 May 2016

— Funding Technocrats Agencies NGO

Fence Sitters

Source: Prepared by author. 206 Ú M.N. Joglekar

Table 8.1: Rehabilitation Game — Macro Models

S. No. Scenario Beggers Choosers Consequences I Few grants, People Rest People may suff er through more loans arrogance of rest, imposed solutions not people sensitive II Abundance of Rest People Benefi ciaries dictate Authorities/ grants Donor Agencies surrender, Technically ineffi cient compro- mised projects, Quality suff ers III Fair balance None None Replicable projects of grant funding

between grant Lack of abundance and loan – Both linked Source: Compiled by author.

depends upon the managerial ability of the project leader to streamline the process and orient it towards the desired and declared objectives and goals. What matters most in the process is the attitude of the participants, the quality and quantity of participation, its spread amongst the beneﬁ ciaries and the scale of the project. Th e attitude of the participants is governed by their ability to perceive and have a perspective vision. And the spread and reach of the consultations depends upon the ability of the project manager to handle and control the target group who have varying perceptions of their own future — ranging from none to impossible dreams. In most cases, and particularly in the case of rural populations, often the main suﬀ erers of a natural disaster, perceptions are governed by the surrounding environment. Since the surrounding environment has very little that inspires, in a rural set-up the imitation of urban systems becomes a desirable objective, regardless of whether it is good or bad. Th e time element (time available for the completion of a project, whether it is social/economic/physical rehabilitation) also determines the participative character of the rehabilitation projects, in addition to the mode of process initiation at the macro level. Downloaded by [University of Defence] at 01:15 24 May 2016

Micro Models as Derivatives of the Macro Environment

In any calamity a comprehensive process is initiated ﬁ rst at the level of the State government, principally because of its spread and the huge funding needs that are generally beyond the capabilities of even the largest development organisation. So far two main rehabilitation Participative Rehabilitation Ú 207

models and a third which combines the ﬁ rst two have emerged. Par- ticipative networking has invariably been a product of the initiating model, which in turn is a product of numerous variables, prominent amongst these being the arrangement of the huge funding requirements and the role that governmental institutions want to play. Broadly these models are:

(A) State-controlled and specially organised processes involving international funding with direct state responsibility for physical rehabilitation, as adopted in the rehabilitation of the victims of the 1993 Killari earthquake in Latur district of Maharashtra. (B) Routinely organised processes without special funding arrangements and no direct state responsibility for physical rehabilitation as in case of the Uttarkashi earthquake (1991) and Chamoli earthquake (1999), both in the State of Uttaranchal. (C) Th e third model is a combination, in varying degrees, of Models A and B. Th e rehabilitation process for the 2001 Kachchh earthquake in Gujarat is heading in that direction.

The participation systems seen in individual projects emanate from the macro models adopted, as mentioned above. For example, in Model A, politicians, bureaucrats and technocrats may dominate, with people’s participation being defi ned according to the political compulsions and conditions of international funding. An ideal variation of this model would be implementation responsibility with benefi ciary- controlled implementation agencies, with statutory expert technical advice, in order to achieve a fair balance between people’s perceptions and visionary development goals. Model B would be ideal from the people’s point of view, would be convenient for the administrative machinery and would also provide an ideal platform for politicians. Donor NGOs, however, would be unhappy due to a lack of their involvement. Th ere is also no guarantee that shelters Downloaded by [University of Defence] at 01:15 24 May 2016 would actually come up, since the cash distribution of grants could easily get diverted to other priorities that the benefi ciaries may have, leaving the residents in unsafe houses still. An ideal variation of this theme would be to grant assistance distribution in kind, i.e., building materials/ components, through a responsible agency, maybe an NGO, who may also provide technical and organisational assistance to the benefi ciary or to his implementation agency. Suggested variations to both themes would constitute the third model. As can be seen, each model has variable roles for the same actors/ 208 M.N. Joglekar

participants in the rehabilitation process. Similarly, participative processes at the project level would also differ in emphasis and character. At the project level, participants in the rehabilitation process can be divided into two categories: beneficiaries on one side and the rest on the other. In a successful participative process the participants' attitudes towards each other and their objectives would be critical. Attitudes could be of three types. First, each could say that I know everything, they know nothing. The second attitude could be that they know everything but I know very little. The third attitude could be that both of us know something and we must pool our resources for mutual benefit. The spread of participation (its magnitude) and the time required to enable the participation process depends on the attitudes that the two main participants adopt. Categorising the participants into two types as above is a simplified version of the proposed participative model. If participants in the 'Rest' category are further detailed, increasing their numbers, a complex situation emerges — one that makes or breaks the participative process. Figure 8.2 will show the implications of the attitudinal variations on two other parameters, i.e., time and spread.

Figure 8.2: Extent of Consultations and the Number of Beneficiaries Acceptable Zone Acceptable Zone

CD -3 1 £? "53 '53 a a u u G t Downloaded by [University of Defence] at 01:15 24 May 2016 t 0% 100% CD 0% 100^ CD H I II III IV I II III IV Extent of Consultations No. of Beneficiaries

Source: Author. Note: I to II: Attitude that I know everything they know nothing. No consultations. II to III: Attitude that both know something — also zone of representative consultations. III to IV: Attitude that I know nothing they know everything. Consultations with all. Participative Rehabilitation Ú 209

Attitudes and the Time Required to Spread Participation

Th e attitude that ‘I know everything and others know nothing’ reduces the decision-making process to no consultations with others. It also means that the time required for decision-making is nil. In the Killari earthquake (1993) rehabilitation process, donor agencies were quick to respond and wanted to commence construction even before the State government could come up with a comprehensive World Bank assistance programme. Being conscious of the state-controlled nature of the rehabilitation programme, donor agencies initially opted for state- approved building plans and village layouts. Th e government too was quick to decide that ‘they know everything’ and forwarded the typical State Housing Board designs and State Town Planners’ Layouts to the donor agencies. It was a diff erent matter that the decision-making process was fast because of the lack of benefi ciary consultations, but environmental quality suff ered as the house designs as well as layout quality were highly urbanised and as such not acceptable to the people. Th e other extreme attitude could be that ‘I know nothing and they (benefi ciaries) know everything’. A donor NGO may decide that the perceptions of people are supreme, irrespective of their quality. In both the Killari and Kachchh (2001) earthquake shelter rehabilitation projects, it was not uncommon to see the benefi ciaries’ insistence and its acceptance by the donor agencies/NGOs that house construction, even for single or double-storeyed development, must be based on column– beam frame construction, despite the fact that the more economical option of load-bearing wall construction, with reinforcement at the corners and around openings was ideal. As people’s perceptions are qualitatively not only not up-to-date, but are also not futuristic, environmental quality suff ered. Th is happened at the settlement planning level as well as the level of planning house expansions in case of a core house,

Downloaded by [University of Defence] at 01:15 24 May 2016 where the need to structurally integrate the core and the expanded construction was not foreseen. An example of this is quoted later. However, in this participative attitude, unlike the fi rst, the time taken and the spread of consultations has to be more. Unless more benefi ciaries are consulted, their wishes cannot be ascertained. However, the time taken for consultations can also be reduced by following a diff erent strategy. Th is brings us to the third attitude. ‘I know something and they also know something and the decision- making process must follow discussions on each others’ views’ is the 210 Ú M.N. Joglekar

third attitude. With this attitude the consultation process becomes much easier. Th e spread of consultations has to be selective; representatives of people should be involved in consultations rather than consulting none or all. Th is reduces the time taken for consultations and is therefore the most acceptable mode. It is important, however, to ensure that there are no predetermined notions of supremacy of either of the two parties.

The Why, Where and How of Participation

Is a participative process necessary? Most disaster rehabilitations generally will have two parties: the aff ected people and the rest who help them. Usually people expect grant assistance. But fi nancial institutions like the Housing and Urban Development Corporation (HUDCO) have not succeeded in providing loan assistance on a large scale either in the 1993 Killari or the 2001 Kachchh earthquake. Even in the Killari earthquake rehabilitation project the loan component (for benefi ciaries) envisaged by the World Bank was a State liability. Normally, the givers of the grant are the ‘choosers’, while the benefi ciary recipients are often equivalent to ‘beggars’. Th eoretically, choosers can dictate the terms, but the rehabilitation scenario is the reverse of the normal process. Humanitarian issues as well as political needs makes it impossible for donors to dictate terms even when the supply of grant assistance falls short of demand. The Gujarat earthquake saw the strange phenomenon of grant supply being in excess of the demand for it, particularly in the NGO sector. Participative rehabilitation in its true sense becomes impossible in such a situation. Th e benefi ciaries are in a position to choose and dictate their terms. Some NGOs who anticipated a large rehabili-

Downloaded by [University of Defence] at 01:15 24 May 2016 tation programme and collected funds accordingly, had to reduce their involvement to around 50 to 60 per cent of their original targets. Th is also forced them to adopt a policy of regarding the people’s wishes as supreme and inviolable. Th is can make the participative process entirely one-sided, and the likely casualties will be optimum technical solutions and environmental quality in settlement planning. Figures 8.3 and 8.4 demonstrate the weaknesses of people’s perceptions. Figure 8.3 shows the layout plan proposed by the architectural consultant as per the requirement of Trombey village, located near Rapar Participative Rehabilitation 211

Figure 8.3: Proposed Layout — Trombey Village Rapar

NORTH 20_ 0_20 40 60 8C

I;': id .Ho M- ?'•;: $SS? Community Center jWHf Panchayat Office j XS

Source: Author. Note: A FICCI/CARE project for the rehabilitation of the Kachchh (Gujarat) earthquake (January 2001).

town. The layout is based on a linear cluster with internal open spaces as the unit of planning. Each plot has a separate rear access without service lanes. Open spaces are planned to facilitate continuous pedestrian movement across the village. All village amenities are planned and located. Figure 8.4 shows a layout prepared by the village surveyor to obtain approval for converting agricultural land to non-agricultural use. The layout consists of back-to-back row plots with no identification or dis-

Downloaded by [University of Defence] at 01:15 24 May 2016 tribution of open spaces and amenities. The villagers had very little understanding of the qualitative aspects of planning, such as open spaces and amenity provision and location, hierarchy upwards from a house plot, cluster and settlement, etc. The layout sent to the Collector for approval for use-conversion was considered sacrosanct by the villagers, who were in no mood to consider alternative options. Consultants' explanations as well as NGOs' persuasions were of no help. The abundance of funds for rehabilitation meant that villagers were prepared to seek another funding organisation if their demands 212 M.N. Joglekar

Figure 8.4: Layout of Trombey Village as Prepared by the Village Surveyor

NORTH 10 02040 5 10 30 50 Scale (in metres)

Source: Author. Note: Prepared by the village surveyor for obtaining conversion from agricultural to non-agricultural land.

were not accepted. On the other hand the NGO was flush with money, and the villagers accepting funds from others would mean a fall in the NGO's targets and unutilised funds. The NGO therefore succumbed to the wishes of villagers, without any serious attempt to introduce a good quality environment through a better layout. The perceptions of the beneficiaries as well as their surveyor/ draughtsman/architect, are conditioned by the physical reality that Downloaded by [University of Defence] at 01:15 24 May 2016 exists in and defines an urban settlement — comprising gridiron layouts that are regimented and incapable of being a unit of either physical or social/community organisation. If the implementing agency is not able to perceive the right solution, settlement planning suffers — it becomes a casualty. There are innumerable examples of such undesirable physical transformations of rural settlements, particularly in relocated villages. In that context, in-situ rural development is better as it allows the community to retain the original socio-economic ties. Participative Rehabilitation Ú 213

Benefi ciary consultation processes often get infl uenced by the indirect and at times perceived weight of the controlling administrative authority. In a project after the 1993 Killari earthquake, HUDCO prepared a layout plan for the relocated village of Killari. Realising that the relocated Killari will no longer be a small village but a growth centre, concepts based on the incremental growth of individual houses as well as that of the settlement, housing clusters, etc., were incorporated to facilitate the transformation of the settlement. A date and time were fi xed for discussions with the panchayat chief, who was an educated person, in fact a medical doctor. Th e proposals were well received and the HUDCO team was happy. Th en entered the State team. Th e HUDCO team informed the Deputy Collector happily that the discussions were over and the benefi ciaries were happy with the proposal. Not impressed, the Deputy Collector reinitiated discussions and asked the Town Planners to spread their huge drawing (2.5m × 2.5m) on the ground. Th ereafter, the State government offi cers tore to shreds a technically superior proposal arguing that it had nothing in common with conventional Town Planning wisdom as understood by them. It was obvious that the benefi ciaries, who were receiving their daily bread and butter ration from the government at that time could not stick to their earlier views. However, it was a diff erent story that the World Bank appraisal system rejected the Town Planner’s plans, and the district authorities had to invite the HUDCO team once again for an interaction with the World Bank team, which had no hesitation in accepting HUDCO proposal. Th is is again an example of the fallibility of the people’s perceptions. It also indicates a limitation to the theory that the people’s wishes are supreme. While ‘authoritative egos’ were overcome through World Bank intervention, the success story ends there. Administrative arrangements ensured that project management was routine and unimaginative. A planning proposal that included concepts like integrating house design with layout, and facilitating a gradual change from a rural to an urban environment, was destroyed as the in-

Downloaded by [University of Defence] at 01:15 24 May 2016 tegrated house designs were replaced by the housing board’s designs, while the layout proposed originally remained, thus demonstrating the inability of the conventional technocrat in understanding emerging design sensitivities. Th is raises the issue of the ‘where’ of participation. In physical rehabilitation (shelter) participation is possible in design, construction and maintenance activities. Design activities involve designing the shelter, designing shelter groups, and designing the pattern of grouping shelter groups in a settlement with adequate provision of amenities and services. 214 Ú M.N. Joglekar

As one goes up in the hierarchy of these activities, people’s ability to perceive and understand the right choices reduces. At the level of individual shelter design people’s views should be regarded as supreme because their personal lifestyles are involved. Th e clustering of houses also involves considering neighbourly relations, and given a choice people would like to choose their neighbours. However, to have a socially successful cluster of houses, an appropriate physical framework is also necessary. Th e process of choosing a neighbour is as much a function of physical planning as it is of the allotment procedure. Figure 8.5 shows a cluster grouping adopted in the Bhopal Gas victims’ rehabilitation programme. It became such a powerful unit of community organisation that it evoked both acceptance as well as rejection. Where insensitive allotment systems had grouped diametrically opposed groups in a cluster it invited the rejection of that form. Alternatively, where allotment procedures allowed people to choose their neighbours, the physical form was highly appreciated as it provided the right environment through its compact scale and size in terms of the number of units and a place within a manageable group.

Figure 8.5: Housing Project for Gas Victims at Bhopal Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Participative Rehabilitation Ú 215

In choosing construction material and construction systems, traditional practices and people’s ability to handle them is important to keep in mind but these choices should not override the demands of earthquake resistant construction technology. Benefi ciaries are primarily concerned about safety measures. Unlike normal housing, which is constructed for an as yet unidentifi ed customer, disaster rehabilitation benefi ciaries are not only identifi ed but are also present on site. Gaining their confi dence in the use of technologies is of prime importance and can only be achieved by making these visible in the end product. Misinformed benefi ciaries, on the other hand, can spell trouble and cause the projects to get scuttled. At the settlement’s level people’s opinions tend to lose their edge, and a comparatively greater role needs to be played by professionals through a mutual dialogue that also involves the benefi ciaries of the project. Shelter design is an area that determines the possibility of benefi ciary involvement in the construction process. Simplicity of construction systems and the materials that are chosen are the key to initiating and strengthening the self-help process. On the whole, however, the experience of most projects indicates that very few benefi ciaries like to get involved in construction work, unless they already are construction workers.

Project Managers as Managers of the Participative Process

Benefi ciary-sensitive technical professionals have often proved to be better managers of participative processes than non-technical NGOs. Th e Building Centre movement, pioneered through technical manpower, has been playing a leading role in conducting project management through participative processes. Several successful benefi ciary-oriented projects managed by technical persons, particularly of the younger generation, can be cited as examples. Such persons have the benefi t of

Downloaded by [University of Defence] at 01:15 24 May 2016 technical knowledge and familiarity with construction management processes. In construction processes people are most convinced by those who can stand at the site and actually make a strong building come up above the ground. Th ese people can also play a major role in educating the beneﬁ ciaries about technological necessities, the cost-eﬀ ectiveness of various products and their manufacture and distribution. Th e slum upgradation project in Gopeshwar after the 1999 Chamoli earthquake, described below, is one such project, undertaken entirely by HUDCO, with the author as adviser. 216 Ú M.N. Joglekar

A Case Study: Slum Upgradation Project in Gopeshwar, Headquarters of Chamoli District, Uttaranchal

Th e project was taken up essentially through HUDCO, a leading techno- fi nancing institution in the Indian housing sector which is also playing an important demonstrative role in disaster mitigation. Th is project, situated in an urban sector, was selected to ensure that the benefi ts percolate to the neediest at the lowest level. Th e emergence and growth of slums is a major problem, cancerous to the urban fabric of today’s cities. Gopeshwar town, though located in the high hills at an altitude of nearly 2,000 metres, was closest to the epicentre. Th e altitude restricted damage to buildings constructed with kachcha materials like mud, stone, bamboo, etc. Th us, poorly constructed buildings in slums were the fi rst to suff er. Th e basti identifi ed for upgradation was a Harijan basti, not too far from the famous Shiva temple in the heart of town. Th e ancestors of this basti are believed to have been the owners of the land around the temple, but their descendants today have been reduced to the status of basti occupants. Th e basti itself is a high-density settlement with 1,161 persons per hectare. It had 336 persons from 39 families living in the most unhygienic conditions. Located below the bazaar, the basti was the destination of all the waste-water fl owing down from the settlement above. With no drains and paths, the area used to become virtually impossible for human habitation. Houses were constructed with Random Rubble Masonry (RRM) in mud mortar, using wooden members for a mud/stone fl oor and a traditional slate roof. Some of the older houses had been built using plans of traditional heritage house types. Th ese old houses were so badly damaged that they had to be replaced. Some houses of recent origin were identifi ed for retrofi tting.

Downloaded by [University of Defence] at 01:15 24 May 2016 The slum’s population did not have the capacity to throw up a leader from within. Th e literacy level was 30 per cent and average monthly income per capita was Rs 1,500. HUDCO’s project manager — an architect trained in the Laurie Baker tradition — assisted by his young Habitat Engineers had to fi ll the gap. Th is challenge was further complicated by intra-basti confl icts and resultant attitudes made even normal tasks impossible to execute. Th e identifi cation of ownership depended on mutual agreements between neighbours as legal titles were not available. Th e small size of plots made each occupant fi ght to grab Participative Rehabilitation Ú 217

more land. Normal construction got embroiled in boundary disputes and no one was ready to even consider sacrifi cing land for either road widening or a common toilet or a community hall. Benefi ciary attitudes towards each other were thus fraught with the emerging possibility of a dangerous confl ict when the HUDCO team stepped in. Th e fi rst task was to make people understand the value of a house and inculcate in them the pride of a legally-owned ‘own sweet home’. To ensure that the benefi ciaries feel a sense of attachment to the environment being created, their positive involvement in the process of its creation had to be given priority. Construction through benefi ciaries, under the guidance of HUDCO’s project management team, was the implementation mode adopted. However, this was easier said than done. People had no background in construction, and lacked leadership and motivation, which was crucial and had to come from within. As a beginning a committee of basti occupants was formed. After long and serious deliberations, a consensus evolved with respect to plot boundaries, house sites, the widening of drains and paths, and the site for a community hall, as well as infrastructure such as standposts for water supply and their location, etc. Th e next bottleneck was in the release of money to the benefi ciaries for implementation. Th e male members of the community had absolutely no idea of the value of money as most of them were habitual alcoholics, in spite of the fact that the entire Himalayan State as the ‘Abode of God’ (dev bhumi) has prohibition. It was believed that if given the money, these male members would not even use it for construction. Th erefore the idea of giving the grant money to female member/housewife was accepted and that too in small instalments. Th is type of women’s empowerment and their involvement in the participative process was the key to successful project implementation. Th e District Magistrate of Chamoli extended all the necessary administrative and logistical support. Some basti-dwellers were chosen to act as catalytic agents in the development process and were trained Downloaded by [University of Defence] at 01:15 24 May 2016 at the nearest Building Centre at Birahi. Teaching them earthquake- resistant building techniques was a job involving much patience and per- severance. Th e project management staff took an interest in teaching as well as creating an awareness in them about earthquake-resistant construction methodologies both for the new construction as well as for retrofi tting old structures. Habitat Engineers virtually lived in the basti. Some of them even jokingly mentioned that if they were thrown out of their contract jobs, they could now take up teaching jobs in schools or in polytechnics. 218 Ú M.N. Joglekar

In the fi rst phase, the reconstruction of G4 and G5 (for classifi cation criteria see Table 8.2) houses was undertaken because these families were currently living in tents. Local media as well as the citizens of Gopeshwar became very serious about the project once they saw that the project was on the road to success. Some even commented that they had never seen technical persons working with the community with such a feeling of commitment. Th e basti population was initially lured by the cash grant, but soon realised that unless the given small sum of money was used properly, more of it would not come. Th is was a lengthy and painstaking process for the Habitat Engineers, who otherwise, in the normal course, would have released money to a contractor. People also began to realise the value of a home. And women, who were earlier helpless , now developed confi dence and self-respect.

Table 8.2: Classiﬁ cation of Damage to Buildings

Grade 1 (G1): SLIGHT DAMAGE Fine cracks in plaster; fall of small pieces of plaster. Grade 2 (G2): MODERATE DAMAGE Small cracks in walls; fall of fairly large pieces of plaster; pan tiles slip oﬀ ; cracks in chimneys; parts of chimneys fall down. Grade 3 (G3): HEAVY DAMAGE Large and deep cracks in walls; fall of chimneys. Grade 4 (G4): DESTRUCTION Gaps in walls; parts of buildings may collapse; separate parts of the building lose their cohesion; and inner walls collapse. Grade 5 (G5): TOTAL DAMAGE Total collapse of the building. Source: Compiled by the Author.

Retroﬁ tting raised many questions initially, as the people did not have much faith in such a technology. Some of them even compared such ef-

Downloaded by [University of Defence] at 01:15 24 May 2016 forts to the repair of old torn clothes, saying that such repaired clothes would not last long. Families whose houses had been identifi ed for retrofi tting started fi ghting for the reconstruction of the entire house. Th ey could be convinced only through demonstrations. Th ere were numerous occasions when working with people meant going beyond offi cialdom and showing personal initiative. Formalities were completed and a cheque was given to a pregnant woman at her home as she was unable to come to the offi ce. HUDCO’s offi ce work even included writing applications on behalf of the benefi ciaries for the release Participative Rehabilitation Ú 219

of money. Work ethics were thus based on humanitarian considerations. Strict adherence to the principle of not giving the funds in advance made people realise that instalments would not be released without fi rst completing the specifi ed tasks. Th is discipline mixed with humanitarian considerations allowed the project to move ahead smoothly. People, however, at diff erent stages, resisted any technology with which they were not familiar. Th ey initially had a low opinion of the Reinforced Cement Concrete (RCC) solid plank and joist system, but these very people can now explain the system in detail. Th ey also had an aver- sion for retrofi tting techniques, but can now practice these techniques to improve their houses. Th is is the real ‘lab to land’ transformation. Similarly, there was no land for a community centre. Land near the temple was identifi ed as a suitable site. Even then disputes arose. To resolve the issue HUDCO agreed to retrofi t the temple. Th e community participated in retrofi tting the temple and a community hall came up alongside. Initially there was only one water standpost, now they have three. Initially there was no bathroom, now each house has one. Th ere was no toilet with people using open land. Ladies and handicapped persons suff ered the most. Now most of them have individual toilets. Two common toilets were provided for those who could not construct individual toilets due to the inavailability of land adjacent to their house. Th e waste- water channel was enlarged to avoid overfl ow and pedestrian paths have been created on top. Th ere were no trees as people were unwilling to give even a small place for tree planting. Th e consultative mechanism of a People’s Committee ensured the provision of land suffi cient to build chabutaras (platforms) at 30 places and plant trees. Th is will not only provide shade but will also change the physical complexion of the basti. All the houses now have one smokeless chulah and the basti as a whole now has 12 electric poles with sodium vapour lamps.

The Implementation Model of this Participative

Downloaded by [University of Defence] at 01:15 24 May 2016 Slum Upgradation Project

Th e rules laid down and the procedure observed in the Slum Upgradation Project detailed above were as follows:

(i) HUDCO grant aid was provided through its Project Manage- ment Oﬃ ce located in Gopeshwar. (ii) HUDCO’s role was to guide and supervise the construction process that would be undertaken by people. 220 Ú M.N. Joglekar

(iii) HUDCO’s fi nancial assistance of up to Rs 54,000 per family per house was to be provided for those whose damaged houses fell in categories G4 and G5 and needed total reconstruction, Rs 17,500 was earmarked for the houses in the category G3, and Rs 15,000 and Rs 10,000 for G2 and G1 categories respectively. Th ese amounts were disbursed to people according to the progress of work at the site. (iv) Th e benefi ciary had to execute the work under the supervision of a HUDCO/Building Centre Site Engineer. He had to implement earthquake-resistant features like plinth, lintel and roof bands, vertical reinforcement at corners and around openings, and through and cornerstones in RRM. (v) Benefi ciaries whose houses were categorised as G3, G2 and G1, have to retrofi t their houses as per technical guidance provided by HUDCO. (vi) Benefi ciaries had to construct a plinth area of minimum 24 sq. m. with the Rs 54,000 grant. Plans and specifi cations would be provided by HUDCO. Th ey could also provide additional area with their own funding, for which HUDCO’s technical assistance was available. (vii) Th e Building Centre provided materials such as concrete blocks, RCC planks and joists, precast stairs, etc. Sand, cement and steel was to be procured by the benefi ciary. Benefi ciaries were also encouraged to use materials available at site, such as random rubble, old doors/windows, etc. ( viii) To organise the skilled and unskilled labour and to co-ordinate this work with other villagers, HUDCO and the district authorities, a six-member village-level committee was formed. A local NGO Siwani Bahu Uddesh Prashikshan Sansthan stationed a woman volunteer at site to intercede in quarrels that might erupt during construction, so as to defuse any potentially dangerous Downloaded by [University of Defence] at 01:15 24 May 2016 situation. (ix) Each benefi ciary had to open an account in the name of the women in the family. Instalments of grant assistance were deposited in this account by crossed cheques. Th e fi rst release was Rs 2,000 for the G4 and G5 category houses and Rs 1,000 for the rest. Even this was a big amount for the slum-dwellers. Th e benefi ciaries’ own contribution and the State grant was to be utilised along with the HUDCO grant. Participative Rehabilitation Ú 221

(x) Th e value of materials supplied by the Building Centre was assessed and deducted from the cash grant being provided. It was the responsibility of the beneﬁ ciary to collect the material from the Building Centre.

Th e resulting physical transformations not only improved the physical environment but also resulted in changed clean and hygienic personal habits. Eventually, a cleaner and more inspiring physical environment is expected to result with better economic standards of living. Th e overall view from the top of the Gopeshwar hill is so diﬀ erent now. Earlier the basti was identiﬁ ed by blight, dirt and an unhygienic environment. It now resembles a middle-class settlement. Th rough a people-inspired renam- ing process, a Malin Basti (slum) has become Shiv Shakti Nagar.

Lessons Learnt

Th e rehabilitation of disaster-aff ected people is a complex process. Achieving excellence in all aspects of rehabilitation is very diffi cult. Success also depends upon the objectives and the defi nition of excellence. But how does one defi ne the objectives? From whose point of view, and should objectives be political, administrative, technical or social? If all these have to combined or cumulative, would there be a common ground at all or only confl icting interests? While there is no doubt that the priority goes to meeting the aspirations of the aff ected people, in a rehabilitation project, where technical aspects must also have equal importance, whose job will it be to determine rational objectives and get them accepted by actors who might have confl icting views? Will it be the responsibility of the participating government — political or administrative? Will it be the co-ordinating NGO’s job? Or can an enlightened technocrat with community development skills take up the task? Or, lastly, can we perhaps consider an

Downloaded by [University of Defence] at 01:15 24 May 2016 enlightened NGO with proper technical understanding to be the ideal decision-maker? A donor with fi nancial commitment is essential to the needs of rehabilitation, but money power can aff ect the decision-making process adversely. Th is is seen sometimes even in the attitudes of international funding agencies, who may impose the wrong (from the point of view of the recipient) consultants or stipulate impracticable reform conditions. An excess of fi nance — particularly grants — over actual needs has forced experienced NGOs to surrender to people’s wishes, as in Gujarat, while the inadequacy of funds encourages authoritative 222 Ú M.N. Joglekar

attitudes, as seen in Latur during the 1993 earthquake rehabilitation programme. Avoiding a free lunch attitude, ensuring that the beneﬁ ciary has a stake and participates, and insisting on correct technical solutions may be the key to successful implementation of rehabilitation projects. It is, however, important to identify a minimum technical threshold below which community aspirations may need to exercise ﬁ rm control. Perhaps the right combination of community development and technical skills, with a benevolent authoritarian attitude, will be the proper framework for the decision-making process, aiming at excellence in shelter delivery to the victims of natural disasters. Shiv Shakti Nagar of Chamoli, Uttaranchal may be one such approach. Downloaded by [University of Defence] at 01:15 24 May 2016 9 Lessons from the Gujarat Experience For Disaster Mitigation and Management V. Thiruppugazh and Sudhir Kumar

You cannot acquire experience by making experiments… You must undergo it. — Albert Camus

O n 26 January 2001, an earthquake measuring 6.9 on the Richter scale took place in western India. Th e epicentre was located near Chaubari village of Bhachau taluka in Kachchh district, at a distance of about 250 km. west of the city of Ahmedabad. Th e quake lasted for about two minutes, with strong motion lasting for about 25 seconds. Th e eff ect of the earthquake was felt throughout the Indian subcontinent. Th is was followed by hundreds of aftershocks with a magnitude of 3.0 and above. Lessons from the Gujarat experience of the 2001 earthquake are tales that can be narrated in diff erent ways — the struggle of an individual to live after losing his family members, home, etc.; hardships faced by a family whose only breadwinner had died in the quake; the eff orts made by a non-governmental organisation (NGO) or a community-based organisation (CBO) in mobilising resources during the relief and rescue phase; and the process of formulating, implementating and monitoring the gi- gantic Repair and Reconstruction (R&R) project by the Gujarat State Disaster Management Authority (GSDMA),1 the nodal agency in Gujarat

Downloaded by [University of Defence] at 01:15 24 May 2016 for disaster management. Th is paper has tried to ink the learnings from the Gujarat experience in terms of the massive R&R Programme being implemented by GSDMA.

1 GSDMA was created on 8 February 2001 for the implementation and monitoring of R&R programmes and long-term disaster preparedness. 224 Ú Thiruppugazh and Kumar

Damage and Losses

Th e earthquake caused major loss of life, injury and damage to property and infrastructure. Th e death toll was 13,805 and it was estimated that about 167,000 people had been injured. About 1.2 million houses were damaged or completely destroyed. Social infrastructure was also severely hit, with destruction or damage to over 1,000 health clinics and about 12,000 schools. Civil administration was badly hit, with offi cial buildings and records being destroyed and members of the staff either getting killed themselves or suff ering the loss of family members. Physical infrastructure was also severely damaged, including the small and medium earthen dams that provide drinking water and water for irrigation to the local communities, urban and rural water supply systems, public buildings, municipal infrastructure, roads, power and telecommunications. Th ousands of small and cottage industries and agricultural assets were damaged or destroyed. Kachchh district was the worst aff ected, nearly 70 per cent of the buildings there were destroyed. Other districts that were seriously affected included Ahmedabad, Patan, Jamnagar, Rajkot, and Surendranagar. Th ese six districts accounted for over 95 per cent of the deaths caused by the earthquake. Towns such as Bhachau and Rapar of Kachchh suff ered nearly 100 per cent damage. Primary losses have been estimated at US$ 3,189 million, secondary losses at US$ 635 million and tertiary losses at US$ 2,097 million. Details are mentioned in Table 9.1.

Table 9.1: Losses Resulting from the Gujarat Earthquake, 2001

Primary losses Secondary losses Tertiary losses

z Human lives z Export/import z Long-term development

Downloaded by [University of Defence] at 01:15 24 May 2016 z Livestock, other animals z Agricultural output z Overall investment climate z Private property z Industry/services z Funds reallocation output z Municipal infrastructure z Remittance income z Community migration/ relocation z Power and telecom z Fall in earning potential infrastructure due to disability, trauma z Health and education z Unemployment assets z Health hazards Source: Compiled by the author. Lessons from Gujarat 225

Lessons Learnt

Lessons learnt from any event/programme can be broadly inferred on two fronts — an analysis of the end result against the set target, and an understanding of the process. Though both are important and complementary, the latter is crucial because it lays the foundation for the result and hence, in this article, greater focus is on the latter.

The Institutional Set-up

The GSDMA was created on 8 February 2001 as a 'society' under the provisions of the Societies Registration Act and the Bombay Public Trust Act. It reflects the commitment of the Government of Gujarat (GoG) that such a body was created within 15 days of the earthquake in order to implement and monitor the R&R project. The GSDMA was created as a nodal agency which would plan and implement mitigation and pre- disaster preparedness and post- disaster activities such as relief, recovery, R&R activities, etc. The institutional set-up is given in Figure 9.1.

Figure 9.1: Institutional Set-up of GSDMA Government of Gujarat

• Best practices • Policies • Best practices • Expertise !H CO • Funds • Reconstructior > & % • Experiences etc. • Expertise data o p • Project • Damage data monitoring ill

.2 • Damage data • Damage data s? • Risk data • Project plan • Project plans • Progress report 3 g 11 • Initiatives etc. • Capacity creation plans Downloaded by [University of Defence] at 01:15 24 May 2016

CO • Damage • Project guidelines suffered • Project reports 11 • Compensation • Status reports 11 claims, etc. • Funds PQ GSDMA Source: Prepared by the author. 226 Thiruppugazh and Kumar

The chief minister of the State is the ex offido chairman of the organisation's governing body and its other members include two senior ministers, senior secretaries, the Commissioner of Relief, etc. A decision of the governing body is considered at -par with a decision of the cabinet and this helped in taking speedy and timely decisions. The advisory committee reviews the progress of R&R work, and ensures a participatory approach for the same by involving all the stakeholders. It comprises experts from different fields, such as industries, education and health, and also includes two voluntary agencies. To ensure time-bound implementation, a State-level central implementation review group was formed under the chairmanship of the chief secretary of Gujarat. The chief executive officer/additional chief executive officer (CEO/ACEO), GSDMA, the secretaries of line departments, president of the Chambers of Indian Industry (CII)-Gujarat and the Gujarat Chamber of Commerce and Industry (GCCI), etc. are the members. GSDMA's co-ordination with the various stakeholders is presented in Figure 9.2. The institutional set-up and co-ordination between the various bodies and people involved facilitated the participatory, hassle-free and speedy implementation of the programme.

Figure 9.2: Co-ordination of Stakeholders Governing Body GSDMA

CEO GSDMA Central Advisory Committee Implementation Review Group

ACEO GSDMA *ACEO GSMA

Downloaded by [University of Defence] at 01:15 24 May 2016 Joint CEO GSDMA Disaster Management Community Participation

Director Director Controller Chief Engineer Director (Administration) (Finance) of Accounts (Procurement) (H&EC)

Source: Prepared by the author. Lessons from Gujarat Ú 227

Legislative Measures

Disaster Management Policy

Gujarat is vulnerable to such natural disasters. Th erefore, as a proactive step the GSDMA formulated a ‘Disaster Management Policy’ for the State. Th is policy was approved by the Gujarat cabinet and is already in force. Th e primary objectives of this policy are:

z To assess the risks and vulnerabilities associated with various disasters and develop appropriate mitigation and prevention strategies. z To provide clarity on the roles and responsibilities for all stakeholders concerned with disaster management, so that disasters can be managed more eﬀ ectively. z To create awareness and preparedness, and provide advice and training to the agencies involved in disaster management and to the community.

Disaster Management Act

In 2003, Gujarat also enacted a Disaster Management Act that came into force on 13 May 2003 — thus becoming the fi rst Indian State to provide a legal and regulatory framework for disaster management. Th e purpose of this Act is to provide for the eff ective management of disasters, mitigation of the eff ects of disasters, administering emergency relief during and after the occurrence of disasters and implementing, monitoring and co-ordinating measures for reconstruction and rehabilitation in the aftermath of disasters. And, for these purposes, to establish the GSDMA and specify other agencies for matters connected therewith or Downloaded by [University of Defence] at 01:15 24 May 2016 incidental thereto. Th e objectives of the Act are:

z To clarify the roles of the principal actors in disaster management. z Emphasise moving from relief to all phases of disaster management — mitigation, relief, reconstruction, and rehabilitation. 228 Ú Thiruppugazh and Kumar

z To enable the appropriate parties to use the existing powers, authority and resources to implement the disaster management framework and provide additional powers wherever necessary. z To provide a framework for sourcing and utilising funds, and sup- plementing existing legislations.

Revision of Building Codes

A large number of engineered buildings, including RCC structures, collapsed because of the earthquake and caused a large number of deaths. Inappropriate design and construction technology can be cited as the reasons for this. Hence, immediately after the quake, the existing Gujarat Development Control Regulations (GDCR) were amended to include mandatory provisions of the Bureau of Indian Standards (BIS) concerning codes for seismic, wind and fi re safety. A review of the existing building code provisions of the BIS is also underway, the broad objective of which is to identify inadequacies, if any, and recommend changes. In addition to this, some of the above-mentioned legislative measures that pertain specifi cally to disaster management have been taken so as to give legal teeth to the managers of the project. It also refl ects the commitment of the GoG towards disaster mitigation.

Housing Reconstruction

Housing was the worst aff ected sector after the quake — nearly 12 lakh houses suff ered serious damage, of which 2.1 lakh required complete reconstruction and the rest required repair or retrofi tting. To deal with this the government announced a comprehensive R&R policy, titled the ‘Gujarat Earthquake Reconstruction and Rehabilitation Policy’. Th e salient features of this programme were:

Downloaded by [University of Defence] at 01:15 24 May 2016 (i) Benefi ciaries were given a wide range of choices for reconstruction, such as the owner-driven approach of reconstruction in in-situ sites and the Public–Private Partnership (PPP) Programme for partial relocation or complete relocation. In the owner-driven approach, reconstruction activity was to be undertaken by the benefi ciary whereas in the PPP Programme, NGOs were to work in consultation with the government and the benefi ciary for the task of reconstruction. Th e bottom line, therefore, was choice to the benefi ciary. Lessons from Gujarat Ú 229

(ii) Community participation was the key in the housing reconstruction programme. Th e approval of the Gram Sabha, comprising all the adult members of the village, was mandatory before deciding on a site for the relocation of an aff ected village. (iii) Th e reconstruction programme was spread across 21 districts. As a result, a decentralised implementation approach was used. Th e programme was co-ordinated by the concerned district collectors/district development offi cers at the district level and by the mamlatdar/taluka development offi cers at the taluka level. (iv) Registration of the reconstructed house under the joint- ownership of husband and wife was made mandatory. (v) All reconstructed houses were insured against risk from 14 types of perils, including earthquake, cyclone, fi re, etc. (vi) Th e reconstructed houses are multi-hazard resistant. (vii) To keep a tab on the quality of construction, a two-pronged strategy has been deployed. Th e assistance amount promised to the benefi ciary (except G-1 category) was given in two or three instalments. Th e fi nal instalment was to be released to the benefi ciary only after quality certifi cation by the engineer. ( viii) Apart from inspection and certifi cation by the engineer, the National Council for Cement and Building Materials (NCCBM), an independent, quality audit agency has been hired. All reconstructed houses were audited by the agency at various stages of reconstruction and a report was submitted to the concerned district administration and the GSDMA for ratifi cation. (ix) Th e four worst-aff ected towns of Kachchh were to be reconstructed as state-of-the-art towns. High quality infrastructure was laid in the old unplanned and unregulated towns. Develop- ment plans were prepared with a 50-year perspective and town planning with 20 years of growth in view. Downloaded by [University of Defence] at 01:15 24 May 2016 (x) Material banks were set up so that the benefi ciaries could get cement at an aff ordable price. Materials were procured by ne- gotiating directly with the manufacturers, with the promise of exemption from the incidence of sales tax. Th us allowing the benefi ciary to get it at a reduced price. Nearly 1,180 material banks were established and more than 18.5 million cement bags have been disbursed. 230 Ú Thiruppugazh and Kumar

(xi) To address grievances, an ombudsman, a village-level committee, a district -level committee, etc., have been set up. Th e District Judge has been declared as the ombudsman and more than 46,000 applications have been already addressed by the same. (xii) Th e multi-hazard resistant reconstruction task could not have been completed in the stipulated time with the existing number of engineers and masons. Hence, massive training programmes were undertaken. More than 6,000 engineers and 26,000 masons were trained.

Th ere were certain other features as well. Th e assistance amount had been derived through unambiguous guidelines and elicited very few grievances. Th e aff ected houses were divided into fi ve categories, according to the severity of the damage: from G1 — buildings with minor damage of about half-inch cracks, up to G5 — buildings with damage such that more than 50 per cent of the structure had fully collapsed. However, repeat surveys were also done if any benefi ciary asked for the same. Assistance was calculated based on the extent of damage/carpet area of the aff ected house. Th e ceiling amount for assistance was fixed in diff erent packages: for example, cash assistance up to Rs 1.75 lakh was provided for house reconstruction in urban areas, and up to Rs 90,000 for rural houses.

Initiatives for Livelihood Restoration and Sustainability

As a result of the earthquake, over 84,000 self-employed persons, such as artisans, craftsmen, and those owning and employed in small business, etc., lost their livelihoods, and the work of around 80,000 farmers was aﬀ ected. Restoration of livelihood and ensuring its sustainability were

Downloaded by [University of Defence] at 01:15 24 May 2016 the two principle objectives in the formulation of livelihood restoration packages.

Cottage Industries

To restore the livelihoods of the aﬀ ected people, over US$ 15 million have been spent. Six packages, as described in Table 9.2, were formulated for this purpose. Lessons from Gujarat Ú 231

Table 9.2: Packages for Livelihood Restoration

Package I Toolkits to artisans Package II Distribution of looms to handloom weavers Package III Toolkits to handicraft artisans Package IV Loan subsidy to self-employed persons Package V Tools to masons Package VI Working capital assistance to handloom weavers Source: Compiled by the author.

For sustainability, measures were undertaken to hone the skills of artisans. Students of the National Institute of Design (NID) and National Institute of Fashion Technology (NIFT) were involved in the upgrading and improving of the design and quality of products. Th e Central Leather Research Institute (CLRI) was involved in training artisans in making leather products using the latest technology. NID was involved in training leather-workers on how to improve products, and a design school is to be set up in Kachchh. Th ree-month courses will be introduced in this institute for design upgradation.

Women’s Livelihood

Th e Women’s Livelihood Restoration Project (WLRP) was started to ensure that the 20,000 women in 15 talukas of the worst-aﬀ ected districts have access to a wide range of services connected with their livelihoods. Th e uniqueness of this project is that it aims to look for integrated solutions that involve capacity building and enabling women to earn higher and more sustainable levels of income through a process that empowers them. Some important activities undertaken under this project are agri-inputs, tailoring, handicraft, self-help group (SHG) formation, dairying, training in embroidery, etc.

Self-Employment Activities Downloaded by [University of Defence] at 01:15 24 May 2016

Women were assisted as per their requirement of, for example, working capital assistance or entrepreneurial skills. Four projects were taken up especially with the objective of spreading the movement of co-operatives and best practices in the milk sector. Th ese activities helped create self-employment opportunities for 5,000 poor women, leading to their sustainable development. Some important activities covered by the programme are: paper dish and cup making, broom-making, ﬂ our mills, 232 Ú Thiruppugazh and Kumar

machine embroidery, catering service, tiﬃ n service, carts, handloom, stitching of rexene bags, potato/banana chips, etc.

Industries

Small industrial units, cabins and shops, and service and trade units that were aff ected by the earthquake were identifi ed. Several schemes such as loans for repair and reconstruction, rescheduling of loans, insurance claims, subsidy and interest subsidies to industrial units, fi nancial assistance to small industrial units, small cabins and shops are being implemented for the rehabilitation and long-term sustainability of these units. Cash assistance up to Rs 3,000–6,000 for each aff ected shop and cabin unit has been disbursed to 13,164 units. Subsidies and interest subsidies were disbursed to 3,465 aff ected service and trade units. Depending on the damage, compensation has been paid to salt farmers, and lease rent and royalty have been exempted on salt production in the aff ected areas.

Agriculture

In order to bring farming activity back to normal, assistance has been provided in the form of input kits and ﬁ nancial assistance for the rehabilitation of farm structures and irrigation assets. Input kits such as tarpaulins, spray pumps, farming tools, storage bins, seeds, chemical fertilisers and pesticides were distributed to the 58,163 aﬀ ected farmers. Assistance was provided for the rehabilitation of 46,872 damaged/ destroyed farm structures such as engine rooms, store rooms, etc.

Awareness Campaign

GSDMA launched a massive awareness campaign incorporating various Downloaded by [University of Defence] at 01:15 24 May 2016 information, education and communication (IEC) activities in the entire region. Almost all available media of mass communication were utilised for this purpose, including advertisements in newspapers, hoardings, pocket-sized booklets, shake-table demonstrations, folk dance and music, jokes in the local language, etc. A few notable examples are:

z Messages about hazard-resistant construction and retroﬁ tting were displayed on State transport buses (500 in number) plying in the aﬀ ected areas. Lessons from Gujarat Ú 233

z Two video cassettes on the technical aspects of reconstruction and retroﬁ tting of houses were prepared and shown in public gatherings in over 2,500 villages. z A video cassette of the local folk dance ‘Bhavai’ was prepared and shown. z Street plays were used to inculcate a disaster management culture. z An essay writing competition for children to improve awareness about earthquakes. z Photography and videography competitions on destruction caused by the earthquake. z GSDMA co-ordinated with and involved a number of NGOs (about 100) in an awareness campaign, and a camp oﬃ ce was set up by GSDMA in Bhuj town for this.

Th us, evidently, the underlying theme of the above-mentioned initiatives was to introduce a disaster management culture into people’s day-to-day lives.

Capacity Building

Gujarat Institute of Disaster Management [GIDM]

GIDM was established as a training and research wing of GSDMA on 26 January 2004 by the Gujarat government. With state-of-the-art facilities, the principal thrust of the institute is to provide proper insight, knowledge and skills to stakeholders at various levels in the State, so as to accomplish the appropriate disaster management goals. GIDM also conducts research in the ﬁ eld of disaster management. Th e major objectives of the institute are:

z To improve overall awareness regarding disaster management and

Downloaded by [University of Defence] at 01:15 24 May 2016 serve as the apex institute in the State for disaster management. z To hone the skills of disaster managers and provide necessary training to all stakeholders. z To facilitate partnerships with eminent national and international organisations, universities, institutions, bodies, and individuals specialising in disaster management. z To conduct research in the ﬁ eld of disaster reduction and prevention, and act as a resource centre and clearing house of information, including case studies, lessons learnt and best practices. 234 Ú Thiruppugazh and Kumar

Institute of Seismological Research (ISR) Gandhinagar

ISR was established by the Science and Technology Department, Government of Gujarat, in 2003. ISR is the only institute in India fully dedicated to seismological research and is planned to be developed into a premier international institute with deemed university status in a few years time. Th e Institute will network with similar centres within the State, country-wide and all over the world for research and sharing of best practices. Major areas of emphasis are earthquake information, earthquake mechanics and engineering, monitoring and computational geo-science, global geodynamics and earthquake modelling, and earthquake mitigation strategy.

Feasibility Studies/Plans on Disaster Management Themes

GSDMA hired international experts/consultants for carrying out feasibility studies, preparing plans, models, etc., on various aspects of disaster management. Based on the ﬁ ndings and suggestions of the consultants, further courses of action were chalked out. Some of the studies that have been conducted are hazard risk and vulnerability assessment, feasibility of seismic microzonation, damage and loss assessment methodology, early warning system and emergency communication, beneﬁ t monitoring, revision of building codes, and emergency response centres.

Workshops/Seminars

A number of international and national workshops/seminars have been organsied by GSDMA on various themes of disaster management. Th e objective of these was to share knowledge and experiences, discuss and Downloaded by [University of Defence] at 01:15 24 May 2016 evolve plans, spread awareness, and impart training, etc.

Masons’ Training

Training modules and materials (which included both theoretical and practical components) were standardised to maintain uniformity of content. Training was arranged at villages and taluka headquarters to facilitate easy access for the masons attending these programmes. Lessons from Gujarat Ú 235

Considering the loss of income to the masons during the training period, food, travel expenses, and a daily per-diem allowance was arranged to compensate for their loss of wages.

Certiﬁ cation of Masons

It is very common in the rural areas of Gujarat that masons not only construct but also design the houses. A large number of practicing masons in Gujarat learnt these skills from hands-on experience in the ﬁ eld. Th e practice of masonry in Gujarat is more by inheritance than by scientiﬁ c training. Th e inadequate skills and the role of existing masons in the rural areas call for an enhancement of their skills. A system has also been evolved for the licensing of masons.

Engineers’ Training

Specialised training was provided to government engineers, private engineers and engineers working with NGOs. Training programmes were conducted at various places, including reputed institutes such as the Indian Institute of Technology at Kanpur and Mumbai. Th ree multidiscip- linary teams have been given advanced training in rescue and relief in the Netherlands and Germany. Th e strengthening of municipalities by imparting training and providing equipment is also underway. Mod- ernisation and strengthening of the ﬁ re department is in progress, and advanced search-and-rescue equipment have been acquired for the same. Th ese initiatives (GIDM, ISR, masons’ certiﬁ cation, workshops, etc.) are being undertaken to update and exchange information and ultimately create roots for a disaster management culture among all the stakeholders.

Conclusion Downloaded by [University of Defence] at 01:15 24 May 2016 The processes and outputs of the Gujarat Emergency Earthquake Rehabilitation Programme (GEERP) provide some useful insights for future programmes. Notably, a comprehensive R&R policy, providing a framework for the recovery phase, has to be in place before the actual implementation of the programme. Th e policy must be supported by an eﬀ ective legal and institutional framework. In Gujarat, for instance, GSDMA, the Gujarat State Disaster Management Act, the Gujarat State 236 Ú Thiruppugazh and Kumar

Disaster Management Policy, etc., were put in place under such a legal and institutional framework. Th e R&R programme always off ers an opportunity to include risk reduction features in reconstruction works and lay the foundation for long-term disaster preparedness. In the case of GEERP, the revision of building codes, mass IEC campaigns, specialised studies on diff erent themes of disaster management, setting up GIDM and ISR, etc., are some of the long-term risk reduction interventions. As a rehabilitation strategy, relocation should be undertaken only as the last option. It should be undertaken in cases such as unsafe ground conditions at the existing site. In the case of relocation, it should be done with the active involvement of NGOs and the communities. Th e relocation plans in the rural areas of the GEERP refl ect such an approach, with the active participation of the community. In the case of urban areas, with their specifi c complexities, the problems of relocation become even more serious. Community involvement and the facilitation of these processes through NGOs/CBOs is therefore the key in relocation. Th e R&R programme should also ensure community participation so that the varied needs of the community can be addressed. In case of the GEERP, the reconstruction of houses used a varied approach from owner-driven to PPPs, which gave people choices for their restoration of their houses. Th e Gujarat experience also clearly pointed out that NGO’s have a profound role to play in any R&R programme as they facilitate and enable community participation and involvement in the various decision- making processes, thereby getting help from the community for successful implementation whether it is relocation, urban planning, identifi cation of benefi ciaries for livelihood programmes or any other such contentious issue. It is important also to invest in the capacity building of engineers, architects, masons, etc., during the R&R programme as this trained

Downloaded by [University of Defence] at 01:15 24 May 2016 manpower will help in adding to the safe habitats that are constructed in large numbers after the disaster. Also, this trained manpower will be a long-term asset for society. Th e challenge is to create an institutional mechanism for the training and certiﬁ cation of masons. Th e question of livelihood restoration is another key indicator of success of any R&R programme. Such programmes oﬀ er an opportunity not only to restore livelihood but also to invest in the enhancement of skills of the artisans. A mechanism can be set up to establish new linkages with market, as done in the GEERP. Lessons from Gujarat Ú 237

Disasters of such a scale as Gujarat, and post-disaster reconstruction and rehabilitation programmes generate a huge amount of debris. It is important to have an environment management framework in the R&R programme that includes a debris management plan that provides guidelines for estimating debris quantity, removal methodology, identifi cation of disposal sites and closure plans. An eff ective programme monitoring and evaluation system provides insight/feedback regarding various complexities and issues at the ground level. Th ese insights/feedbacks help in the mid-course correction of policies. Hence, a combination of top-down and bottom-up approaches in information fl ow and decision-making and should be an integral part of R&R programmes. In conclusion it can be unambiguously stated that the success of any R&R programme cannot be judged simply on the basis of the physical numbers achieved, the pace of implementation or geographical spread, but must also consider the long-term sustainable initiatives that are taken up in the form of various IEC campaigns and strategies, leading to capacity building of the community and the State at large. Th is is what results in a disaster resilient society and in sustainable development. Downloaded by [University of Defence] at 01:15 24 May 2016 10 Disaster Management in India Policy Interventions and Recent Initiatives Anil K. Sinha, Dr Satendra and Shikha Srivastava

Disasters occur in India with rather grim regularity, causing enormous loss of life and property. Almost 85 per cent of the country is vulnerable to one or multiple disasters, and about 57 per cent of its area — including the capital — lies within a high seismic zone. Approximately 40 million ha. of the country’s land is prone to fl ood, about 8 per cent of the total land mass is prone to cyclone and 68 per cent of the area is susceptible to drought. Of the 35 States and union territories in the country, 27 are disaster prone; and one State, West Bengal, is prone to disasters of four types. In the recent past, global warming and climate change have added a new dimension to community risks and vulnerabilities to disasters. Th e threat of natural disasters has further increased in terms of the intensity and frequency of fl oods, cyclones and droughts, coastal inundation and erosion, saline intrusion, deforestation, loss of bio-diversity and agriculture, etc. A number of special programmes have been in operation for many years to mitigate the impact of natural disasters. And, over the centuries, local communities have also developed their own indigenous survival mechanisms. Th is rich storehouse of knowledge is also a part of the country’s inheritance and its legacy. Offi cially, disaster management in the country dates back to 1878, when a Scarcity Relief Division was set up in the Ministry of Agriculture to deal with the problem of food scarcity. With the changing pattern Downloaded by [University of Defence] at 01:15 24 May 2016 of disasters and their expanding spectrum in terms of types, this set- up, originally dealing only with droughts and famines, later came to encompass the management of all disasters. In the post-Independence period, the country developed an integrated administrative machinery for disaster management at the national, State, district, and block levels. Th e responsibility for carrying out rescue and relief operations in the event of natural calamities lies with the concerned State government; the union government supplements the State’s Disaster Management Policy and Initiatives in India Ú 239

relief efforts by initiating supportive action. Elaborate procedural mechanisms are outlined in relief manuals and codes, and are reinforced by contingency action plans that facilitate emergency management operations. However, these eff orts remained primarily relief-oriented, with little effort put into disasters mitigation, prevention and preparedness. A paradigm shift took place in the approach to disaster management with the inauguration in 1990 of the United Nations International Decade of Natural Disaster Reduction (IDNDR). Hereon, the focus increasingly shifted towards prevention and mitigation of disasters. In India, the eff orts towards a more holistic and comprehensive disaster mechanism took on concrete shape in 1999, with the constitution of a High Power Committee (HPC) on Disaster Management. Th e mandate of the HPC was to suggest institutional measures for the eff ective management of disasters, both natural and man-made. It was the fi rst serious attempt made in India towards evolving a systematic and broad-based approach towards all disasters.

Important Recommendations of the HPC

Th e vision of the HPC was to create a disaster-free India through adherence to the principles of preparedness, quick response, strategic thinking, and prevention. Th ese principles rest on the premise that while hazards — natural as well as man-made — are inevitable, the disasters that follow need not be, and that society can be prepared to deal with them eﬀ ectively whenever they occur. On the basis of a long and comprehensive exercise, the HPC came out with a large number of recommendations dealing with the constitutional and legal framework, organisational structures and institutional mechanisms in the overall system of disaster management in the country. Th ese are listed below.

Downloaded by [University of Defence] at 01:15 24 May 2016 (i) Appropriate mention of disaster management in any one list of Schedule 7 of the Constitution of India. (ii) Suitable legislation at the national and State levels. Drafts of a National Act for Calamity Management and a model State Disaster Management Act were prepared and submitted to the government for consideration. Subject to the enactment of suitable legislation, it would be required to evolve detailed regulations to help in the enforcement of law. 240 Ú Sinha, Satendra and Srivastava

(iii) Establish continued and sustained focus by constituting a Cabinet Committee on Disaster Management. (iv) Th e All-party National Committee (under the chairmanship of the Prime Minister) should be re-christened National Council on Disaster Management and its scope enlarged to include man-made disasters as well. Subsequently, the Council and the Working Group should be institutionalised as permanent standing bodies. (v) A nodal Ministry of Disaster Management to be set up for sustained and focused eff orts in the area of disaster preparedness, mitigation and management. (vi) A National Centre for Calamity Management, as suggested by the Eleventh Finance Commission and its structure as evolved by the HPC, needs to be set up at the earliest. (vii) A National Institute for Disaster Management to be established as a centre of excellence for the creation of knowledge and its dissemination including training and capacity building through symbiotic linkages with other institutions. (viii) Integration of development plans with disaster mitigation strategies. (ix) At least 10 per cent of the plan funds at the national, State and district levels should be earmarked and apportioned for schemes that specifi cally address prevention and preparedness. Th e HPC was of the view that the Calamity Relief Fund (CRF) be reconstituted and the funds provided by the Tenth Finance Commission be earmarked for State-level capacity building for disaster management and to act as a buff er for handling State- and district-level emergencies. (x) Precision Geographical Information System (GIS) and digital maps of all States, districts and urban centres, with spatial and non-spatial data, be made available at appropriate scales. Th e

Downloaded by [University of Defence] at 01:15 24 May 2016 HPC also felt the need to give a ﬁ llip to forecasting and early warning systems. (xi) Set up a network of state-of-the-art control rooms and Emer- gency Operations Centres (EOCs) in national and State capitals and in the headquarters of disaster-prone or vulnerable districts. (xii) A National Disaster Knowledge Network should be established to cover natural, man-made and biological disasters and should serve the needs of disaster managers, decision makers, aﬀ ected communities, and the media. Disaster Management Policy and Initiatives in India Ú 241

(xiii) Training for the various players involved in disaster management and the response mechanism to be worked out in detail for every situation. (xiv) A National Disaster Mitigation Strategy, comprising the creation of policy supports, improving public awareness and strengthening institutional infrastructure, among other things. (xv) Insurance should be made mandatory in disaster-prone areas and hazardous areas should be announced, notiﬁ ed and publicly displayed.

The Changing Face of Disaster Manage ment

In the wake of the Gujarat and Orissa disasters, the Government of India (GoI) gave serious thought to the recommendations of the HPC and the Eleventh Finance Commission on disaster management and took certain policy decisions. Disaster management was moved from the purview of the Ministry of Agriculture and Co-operation to the Ministry of Home Aﬀ airs. Th is shift heralded a new chapter in the future course of disaster management in the country.

Shift to the Ministry of Home Affairs

To legalise this transfer of subject, in February 2002, the Government (Allocation of Business) Rules, 1961 were accordingly amended and in June 2002, the formal transfer took place. Based on the HPC’s recommendations and lessons learnt from past disasters, the GoI made major changes to the institutional framework of disaster management, giving due consideration to the major recommendations of the HPC relating to the constitutional and legal framework, community participation,

Downloaded by [University of Defence] at 01:15 24 May 2016 capacity building, fi scal administration, and relief. Th e fi rst step taken in this direction was to draw up the National Disaster Framework. Th is framework gives a broad outline of government policy and covers major aspects of disaster management, like the institutional framework, disaster prevention strategies, early warning systems, disaster mitigation, and preparedness. It identifi es and discusses, at a macro level, the areas of intervention, the inputs required and the agencies that need to be involved at various stages. 242 Ú Sinha, Satendra and Srivastava

HPC Recommendations and Follow-up

Th e disaster management policy and institutional framework that existed in India were fairly elaborate and comprehensive as far as response, relief and rehabilitation are concerned, however, little importance had been given to prevention, mitigation and preparedness. Th e HPC, giving due consideration to these aspects, made several recommendations to upgrade institutional and legal frameworks, policy and planning methods, capacity-building strategies, technological developments, etc., at the State and regional levels. Following the recommendations of the HPC, the GoI took a number of steps to strengthen disaster risk management systems at the national, regional and local levels.

National Level

To handle disaster management more effectively, the HPC recommended setting up a separate department for this at the level of the union government. Following this, on 28 September 2005 the National Disaster Management Authority (NDMA) was set up, headed by the Prime Minister, as the apex body for disaster management. Th e NDMA has been established with the singular and exclusive objective of developing a safer and a more disaster resilient India by developing a holistic, proactive, multi-hazard and technology-driven strategy for disaster management. Th e NDMA has taken many steps to evolve such strategies and strengthen the existing framework of disaster management. Some of the main responsibilities and areas of involvement of the NDMA are as follows:

(i) Enactment of the National Disaster Management Act, 2005. (ii) Formulate policies on disaster management.

Downloaded by [University of Defence] at 01:15 24 May 2016 (iii) Carry out vulnerability assessment for speciﬁ c disasters in all States. (iv) Approve the National Plan and other plans prepared by the ministries and departments of the Government of India. (v) Lay down guidelines to be followed by State authorities while drawing up State plans. (vi) Recommend the provision of funds for the mitigation of disasters and also utilise the same. Disaster Management Policy and Initiatives in India Ú 243

(vii) Lay down the guidelines to be followed by diff erent ministries or departments of the government for the integration of measures for the prevention of disasters or the mitigation of their eff ects in their development plans and projects. (viii) Institute a large-scale, nation-wide eff ort aimed at capacity development and the training of all functionaries, stakeholders and communities, with the active involvement of the NIDM and National Disaster Response Force (NDRF) training institutions and battalions. (ix) Equip and train an eight-battalion-strong NDRF to provide specialised response in all type of contingencies. (x) Position the National Disaster Mitigation Reserves to cater to the needs of 300,000 aff ected people in the fi rst 48–96 hours, jointly with the NDRF battalions at various strategic locations. (xi) Upgrade fi re services and re-orient civil defence. (xii) Co-ordinate enforcement and ensure the implementation of disaster management policies and plans. ( xiii) Take all such measures that are deemed necessary in order to prevent future disasters and to ensure, in the event of a disaster or a similar threatening situation, capacity development and preparedness in order to mitigate impact. (xiv) Provide similar support to other countries aff ected by major disasters as determined by the central government. (xv) Exercise general superintendence, direction and control over the NDRF. (xvi) Provision of a user-friendly forecasting and early warning system and ensuring eff ective dissemination through a fail-safe communication system that runs right down to district level and below. (xvii) Promote use of disaster-resistant technologies in the design of

Downloaded by [University of Defence] at 01:15 24 May 2016 habitats, with special attention to lifeline structures (existing as well as well as those in the pipeline), including schools and hospitals. (xviii) Recommend norms and guidelines for the minimum standards of relief to be provided to persons aﬀ ected by disasters. (xix) In case of disasters of severe magnitude, recommend relief in the repayment of loans or the grant of fresh loans to persons aﬀ ected, on such concessional terms as may be appropriate. 244 Ú Sinha, Satendra and Srivastava

Th e NDMA has been established as an apex body not only to strengthen and develop the disaster management system at the national level, but also to work closely with States and proactively support them in their work.

State Level

Since disaster management is a responsibility primarily of the State government, they have been advised to set up disaster management authorities (State Disaster Management Authorities or SDMAs) under the chairmanship of the Chief Minister. Th e purpose behind setting up such an authority is to ensure that disaster mitigation and preparedness are seen as the joint responsibility of all the concerned departments and disaster management concerns are therefore included in the State’s programmes. Th e National Disaster Management Act, 2005 makes it mandatory for all State governments to prepare their own State and district-level disaster management plans. As per the Act, all the developmental departments of State governments too must prepare and implement disaster management plans, in consultation with the SDMA. A number of States following this directive have already set up such an authority and have also come out with their own disaster management plans and acts. Th e Government of India has also suggested to the State governments to convert the departments of Relief and Rehabilitation into departments of Disaster Management, with an enhanced area of responsibility that includes mitigation and preparedness in addition to relief and rehabilitation. Th is has already been implemented in many States and union territories. Th e States have also been advised to involve local community-based organisations and NGOs in all stages of disaster management.

Downloaded by [University of Defence] at 01:15 24 May 2016 District Level

Th e district administration is the focal point for the implementation of all State plans and activities at this level, including those relating to disaster management. As per recent initiatives, the district magistrate will be responsible for co-ordinating all activities relating to disaster prevention, mitigation and preparedness, in addition to his existing responsibilities — response and relief. Th e National Disaster Management Act, 2005 provides for establishment of District Disaster Management Disaster Management Policy and Initiatives in India Ú 245

Authorities (DDMAs) with the district magistrate as the ex oﬃ cio chairperson, the elected representative of the local authority as the ex oﬃ cio co-chairperson and other department heads as members. Th is authority will act as the district planning, co-ordinating and implementing body for disaster management and will take all the measures necessary for the purpose of disaster management, in accordance with the guidelines laid down by the national and State authorities.

Local Level

At the block and taluka level, Disaster Management Committees (DMCs) have been created in many of the multi-hazard prone districts in many States of the country. In these districts, the government is also taking steps to constitute Disaster Management Teams (DMTs). Each village will have a Disaster Management Plan. Th e process of drafting the model disaster management plan has already begun. Th e DMC, which draws up the plans, consists of elected representatives at the village level, local authorities and government functionaries, including doctors and paramedics of primary health centres located in the village as well as primary school teachers. Th e plan encompasses prevention, mitigation and preparedness measures. Volunteers from various organisations like Nehru Yuva Kendra, National Social Service (NSS) and other NGOs will also be made members of the DMTs. Th e teams are being provided basic training in evacuation, search and rescue, etc. Th e committee will review the plan at least once a year. Generating awareness among people in the village of speciﬁ c risks and hazards, based on the vulnerability of the village, is the responsibility of these committees. Panchayati Raj Institutions (PRIs) in rural areas and the local municipal bodies in urban regions too, similarly, have deﬁ nite roles in all spheres of disaster management and the States are expected to ensure their active involvement. Th e GoI–UNDP Disaster Risk Manage-

Downloaded by [University of Defence] at 01:15 24 May 2016 ment (DRM) Programme, being implemented in 17 disaster-prone States and union territories, is also helping State governments ensure maximum community involvement in disaster risk reduction activities.

Disaster Management Acts and Policies

To make the above-mentioned institutional system more eﬀ ective and responsive, there is a need to equip the authorities with some legislative powers. With this objective in mind the HPC emphasised the need for 246 Ú Sinha, Satendra and Srivastava

a National Calamity Management Act, and similar State-level acts, relating to the prevention and mitigation of disasters. Following this recommendation, the union government formulated a comprehensive legislation providing for a legal and institutional framework of disaster management at all levels in the country in the form of the Na- tional Disaster Management Act, 2005. The Act concentrates very comprehensive powers and functions at the national level to deal with disasters, providing adequate powers to the various authorities co- ordinating mitigation, preparedness and response. A number of States have already enacted their own acts or adopted the central Act. Th ey have also been advised to convert their relief codes into disaster management codes by including aspects of prevention, mitigation and preparedness. Th e GoI has decided to articulate a national policy on disaster management, covering almost all spheres of central government activity, and which will take precedence over all existing sectoral policies. Th e broad objectives of the policy are to minimise the loss of lives and social, individual and community assets because of natural or man-made disasters and contribute to sustainable development and better standards of living for all, and more speciﬁ cally for the poor and vulnerable sections, by ensuring that development gains are not lost through natural calamities. Th e policy framework is in conformity with the International Strategy for Disaster Reduction, the Rio Declaration, the Millennium Development Goals and the Hyogo Framework for Action 2005–15. Central ministries, the States and other stakeholders have been involved in the participatory and consultative process of evolving the necessary policies and guidelines, keeping in mind the following objectives.

(i) Adopting a holistic and proactive approach for disaster prevention, mitigation and preparedness. (ii) Promoting a culture of disaster prevention and preparedness,

Downloaded by [University of Defence] at 01:15 24 May 2016 by making disaster management an overriding priority at all levels and at all times. (iii) Encouraging mitigation measures based on state-of-the-art technology and environmental sustainability. (iv) Mainstreaming disaster management concerns into the development planning process. (v) Putting in place a streamlined institutional techno-legal framework in order to create and preserve the integrity of an enabling regulatory environment and a compliance regime. Disaster Management Policy and Initiatives in India Ú 247

(vi) Developing contemporary forecasting and early warning systems backed by responsive and fail-safe communications and information technology (IT) support. (vii) Promoting a productive partnership with the media, NGOs and the corporate sector in the areas of awareness generation and capacity development. ( viii) Ensuring an eﬃ cient response and that relief work is done with a caring and humane approach towards the vulnerable sections of society. (ix) Making reconstruction an opportunity to build disaster-resilient structures and habitats.

Disaster Mitigation, Prevention and Preparedness

Mitigation, prevention and preparedness are the most essential components of disaster management. Due to various reasons these aspects had not been given due consideration in earlier policies and plans. How- ever, in its recent policy, the GoI has taken some essential steps to take proper care of these components; the state has adopted mitigation and prevention as essential aspects of its development strategy. Th e Tenth Five Year Plan document has a detailed chapter on disaster management. It emphasises the fact that development cannot be made sustainable unless mitigation is built into the developmental process. Giving this aspect due consideration, a comprehensive programme has been evolved for disaster-speciﬁ c mitigation, preparedness and response. Th e important components of this programme are discussed here. Although the Bureau of Indian Standards (BIS) has laid down certain standards for construction in diﬀ erent seismic zones, these were not being followed. It is the Town and Country Planning Acts and Building Regulations that regulate building construction in urban and suburban

Downloaded by [University of Defence] at 01:15 24 May 2016 areas however, in many cases, the Building Regulations do not incorporate BIS codes. Th e lack of knowledge regarding seismically safe construction among architects and engineers, as well as the lack of awareness among the population regarding their own vulnerability has led to most of the construction in the urban and suburban areas being done without reference to BIS standards. In rural areas also, the bulk of the housing is non-engineered construction. In the recent past, due to various factors, the mode of construction in rural areas has changed from the use of traditional indigenous material to brick and concrete construction, 248 Ú Sinha, Satendra and Srivastava

thereby increasing their vulnerability. Moreover, demographic expansion has led to the emergence of settlements in vulnerable areas, close to the riverbed, for instance, which are prone to soil liquefaction. Keeping these aspects in mind the GoI has taken many decisions and is now in the process of implementing them. To deal with the problem eff ectively, a National Core Group for Earthquake Mitigation has been constituted, consisting of experts in earthquake engineering and government administrators. Th e objective of this group is to draw up a strategy and plan of action to mitigate the impact of earthquakes and to provide advice and guidance to States on various aspects of earthquake mitigation. In addition to this, a national programme for capacity building for earthquake mitigation has been fi nalised to impart training to engineers. Since earthquake engineering is not part of the curriculum in engineering colleges at the undergraduate level at present, it has been proposed to select some leading engineering colleges in each State and train the faculty members of their civil engineering departments in earthquake engineering at the Indian Institute of Technology and other institutes. Th ese faculty members will take up the training of municipal engineers as well as the training of engineers and architects in the private sector in RCC and masonry construction. An earthquake mitigation project has also been fi nalised and is being implemented. It includes a detailed evaluation and retrofi tting of lifeline buildings such as hospitals, schools, water and power supply units, telecommunication buildings, airports, airport control towers, railway stations, bus stands and important administrative buildings. Th e programme also includes the training of masons in earthquake- resistant construction. Th e Indian coastline, measuring approximately 8,000 kilometres, is highly susceptible to cyclones. For this reason a cyclone mitigation project has been formulated, which includes components on the strengthening Downloaded by [University of Defence] at 01:15 24 May 2016 of monitoring and warning systems, coastal shelterbelt and mangrove plantations, the construction of cyclone shelters, storm surge modelling and water envelope studies. Th e focus will be on regenerating coastal shelterbelt and mangrove plantations wherever these have degener- ated. Th e cyclone shelters will be multipurpose units to be run as schools or community centres in normal times. In recent years many international organisations working in the fi eld of disaster management have been coming forward to co-ordinate and Disaster Management Policy and Initiatives in India Ú 249

co-operate with governmental eff orts in disaster mitigation, prevention and preparedness. Th e DRM Programme that has been taken up, with assistance from UNDP in 169 of the most hazard-prone districts in 17 States of the country, aims to contribute to the social and economic development goals of national and State governments — enabling them to minimise losses to development gains and reduce vulnerability to natural disasters. Th e programme relies on a community-based approach to disaster management and seeks to build the capacities of communities, government functionaries (at all levels) and all other stakeholders involved in the process in an organised manner. Th e programme components include awareness generation and public education; preparedness, planning and capacity building; developing appropriate policies and plans; institutional, administrative, legal and techno-legal regimes at the State, district, block, village, urban local body and ward levels for vulnerability reduction. Under this programme State and district disaster management plans are being prepared. Th e constitution and training of DMTs and DMCs at all levels, with adequate representation of women, is another important component of the DRM Programme. Giving due consideration to gender issues, special programmes like training women in fi rst aid, shelter management, water and sanitation, rescue and evacuation, etc., have also been included in the programme. Capacity building through training is also considered an integral part of this programme. Elected representatives of PRIs, members of voluntary organisations, government offi cials, engineers, masons and architects, all have been trained in disaster mitigation and preparedness at diff erent levels. Under the programme, DMCs, consisting of elected representatives, civil society members, civil defense volunteers and government functionaries, have been constituted at all administrative levels, including village, urban local body, and ward levels. Th e Urban Earthquake Vulnerability Reduction Programme (UEVRP), a sub-component of the DRM Programme, is also being implemented Downloaded by [University of Defence] at 01:15 24 May 2016 with international support. It aims to strengthen the capacities of communities, urban local bodies and the administration in the realms of mitigation, preparedness and response in 38 Indian cities. Th ese cities have been chosen on the criteria of being located in Seismic Zones 3, 4 or 5, with a population of more than 0.5 million. Th e project will demonstrate a suitable model for the mainstreaming of earthquake risk management initiatives at all levels and help reduce risk in the most earthquake-prone areas in urban India. 250 Ú Sinha, Satendra and Srivastava

Human Resource Development and Capacity Building

Keeping in mind the strategic signifi cance of human resource development (HRD) in disaster mitigation, prevention and preparedness, the GoI has taken many initiatives. Th e most signifi cant being the re-designation of the National Centre for Disaster Management as the National Institute of Disaster Management (NIDM). Th e institute is being developed as a Regional Centre of Excellence in Asia for imparting training, HRD and capacity building, documentation, research and development in the fi eld of disaster management. NIDM is involved in developing training modules at diff erent levels, undertaking the training of to-be trainers and organising training programmes for planners, administrators and command functionaries. The other functions assigned to the NIDM include the development of an exhaustive national level information database on disaster management policies, prevention mechanisms, and mitigation measures; formulation of a disaster management code and providing consultancy to various States in strengthening their disaster management systems and capacities as well as the preparation of disaster management plans and strategies for hazard mitigation and disaster response. At the State level also necessary steps are being taken for capacity building of the various stakeholders. Disaster management faculties have already been created in most State-level training institutes. Th ese are expected to take up several focused training programmes for diff erent target groups, including both the government and non-governmental organisations. NIDM at the national level also provides the support necessary to strengthen State-level institutes in their capacity building eff orts. Generating awareness among local communities is of strategic importance in disaster mitigation. Th e Government of India has constituted

Downloaded by [University of Defence] at 01:15 24 May 2016 a committee for this purpose, which will chalk out an appropriate strategy using diﬀ erent medium of communication. Co-operation of local, grass-roots agencies like NGOs and panchayats has also been solicited for this purpose. Two focused campaigns on earthquakes and cyclones have been initiated in the electronic and print media at the national as well as State level; these campaigns have generated a lot of interest within the community and among other stakeholders. Creating awareness through education is similarly being accorded high priority for mitigating future disasters. Under this, disaster management Disaster Management Policy and Initiatives in India Ú 251

has been introduced in the school curriculum as a subject in social science. Th e Central Board of Secondary Education, which has introduced it in the curriculum, runs a very large number of schools throughout the country and its course curriculum is invariably followed by the State boards. During the Gujarat earthquake, the main fl aw observed in the system was the lack of trained and well-equipped search-and-rescue teams. Th is resulted in much loss of life, which could otherwise have been avoided. Th e government has therefore operationalised the process of training and equipping search-and-rescue teams. Each team will consist of trained personnel like doctors, paramedics and structural engineers. Th ese teams will be located at various centres around the country to ensure a quick response. Th ese teams will have the latest equipment and trained dog squads to locate survivors during such an emergency. A number of such teams have already been trained. Apart from specialist search- and-rescue units, it has been decided that personnel of the central police also be imparted this kind of training so that they too can be requisitioned without delay. Before the arrival of specialist teams, police battalions located near the site of the disaster could thus be deployed immediately. Th e States have also been advised to set up their own specialist disaster response teams. Central assistance will be provided to the State governments for training their personnel at national institutes already designated for the purpose. In order to foster a culture of preparedness and to reach out to stakeholders at the cutting edge level, 23 scenario building exercises in the form of ‘table top’ and mock exercises have been planned and carried out throughout the country for various types of disasters — earthquakes, cyclones, fl oods, fi re and chemical (industrial) disasters. Eight table top exercises and eight mock exercises have already been carried out. Th ese exercises have generated tremendous interest and areas of emphasis for improving response to various hazards in a holistic manner have

Downloaded by [University of Defence] at 01:15 24 May 2016 also been identifi ed. During the Gujarat (2001) and Orissa (1999) disasters, medical response was delayed and this aggravated the suff erings of the disaster- aff ected population. Th e government therefore initiatedsteps to ensure a timely medical response. It is planning to set up mobile hospitals, equipped with all manner of medical and emergency equipment, and which will be located in diff erent parts of the country. Also, for a quick and timely response in an emergency situation a databank of resources, both material and personnel related, is essential. In this regard, a web-enabled 252 Ú Sinha, Satendra and Srivastava

centralised database for the India Disaster Resource Network has been operationalised. Th e network will ensure quick access to resources so as to minimise response time in emergencies. Th e list of resources to be updated in the system has been fi nalised. Th e system will provide, at the touch of a button, the location of specifi c equipments and specialist resources as well as the controlling authority for that resource so that it can be mobilised in the shortest possible time. To help the States in capacity development, running mock drills and facilitating the provision of central assistance, the Indian government is in the process of establishing National Disaster Mitigation Resource Centres (NDMRCs). Th ese centres, to be jointly located with NDRF battalions and set up in nine locations on a pilot basis, will have a reserve of relief stores for 325,000 persons for use in the event of a disaster of severe magnitude. Th ese will include stores for 100,000 persons aff ected in high altitude areas. Th ese reserves are intended to augment the resources of the States in case of major disasters and will be maintained by the mitigation resource centres. For eff ective monitoring and control of a disaster situation, a well- equipped EOC with access to the database and an updated communications system is essential. States are being advised to set up such EOCs at the State and district levels. Assistance for construction and the purchase of equipment for control rooms is being provided by the central government under the DRM Programme. While managing an emergency situation, access to information in terms of the crucial physical parameters for disaster-aff ected areas is of strategic signifi cance. A GIS database is an eff ective and important tool for emergency responders. GIS data is also very helpful in demarcating hazard zones and drawing up maps of vulnerable areas. India is therefore conceptualising a project to institutionalise arrangements to prepare a rich database of such maps, which will provide the required district-wise information. Downloaded by [University of Defence] at 01:15 24 May 2016 Collaboration with the International Community

India is committed to taking all the steps necessary to achieve the goals and objectives of the IDNDR. It has hosted a number of regional conferences on the subject; World Disaster Reduction Day is observed on the second Wednesday of October every year to create public awareness about disasters and the need to adopt preparatory measures. India is Disaster Management Policy and Initiatives in India Ú 253

also a member of the Asian Disaster Reduction Centre, Kobe, Japan. Th e government’s UNDP-assisted NDRM Programme, formulated within the above framework, aims at reducing the vulnerabilities of communities at risk in some of the most (multi-) hazard-prone areas of India. Co-operation between the US and India in the fi eld of disaster management has grown with the launch of a fi ve-year combined eff ort of the United States Agency for International Development (USAID) and the Ministry of Home Aff airs. Th is programme draws in new US partners to share American experiences and expertise on how to mitigate disasters. Th e Indian government believes that co-operation among the various countries, particularly in the Asian region, is essential for achieving eff ective disaster reduction and preparedness. Th ere is consequently a need to have a continuous exchange of experience and know-how. As a part of this exercise, India has also collaborated with other countries of the South Asian region. India with its talented and skilled human resources, information technology expertise and its well-equipped multi-disciplinary research institutes can play a leading role in the fi eld of disaster management, particularly in the South-East Asian region. In this context, a South Asian Association of Regional Co-operation (SAARC) Disaster Management Centre (SDMC) has been set up in New Delhi. Th e centre has a mandate to serve the eight member countries of SAARC — Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka — by providing policy advice and facilitating capacity-building services, including strategic learning, research, training, system development and exchange of information for eff ective disaster risk reduction and management in South Asia. Th e SDMC is a body of professionals working on various dimensions of disaster risk reduction and management in South Asia. Th e centre is networking through the National Focal Points of the member countries with the various ministries, departments and scientific,

Downloaded by [University of Defence] at 01:15 24 May 2016 technical, research and academic institutions within and outside the government working on various aspects of disaster risk reduction and management. Th e centre also conducts studies and research, organises workshops and training programmes, publishes reports and documents, and provides various policy advisory services to the member countries. It has the vision to be recognised as a centre of excellence for knowledge, research and capacity building on disaster management in South Asia and in the rest of the world. 254 Ú Sinha, Satendra and Srivastava

Towards a Safer India

Th e preceding paragraphs discuss some of the policy interventions and initiatives that have been taken in the fi eld of disaster management. With these initiatives, the dream of a safer, more resilient and less disaster-prone India, as envisaged by the HPC, is slowly but steadily being realised. It is hoped that, in time, a new institutional framework will be fi rmly established which can reduce the impact of disasters by focusing on preparedness as well as mitigation. The thinking has gradually gained ground that timely prevention is far more cost-eff ective than post-disaster relief and rehabilitation. Th is transition from an entirely relief-oriented mechanism to a holistic, co-ordinated and comprehensive disaster management system will help in building a safer and more secure India that is capable of resisting both as well as those that are unknown, unforeseen and unexpected. Downloaded by [University of Defence] at 01:15 24 May 2016 11 Seismic Retrofi tting of Existing Non-Engineered Load-Bearing Structures Rajendra Desai

India is a country that is frequented by disasters almost every year, in one part or another. It witnesses much damage, destruction, injury, and loss of life. A resource-starved nation like India loses a lot of scarce and valuable resources on post-disaster rehabilitation. Table 11.1 shows the impact of recent earthquakes in the country.

Table 11.1: Recent Earthquakes and their Impact

Houses severely Magnitude Villages damaged or Houses No. of (Richter aﬀ ected Cities destroyed damaged deaths Earthquake Year scale) (approx.) aﬀ ected (approx.) (approx.) (approx.) Uttarkashi 1991 6.5 400 None — — 1,200 Latur 1993 6.4 1,200 None 35,000 200,000 9,400 Jabalpur 1997 6.1 45 1 5,600 57,000 45 Chamoli 1999 6.8 4,175 None 19,300 86,000 135 Kachchh 2001 7.7 6 150,000 750,000 14,000 Source: Author.

Similar damage and destruction is caused by cyclones that regularly hit the coastal parts of the country. What these numbers do not reveal is that most of the aﬀ ected structures are of the non-engineered load- bearing type. Downloaded by [University of Defence] at 01:15 24 May 2016

Non-Engineered Structures

In a developing country like India an overwhelming majority of houses, as also infrastructure, in towns and villages, is of the non-engineered load-bearing type. Th ese structures, especially houses, built over the past century or more, use locally available materials and locally practiced traditional technologies that are most common in the area. Th e materials 256 Ú Rajendra Desai

used include stone, bricks, earth, lime, and timber for walls, and clay tiles, stone or mud for roofi ng, supported on an under-structure made of local timber such as Teak, Acacia, Neem, Deodar, Pine and Bamboo (see Figures 11.1–11.5). In recently built structures one can fi nd a mix of traditional and new materials/technologies being used, such as cement, concrete and steel. Th e structures have a pitched or fl at roof, and are single-storeyed or double-storeyed. Barring a few exceptions, the majority of these structures are of the load-bearing type. Th ere has been, however, little or no engineering input in the construction of these structures, and the people who build them have no technical understanding of the construction they carry out. Th erefore these structures are called non-engineered. In recent times, a variety of causes have led to a general degradation in the application of such technologies. As a result, most of these buildings have simply not been built to withstand the force of an earthquake. In addition to traditionally built structures, during the past few decades, there has been a great infl ux of structures, load-bearing or otherwise, made out of predominantly modern building materials, like RCC, in rural as well as

Figure 11.1: A Kachchh Bunga Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Figure 11.2: A Typical Uttaranchal House

Source: Author.

Figure 11.3: A North Gujarat Mud-Th atch House Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. 258 Ú Rajendra Desai

Figure 11.4: A Typical Latur House

Source: Author.

Figure 11.5: A Central Gujarat House Downloaded by [University of Defence] at 01:15 24 May 2016 Source: Author.

urban areas. The post-earthquake damage assessment studies for

Jabalpur (TARU 1997) and Kachchh (TARU 2001) earthquakes have revealed a rather disturbing truth, that even multi-storeyed structures, commonly seen in many urban centres, are sometimes non- engineered or semi-engineered. Th is explains the dismal performance of so-called modern structures in both these earthquakes. Seismic Retroﬁ tting Ú 259

Relationship of Non-Engineered Structures to the Local Context

Th e non-engineered traditional construction commonly practiced in diff erent parts of the country depends a great deal on the specifi c local context of the area. In other words, technologies may vary signifi cantly from one area to another. Th e major local factors that dictate what technology is to be used are:

(i) Biomass availability, which dictates the possibility of a fl at or sloping or curved roof, or the possibility of its use in wall construction. (ii) Soil types, which dictate usability as wall-building material, or as mortar or raw material for bricks. (iii) Type of most easily accessible rock, which dictates its use for walling as well as for roofi ng, in the form of slabs for a fl at roof or tiles for a sloping roof. (iv) Weather patterns, including rainfall, extreme temperatures, wind conditions, etc., which dictate the roof’s slope and over- hang, wall thickness, window openings, verandah depth, etc. (v) Accessibility of the area, which dictates the cost of all the materials that have to be carted to the site. (vi) Th e lifestyle of the people.

A study of Gujarat’s traditional building systems, carried out by the

author (Desai and Desai 1990) observed that there are eight distinct building systems within this one State. Th ese building systems are in tune with the local factors described above. Map 11.1 shows the prevalent building systems in various parts of the State.

Signiﬁ cance of Non-Engineered Construction Downloaded by [University of Defence] at 01:15 24 May 2016 Non-engineered structures such as those described above play a major role in meeting the basic yet crucial needs of society. Th is can be better understood by looking at aspects like cost, climate, lifestyle, and economy.

(i) Cost: Th ese systems are the most aﬀ ordable in the short and long term. Local materials are the most easily accessible and hence cost little to transport. Bricks that cost Rs 900 per 1,000 Downloaded by [University of Defence] at 01:15 24 May 2016

Map 11.1: Traditional Building Systems of Gujarat

KILOMETRES

Raiasthan

Gulf of Ka^hcKb

2:--: . '": ivj]is root no v.-Sid

: no w&A 2-iided 3 rand mud 2-sided Arabian. Sea si stone stone 2-sided Gu.f or Khambat ' ""-'• :: : ->_•!•! - stone iud 2-sided S r. ;• ::-& d 7 mud 7mud eanJilfiat

8 stone stone : - •

Source: Desai and Desai (1990). Seismic Retroﬁ tting Ú 261

around Ahmedabad cost Rs 1,400 or more in eastern Kachchh, and possibly Rs 1,700 in western Kachchh. In Uttaranchal this diff erential is even more acute. Bricks that cost Rs 800 in Rishikesh could cost Rs 3,000 in the hills at the roadside but Rs 6,000 or more away from the road. Th e cost of these systems is lower also because the available local labour is able to execute them. In the long run, these systems are also easier and cheaper to maintain. Th e fl at mud-roof of Latur requires only the simple removal of weeds every monsoon and the spreading of a little new clayey, impervious soil to ensure that it remains waterproof. No special waterproofi ng material or skill is required to be brought in from afar. Comparing on an even longer run, the greatest attribute of these systems is that a good portion of the materials can be recycled at the end of the house’s life. (ii) Climate: Latur houses, with thick stone walls and thick mud roofs, require no fans on hot summer afternoons and no blankets on cold winter nights. In Uttaranchal, houses with a pitched stone roof are distinctly warmer than those having an RCC slab roof. In Kachchh, the bunga having a roof made of Dhrub grass is distinctly cooler in the hot summer than one having a Mangalore pattern clay tile roof. (iii) Economy: Investment in a house built using the traditional system that employs mainly local materials and labour contributes signifi cantly more to the local economy. It also generates much more local employment.

One could say with a fair amount of certainty that the greater part of all building construction will continue to be done in this way in the foreseeable future because of the eternal resource crunch and the country’s huge population. Other technical options have become available in the past few decades and more will come with time. But few will be as

Downloaded by [University of Defence] at 01:15 24 May 2016 viable and as optimised for the rural and semi-urban context as these older options that have been in use for all these years.

Large Scale of Vulnerability

A study conducted by Prof A.S. Arya (2003), based on the Vulnerability Atlas of India (1997), shows that in Seismic Zone V alone there are over 11 million housing units that are vulnerable to an earthquake. 262 Ú Rajendra Desai

Similarly, in Zone IV this number could be as much as 50 million. By another count (Bose and Verma 2003) there are over 80 million houses in the country that are vulnerable. Th ese counts, however, do not include non-engineered infrastructure buildings in towns and villages as also those built with modern technologies but without any engineering input. In other words, the number of structures that are vulnerable is extremely large, and the people at risk may amount to as much as 50 per cent of the country’s population.

Performance by Non-Engineered Structures in Recent Disasters

Th e performance of non-engineered structures, built using a variety of building systems and materials, has been studied extensively in the past decade. Th ese include, among others, the following types of construction.

Table 11.2: Building Systems and Materials of Structures Aﬀ ected by Recent Earthquakes

No. Type Walling Roofi ng Region 1 Non-load- Random rubble Flat mud roof on a Latur bearing timber frame 2 Load-bearing Random rubble Corrugated Galvanised Latur Iron (CGI) Sheeting 3 Load-bearing Random rubble Pitched slate; fl at RCC Uttaranchal slab 4 Load-bearing Random rubble Pitched clay tile; fl at Kachchh, RCC slab Saurashtra 5 Load-bearing Bela or Shaped Pitched clay tile; fl at Kachchh, stone RCC slab Saurashtra 6 Load-bearing Concrete block Pitched clay tile; fl at Kachchh, RCC slab Saurashtra 7 Load-bearing Burnt brick Pitched clay tile; fl at Jabalpur, Downloaded by [University of Defence] at 01:15 24 May 2016 RCC slab Saurashtra 8 Load-bearing Mud (Adobe Pitched clay tile Jabalpur, and Cob) Kachchh Source: Compiled by the author.

All these types of buildings, which were built without taking into account the possibility of earthquakes in the area, performed badly (see Figures 11.6–11.9). Figure 11.6: Delamination of Random Rubble Masonry Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Figure 11.7: Racking Shear Cracks

Source: Author.

Figure 11.8: Delamination of Gable Wall Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Seismic Retroﬁ tting Ú 265

Figure 11.9: Gable Wall Failure

Source: Author.

Th e reasons for their bad performance are well known. Th e fi rst reason was the poor quality of construction, which refl ects the violation of the basic rules of construction (see Appendix A) and the second reason was the absence of disaster resistant features (see Appendix B). Even if a structure were to be built ensuring good quality, it is unlikely to withstand a strong earthquake if special features needed to overcome the inherent weaknesses of a non-engineered structure are not incorporated in the construction. It should also be noted that, as mentioned earlier, in Jabalpur and Gujarat, many structures having an RCC frame of some sort also suff ered signifi cant damage. Th is poor performance resulted in massive loss of property, the extent of which can be quickly gauged by translating it into monetary terms (approximate). Th ese would be purely indicative, simply to give some indication of the losses that may be experienced in a future earthquake (Table 11.3). Given these fi gures, clearly it is important to undertake some sort of Downloaded by [University of Defence] at 01:15 24 May 2016 corrective action that will help reduce the vulnerability of several million such structures.

Table 11.3: Indicative Losses of Property (buildings) in Past Earthquakes

Houses Public buildings Total loss No. Earthquake (in US$) (in US$) (in US$) 1. Latur 375,000,000 700,000 375,700,000 2. Chamoli 900,000 1,424,000 2,324,000 3. Kachchh 585,000,000 19,500,000 604,500,000 Source: Author. 266 Ú Rajendra Desai

The Concept of Retroﬁ tting

With a large number of structures being non-engineered and therefore vulnerable, it is important that this vulnerability be tackled in order to reduce the undue hardship and losses that the people will suff er and the economic losses that the country will face in the event of another major earthquake. Th is can be achieved by retrofi tting these structures. A number of tests have been carried out in India to study the eff ect of retrofi tting on a variety of traditional structures. Th ese include those carried out by Prof Arya at Umerga near Latur in 1995, by the author (Desai 1999) at Latur in 1997, and by Prof Arya and the author at Radhanpur (bordering Kachchh) in 2002, as well as those carried out at the Indian Institute of Technology (IIT), Roorkee by Pankaj Agrawal and S.K. Th akkar in 2003. These tests eff ectively demonstrated that the retrofi tting of non-engineered structures with simple elements like a Ferrocement (FC) belt installed at lintel or eaves level, vertical single reinforcing bars in the corners, diagonal bracing in the roof, reduction in roof dead load, etc., signifi cantly increases their resistance to seismic forces. Retrofi tting simply means ‘altering to fi t the context’. It is a time- tested concept that is commonly used in our day-to-day lives. One retrofi ts or strengthens an old worn-out cardboard carton with the help of adhesive tape along the corners or with a string wrapped around it so that it does not fall apart when loaded with heavy things. While retrofi tting one also repairs all the old damage because that too constitutes a weakness. Similarly, a house that is not built to withstand a disaster that it is likely to face in its lifetime, can be retrofi tted to withstand the forces that may be exerted on it in case a disaster strikes. Th e extent of retrofi tting will depend upon (i) the probability and magnitude of the expected forces as well as (ii) the vulnerability of the structure under consideration. In Seismic Zone V, as in Kachchh or the hills of Uttaranchal State, one

Downloaded by [University of Defence] at 01:15 24 May 2016 would have to take greater care than in Seismic Zone III, as in Surat area of Gujarat or the Latur region in Maharashtra. Similarly, in coastal Kachchh and the Saurashtra regions of Gujarat one would have to worry about cyclones more as compared to locations inland. Th e structures that are very poorly built are more vulnerable than those that are well built. Structures that have been damaged are also more vulnerable than those that have not been damaged. Th ese more vulnerable structures need greater improvement than a structure that is already in good shape. In other words, structures that have a high degree of in-built vulnerability must be retroﬁ tted at the earliest. Seismic Retroﬁ tting Ú 267

Retroﬁ tting Measures

For each weakness that has been observed in an earthquake there is a remedy in the form of a retrofi tting measure. Th ese remedies are evolved on the basis of the information available in a number of books (ISET 1989; BMTPC 1999, 2000) and Indicative Specifi cations (IS) Guidelines (Improving Earthquake Resistance of Low-Strength Masonry Buildings [IS: 13828: 1993]; Improving Earthquake Resistance of Earthen Buildings [IS: 13827: 1993]; Repair and Seismic Strengthen- ing of Buildings [IS: 13935: 1993]; Earthquake Resistant Design and Construction of Buildings — Code of Practise [IS: 4326: 1993]) in India. Many of these measures are described in Table 11.4. Th e IS for some of the retrofi tting measures mentioned above are given in Appendix C. Th ese specifi cations have been fi ne-tuned based on retrofi tting work that has been carried out by the author on nearly a thousand structures in various parts of the country, including Latur, Uttaranchal, Kachchh, Saurashtra and North Gujarat.

Principal Steps Involved in Retroﬁ tting

To an experienced engineer the retroﬁ tting of a non-engineered structure may appear to be a simple mechanical process of applying various measures. However, a simplistic approach will invite mistakes that might fail to tackle the vulnerability of the structure. Th e process of retroﬁ tting involves several steps. Th ese can be listed as follows:

(i) Prepare a Graphical Document of the structure to thoroughly understand the existing construction, especially the ‘what’ and the ‘why’ aspects of it, and also study the stress path created by the seismic forces. Downloaded by [University of Defence] at 01:15 24 May 2016 (ii) Assess the weaknesses of the structure, including those resulting from past damage as well as the structure’s age, and assess the impact on the structure of the most powerful earthquake likely to occur in the area. (iii) Using certain rules of thumb evolve a retrofi tting system to counter a likely earthquake impact consisting of various measures for the structure as a whole. (iv) Show all the measures on the Graphical Document. (v) Estimate the cost of the application of each one of these measures. 268 Ú Rajendra Desai erent levels erent tting measure tting Retrofi junction securely connected with both the walls coupled with tie rods Gable FC belt and Guy Lines Install cast in-situ RCC Bond Elements Vertical Rod or FC Strap to connect walls with the roof Same as above, plus a Vertical Rod or FC Strap at wall Same as above FC belt around the structure at diff tting Measures tting Cause Walling wall “through stones” lateral shear connection in-plane shear lateral shear Absence or shortage of Horizontal bending due to Table 11.4: Retrofi Weakness bulging Bearing failureDelamination or Absence of bearing block Introduce Bearing Block of cast in-situ RCC or timber Gable wall collapse Unrestrained free-standing Horizontal cracks Vertical bending due to Cracks along corner Poor wall-to-wall Diagonal cracks Pulling due to racking or Downloaded by [University of Defence] at 01:15 24 May 2016 (UCR) masonry walls 6. 7.Rubble Uncoursed 5. 4. 3. No. Building element 2. 1. Masonry Walls Vertical cracks Seismic Retrofi tting Ú 269 slab angle diagonals Concrete (RC) screed pipe at timber column to beam connections eaves to clamp down tiles and install load walls on top of each gable wall diagonal ties of timber or Mild Steel (MS) elements, or pre-tensioned Galvanised Iron (GI) wire cables understructure elements Vertical reinforcement in walls to connect with Creating diaphragm with MS angle struts and Creating diaphragm with 35 mm-thick Reinforced Install knee-braces made of timber, MS angles or steel Install steel bar or GI wire cable along tile top at roof Creating in-plane diaphragm with timber struts and Strengthen connections between various ng Roofi ness with wall elements and absence of diagonal ties elements and absence of diagonal ties connection is not adequately rigid tiles stiff between members Roof beam to column Absence of anchoring for Inadequate connections at of roof frame eaves separation Roof distortion Absence of in-plane Downloaded by [University of Defence] at 01:15 24 May 2016 Tile (MPT) pitched roof 12. oor Stone on joists fl 13. Separation of elements Stone on joists roof No connection between 14. Separation of elements Timber frame for roof No connection between Tilting or total collapse Source : Compiled by the author. 11. RCC Slab Sliding on top of wall Absence of connection 9.10.member Roof Tiles falling off 8.Pattern Mangalore 270 Ú Rajendra Desai

(vi) Decide the budget. (vii) Decide on the sequence and mix (all or a few) of the measures and their extent (whole house or portions of it) at the given time, based on the budget as well as the convenience of the house owner. ( viii) Identify suitable labour, including skilled artisans. (ix) Procure the necessary materials. (x) Execute the job, keeping a close tab on critical aspects of each measure to get the best results.

Cost

Th e cost of retrofi tting a structure depends upon the kind of corrective measures that need to be taken. Th e structures that are more vulnerable require a larger number of retrofi tting elements. Th us, a slate roof house in Uttaranchal needs a variety of retrofi tting elements for the roof as well as the walls. On the other hand, a typical brick-wall structure with an RCC slab requires fewer retrofi tting elements. Th is diff erence gets refl ected in the cost. Similarly, in case of a higher seismic zone more elements are required since a given structure would be more vulnerable in a higher zone as compared to one in a lower zone. Vulnerability also depends upon the degree of damage in a given structure. Th ese relationships are demonstrated in a study carried out by TARU as shown in Table 11.5.

Table 11.5: Retroﬁ tting Rural Dwellings in North-East Himalayas — Unit Cost (Rs/sq. m.) and Retroﬁ tting Index1

Building Types: Material of Damage Grade Roof and Wall G1 G2 G3 G4 Slate/Earth — 176 (8%) 205 (15%) Slate/Stone 353 (21%) 383 (21%) 400 (21%) 551 (32%) RCC/Stone 158 (7%) 327 (15%) 348 (15%) 598 (31%) Downloaded by [University of Defence] at 01:15 24 May 2016 RCC/CB (Concrete Block) — 20 (1%) 210 (8%) 298 (10%) CGI/Stone — — — 678 (23%) CGI/CB — 394 (18%) — — RCC/Brick — — 132 (4%) — Source: TARU (1999).

1 Retrofi tting Index is the cost of retrofi tting compared to the cost of new construction. Seismic Retrofi tting Ú 271

Similar trends have been observed in other regions of the country where retroﬁ tting work has been carried out.

The Economics and Advantages of Retroﬁ tting

Th e economics of retrofi tting are self evident, as can be seen from Table 11.5 which shows that the retrofi tting index is as low as 5 per cent. For retrofi tting to be most advantageous it must be done in a proactive manner. In other words, it is best if done before a disaster strikes. Th is will help save lives and losses, and prevent much suff ering. Th us far in India the approach to retrofi tting, whatever little that has been carried out, has been retroactive. Past disasters show that retrofi tting has not been a priority for most concerned parties. It has been observed that many house owners, donor agencies, and at times even government agencies, want to demolish structures that are only a little damaged or even absolutely undamaged. But that proposition has been expensive, inconvenient and time consuming. Instead, the option of repair and retrofi tting is faster, more economical, and helps retain the conveniences that existed in the earlier structure, be it a house or an infrastructure building. It is indeed the most economical option that one could exercise for ensuring one’s long-term safety. Th e option of building anew, although eff ective, involves the cost of dismantling the existing structure, carting off debris and, fi nally, the cost of reconstruction. With rapidly escalating costs the expenditure involved may be several times that of what was spent when the structure was fi rst built. On the other hand, retrofi tting costs only a tiny fraction of this. Yet another important advantage of this option is its innate fl exibility. Of the several diff erent measures that need to be applied one has a choice of using one or a few measures at a time to a part of the structure, or one or a few measures to the whole structure, or applying all the options to the whole structure in one go. Th us one could take a decision based on the availability of funds, the availability

Downloaded by [University of Defence] at 01:15 24 May 2016 of time, and ﬁ nally, convenience, while ensuring the technical soundness of the overall scheme.

Hurdles in Applying Retroﬁ tting Measures to Non-Engineered Structures

Th ere are several hurdles in the large-scale application of retroﬁ tting measures in the country, be it in urban areas or rural. Th ese can be listed as follows: 272 Ú Rajendra Desai

(i) Few engineers understand the technicalities of the variety of load-bearing systems used in the country. Most of them consider them unscientific, archaic, unreliable, unsafe, out of fashion, and not modern. As a result, in the event of making a choice between preserving a load-bearing structure through repair and retrofi tting versus building it anew, the choice falls on the latter. (ii) An engineer’s decision to recommend retrofi tting, which is essentially a low-cost option, is dictated at times by its economic attractiveness to the engineer whose advice is sought. (iii) In the aftermath of a disaster, the over-abundance of funds tends to result in the exclusion of this option in favour of the more expensive option of dismantling and reconstruction. (iv) House owners, in general, are ignorant about retrofi tting and they lack confi dence in it. Hence, they do not readily opt for it and do not invest their own funds. (v) Building artisans and contractors do not possess the necessary skills and knowledge, nor do they understand the economics of it. (vi) Th ere have been training programmes for masons and engineers in the past in many areas of the country. Most of them have been classroom programmes. It is vital that training be carried out in the fi eld. (vii) Most training programmes have been ‘one shot deals’. But this is simply not enough to begin work on one’s own. Much follow-up and refresher training is needed to prevent mistakes and the false sense of security that such wrong work may create. ( viii) Necessary materials, such as welded wire mesh (galvanised or otherwise), 16 gauge GI wire, low shrink grout, etc., are not available in most places. Downloaded by [University of Defence] at 01:15 24 May 2016

In other words, the delivery system for retroﬁ tting simply does not exist and needs to be created.

Need of the Hour

It is recognised around the world that retroﬁ tting existing buildings is the most viable option for ensuring the long-term safety of the maximum Seismic Retroﬁ tting Ú 273

number of people against future disasters. Since in developing countries like India the bulk of the houses and infrastructure buildings are non- engineered and are executed in the informal sector, promotion of this option must be aimed primarily towards the people at large, including building artisans. Th e technical community, however, does have a marked infl uence on the psyche of the people and on artisans. Hence, eff orts at promotion must also focus on the technical community. Th e following is a list of things that ought to be taken up for the large scale promotion of the option of retrofi tting: Education of Engineers Since the engineering education curriculum ignores vernacular systems of construction, engineers coming out of these colleges are completely ignorant about them. Load-bearing building systems are also absent from the curriculum of most colleges. On the other hand, technical education over-emphasises RCC and steel construction. As a result, for most engineers these are the only legitimate construction systems worth using. Everything else is useless and must be discarded. It is this attitude that needs to be changed to initiate a process of improvement in the construction scene at the grass-roots. Th is will help people cultivate pride in their own construction systems and, at the same time, welcome the improvements that suitably trained engineers may be able to suggest to make their homes disaster-resistant, rather than opting for alien systems that cannot be sustained by them. Practicing engineers also need to be exposed to vernacular systems and their relevance in modern India, as well as to their weaknesses and strengths, and ways to retrofi t such structures. Special fi eld visits for orientation may also be organised. Such programmes may be organised through local associations of engineers. Videos of shake-table tests could be very eff ective in instilling confi dence in the retrofi tting of such structures. Downloaded by [University of Defence] at 01:15 24 May 2016 In 2005, the author had the shake-table from Kachchh dismantled and shifted permanently to Nirma University, Ahmedabad, so that it became accessible to more people. In the new set-up, the tractor was replaced by a 1.5 tonne pendulum of 2 m height. Th e pendulum can be raised through diff erent angles and released to give impacts of diff ering intensities to the platform. Th e issues that were investigated through the tests included the use of multiple bands as opposed to a single band, the use of RC columns (without RC beams) in the corners (as is commonly 274 Ú Rajendra Desai

practiced today) instead of single vertical rods within masonry, and the use of small diameter, vertical confi nement reinforcement versus no vertical reinforcement. Th e low-cost option of confi nement reinforcement on both faces of masonry walls was found to be eff ective in preventing the collapse of the structure. Multiple bands were clearly more eff ective than a single, lintel-level band. Th ese tests also conveyed

Figure 11.10: Shake-Table Demonstration — Conﬁ nement Reinforcement

Source: Author. Note: Model with conﬁ nement reinforcement (on the left) remains standing, while one built in the traditional way (on the right) has collapsed.

Figure 11.11: Shake-Table Demonstration — Single RC Band Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Note: Model with a single RC band in a state of impending collapse (on the right) while one with multiple bands (on the left) remains standing. Seismic Retroﬁ tting Ú 275

that corresponding retrofi tting measures would be eff ective in improving the performance of the existing structures. Imparting Visibility Through demonstrations on actual buildings in various locations retrofi tting options can be popularised. Th e participation of known experts in the fi eld and the use of media would greatly help increase the eff ectiveness as well as the reach of such demonstrations. Th is must be accompanied with a public awareness programme. Public Awareness Campaigns In order to create a felt-need among the people, creating public awareness is a crucial step in evolving the delivery system for retrofi tting. In the past, post-disaster rehabilitation programmes have shown that unless people feel the need for safety, that too through the option of retrofi tting, no amount of training and other programmes will ensure that existing structures will be retrofi tted. Th is could be done through a variety of means such as TV talk shows, workshops, the publication and distribution of pamphlets, making available simple handbooks, house owner’s guides, etc., at a very low cost in bookshops or at the building supplies stores, for example. Once people have a felt-need for safety they will invest their own funds in retrofi tting their buildings, provided they feel sure that it will work. In other words they must have confi dence in retrofi tting. Confi dence Building Programmes Th ese could be organised through fi eld shake-table programmes in various parts of the country, and by making and airing of videos of shake- table demonstrations and other relevant issues. Th ese programmes could be tailormade for diff erent audiences such as those from the construction industry, the common man, policy-makers, children, etc. Downloaded by [University of Defence] at 01:15 24 May 2016 Training of Petty Contractors and Building Artisans Th is forms the most important link in the whole delivery system. After all, for the common man, the mason and the petty contractor are the ‘engineers’ who help him build his house. Th is is the case not just in villages but also in mid-size towns. An overwhelming proportion of house construction in the country is carried out by them. Intensive hands-on training of this group of individuals is necessary so that they are in a 276 Ú Rajendra Desai

position to plan and carry out the retrofi tting of simple buildings that are built by them in the fi rst place. One-time training is not adequate; it has to be a number of days and followed up by refresher training as well as monitoring of what they actually do in the fi eld. An important point that needs to be recognised is that these types of training programmes are successful only if potential participants feel that they stand to gain from them. Th is is possible only if there is a demand for such expertise. Clear and Positive Government Policies Th e government invests the maximum amount of money as compared to anyone else in construction, as well as in post-disaster relief and rehabilitation work. Hence, the government must be concerned about making sure that its existing buildings do not suff er much damage in the event of an earthquake. It should also be concerned that the other buildings in which people live and work do not collapse. Th is would result in saving a lot of funds and other resources. Th erefore, it must have a policy that promotes retrofi tting. Th is can be done through the following means:

(i) Making it compulsory to retrofi t all public buildings. (ii) Ensuring that only suitably trained contractors and masons work on such projects. (iii) Encouraging people to retrofit their houses by offering them cheap insurance for the retrofi tted houses. (iv) Providing low-interest loans for retrofi tting. (v) Setting up retrofi tting centres where a house owner could get all the necessary, guidance and support. (vi) Commissioning studies of vernacular systems prevalent in the country and identifying retrofi tting measures. (vii) Preparing simple house owner’s guides to retrofitting,

Downloaded by [University of Defence] at 01:15 24 May 2016 manuals for engineers, contractors and building artisans, etc., for seismically active parts of the country in the local language. ( viii) Encouraging Doordarshan and other TV channels to conduct shows for information dissemination. (ix) Help bring relevant international experience to the country.

Th e synergistic eﬀ ect of all these programmes on a sustained basis could bring the desired results in the foreseeable future. Seismic Retroﬁ tting Ú 277

Appendices

Appendix ‘A’

Table 11A: Weaknesses in Non-Engineered Construction

No. Violation Damage Cause Masonry walls 1. Lack of ﬁ lling of vertical Vertical cracks Horizontal bending due joints to lateral shear 2. No breaking of joints Vertical cracks Poor interlocking among between successive masonry units, creation courses of a vertical plane of weakness 3. Walls built one at a time Corner cracks Poor wall-to-wall with toothing left at connection its ends for joint with the adjacent wall 4. Wall (including gable Horizontal cracks Vertical bending due to wall) standing by lateral shear itself without support of adjacent walls 5. No breaking of joints Delamination or Absence of interlocking within each course disintegration of UCR between stones of inner along the length of wall walls and outer faces of wall 6. Inadequate or absence Delamination or Absence of stitching of through stones bulging in UCR walls between the two faces of a wall All types of walls 7. Total opening width Diagonal cracks Pulling due to racking or too large in-plane shear 8. Opening too close to Diagonal cracks Pulling due to racking or corner in-plane shear 9. Absence of bearing block Bearing failure Absence of bearing block under concentrated loads from beams Downloaded by [University of Defence] at 01:15 24 May 2016 Use of materials 10. Improper cement Poor masonry strength Poor quality cement mortar mix and/or sand, use of less cement 11. Use of mortar after Poor masonry strength Initial strength lost due to initial setting disturbance to mortar

(Continued) 278 Ú Rajendra Desai

(Continued) No. Violation Damage Cause 12. Inadequate wetting of Micro-cracks in joinery Shrinkage of mortar masonry units, namely due to loss of water bricks, concrete blocks, in mortar due to stone at the time of absorption into construction masonry unit 13. Inadequate curing Drop in masonry Incomplete hydration of strength cement Roof 14. Inadequate connections Displacement of In-plane shear results in between roof members rooﬁ ng elements relative movement between various elements Source: Compiled by the author.

Appendix ‘B’

Table 11B: Earthquake Damage Resulting from Absence of Disaster Resistant Features

No. Missing special feature Damage type Cause 1. Lintel level or eaves Cracks of all types in Low allowable tensile level band walls strength to counter stresses caused by lateral and in-plane shear 2. Gable band Crack at the gable wall Low allowable tensile base strength to counter stresses caused by lateral shear 3. Vertical corner Horizontal cracks near Low allowable tensile reinforcement wall base strength to counter stresses caused by lateral shear 4. Vertical reinforcement Diagonal cracks in walls Poor bond between roof embedded at top into at roof–wall interface; and walls

Downloaded by [University of Defence] at 01:15 24 May 2016 RCC roof slab lateral slippage of roof slab on top of wall 5. Roof diaphragm in the Severe distortion of Under the impact of lateral form of in-plane pitched roof resulting shear the rectangles diagonal ties and in damage to roof tend to become timber struts in roof members parallelograms 6. Roof diaphragm in the Severe distortion of Absence of restraining form of in-plane pitched roof resulting elements such as roof diagonal ties and in damage to gable diaphragm timber struts in roof wall or its collapse (Continued) Seismic Retrofitting 279

(Continued) No. Missing special feature Damage type Cause 7. Adequate roof member Displacement or blowing Absence of positive con- anchoring onto wall away of roof nection between roof and walls 8. Knee-braces at timber Tilting or total collapse Roof beam to column column to beam of roof frame connection is not connections adequately rigid Source: Compiled by the author.

Figure 11B.1: Knee-Braces for a Timber Frame

Source: Author.

Figure 11B.2: Knee-Braces at Timber Column to Beam Connections Downloaded by [University of Defence] at 01:15 24 May 2016

30x30x3mm MS Angle Knee Bracing

Source: Author. 280 Ú Rajendra Desai

Appendix ‘C’

Suggested Speciﬁ cations

For all major wall types (UCRM, UCRC, UCRL, ASHLR, CB, BBM, BBC, and MUD) and all major roof and ﬂ oor types (pitched MPT, RCC, and stone on joists), all items are to be in accordance with the relevant codes or guidelines published by the concerned authorities. Th e most important consideration in deciding the details of the measures to be taken and their speciﬁ cation is that they should respect local factors. In other words, the local availability of materials and skills, and the overall viability of the option must be looked into. Availability of infrastructure, such as access to the site, options for carting of materials, the availability of electricity (2 and 3 phase), etc., are equally important.

Indicative Speciﬁ cations for Seismic Strengthening

Figure 11C.1: A Retroﬁ tted Non-Engineered Structure

Source: Author. Downloaded by [University of Defence] at 01:15 24 May 2016 Cast In-Situ Bond Element (for UCRM, UCRC and UCRL) Bond elements are to be installed in all UCR walls. Holes shall be made in the walls by removing one stone from both the inner and outer faces of the wall. Th e hole is to be cleaned and wetted. An 8 mm diameter High Strength Deformed (HSD) bar, hooked at both ends, shall then be placed in the hole. Th e hole shall Seismic Retroﬁ tting Ú 281

Figure 11C.2: Cast In-Situ RC Bond Element

Source: Author.

be ﬁ lled with concrete (1:2:4). Th is RCC bond element is to be cured for seven days. Provide a minimum of one element in every 0.8 sq. m. wall area. Ferrocement (FC) or Seismic Belts (for UCRM, UCRC, UCRL, ASHLR, CB, BBM, and BBC) Th e FC belt may be 200 mm wide. It may be made of 150 mm wide, 13 gauge, 25 × 50 mm welded wire mesh (WWM) plus 2–6 mm diameter MS rods fully embedded in 35 mm thick cement mortar (1:3). In Seismic Zones IV and V this is to be applied on both faces of the walls. It may be anchored to the wall at a spacing not exceeding 1.3 m using shear connectors described hereunder. In addition, 6 mm diameter 100 mm-long iron nails with large heads and a washer also may be used for additional connections as deemed necessary. Prior to the installation of the WWM the plaster is to be fully removed along the alignment of the belt and all the masonry joints to be raked to a minimum depth of 12 mm. Th e surface Downloaded by [University of Defence] at 01:15 24 May 2016 is to be cleaned with a wire brush and the dust washed oﬀ with water. Plaster will be applied in two coats. Shear Connector (for UCRM, UCRC, UCRL, ASHLR, CB, BBM and BBC) Th e shear connector is to consist of an 8 mm HSD rod, hooked at one end and L shaped at the other. Th e hooked end should be inserted in a hole measuring 50 mm to 150 mm in diameter, with the bent end projecting out close to the wall Figure 11C.3: Installation of Seismic Belt

Source: Author.

Figure 11C.4: Seismic Belts Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Seismic Retroﬁ tting Ú 283

Figure 11C.5: Seismic Belt Installation

Source: Author.

surface. Th e hole is to be grouted with 1:3 non-shrink grout in case of small diameter holes and with 1:2:4 concrete in case of larger holes which are to be cured. Tie Rod for Ferrocement Belts (for UCRM, UCRC, UCRL, ASHLR, CB, BBM and BBC) When the clear length of the belt from one corner to another is more than 5 m, 12 mm diameter MS bars are to be used as tie rods to restrain the band.

Figure 11C.6: Installing Bond Elements Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. 284 Ú Rajendra Desai

Figure 11C.7: Casting a Bond Element

Source: Author.

Th e space between tie rods not to exceed 3 m. Th e rod is to extend from the belt on one wall to the belt across the room on the opposite wall. Th e tie rod is to be taut, preferably with the help of a turn buckle in the middle or nuts at the ends.

Downloaded by [University of Defence] at 01:15 24 May 2016 In case the length of the tie rod exceeds 3 m it may be advisable to suspend it from one or two points to eliminate sag. Vertical Reinforcement (for UCRM, UCRC, UCRL, ASHLR, CB, BBM, and BBC) Vertical reinforcement shall extend from 300 mm below the fl oor level of the ground fl oor to the roof level. At the bottom it may be anchored in mass concrete or with the foundation wall, if possible. Within a storey’s height it must Seismic Retrofi tting Ú 285

Figure 11C.8: Tie Rod for Seismic Belt

Source: Author.

Figure 11C.9: FC-Belt Mesh Installation Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. 286 Ú Rajendra Desai

be connected at a spacing not exceeding 750 mm to the wall, to the RCC band or to the Ferrocement Belt, whichever is there, and to the reinforcement of the RCC slab. Th e vertical reinforcement may be installed adjacent to the wall-to- wall junction, and on either side of large wall openings. In case of wall length exceeding 3 m it may be provided at a spacing not exceeding 1.5 m. It may be provided in the following diff erent ways: For UCRM, UCRC, UCRL, ASHLR, CB, BBM, and BBC (i) Ferrocement Strap: installed in a fashion identical to that recommended for the Ferrocement Belt, but its alignment will be vertical. (ii) 8-mm HSD Bar: the bar may be installed in wall corners, approximately 50 mm away from both the walls. It is to be encased in micro concrete (1:1.5:3) along its length with all around minimum clear cover of not less than 15 mm. Th e bar is to be fi xed to the wall with the help of shear connectors. For MUD (iii) 16 gauge GI Wires: on both faces of the wall at a spacing not exceeding 300 mm placed in grooves approximately 30 mm deep. Th e wires must be anchored at a vertical spacing not exceeding 450 mm. At each anchor location after drilling a hole through the wall, a 16 gauge GI wire tie is to be installed holding tightly together the wires on the outside and inside faces in order to provide confi nement. Th e wires must be securely tied to the wall plate and to intermediate bands. All wires must be covered in mud plaster. Gable Strengthening — in Pitched Roof Houses Only (for UCRM, UCRC, UCRL, ASHLR, CB, BBM, and BBC) Th is being the most vulnerable part of the structure may be retrofi tted by using two independent items:

(i) Ferrocement Belt of WWM (identical to Seismic Belt), installed along

Downloaded by [University of Defence] at 01:15 24 May 2016 the sloping alignment on the triangular gable wall face just under the gable top. In Seismic Zones IV and V this is to be applied on both faces of the walls. (ii) Securely connect the structural roof members to the Seismic Belt with the help of 2–16 gauge GI wire ties adequately pre-tensioned. (iii) Guy lines made of 4 to 5–16 gauge GI wires to keep the walls from falling out. Th e guy lines are to be anchored into the wall with the help of modiﬁ ed shear connectors anchored securely with a large bearing plate on the other side of the wall. Seismic Retroﬁ tting Ú 287

Figure 11C.10: Gable Guy-Lines

Source: Author.

Floor Level Diaphragm — for Stone on Joist Floor Only Since the ﬂ oor joists are not anchored to a full band or wall plate the diaphragm action may be achieved through the installation of diagonal ties and struts on the underside of all the joists. Th ese could consist of (i) 100 mm × 25 mm wooden members installed with a minimum of four nails at the ends and two nails in between, or of (ii) MS angle 35 × 35 × 3 either screwed in case of wooden joists or welded to the exposed reinforcing bars of the PC/RCC joist. Th e struts shall touch the far walls at their ends. Pitched Roof In-plane Diaphragm Th e diaphragm action in roofs will be achieved by installing diagonal ties and struts. Downloaded by [University of Defence] at 01:15 24 May 2016 Struts: 100 mm × 25 mm timber strut may be installed across the underside of all the rafters, one near the ridge parallel to it, and the other near the lower end, parallel to the wall. Struts to be attached to each rafter it passes under with a minimum of two 5 mm diameter screws/nails 50 mm long. Diagonal Ties: In-plane diagonal ties may consist of 4 to 6–16 gauge GI ﬂ exible wires pre-tensioned by twisting. Tie along its length, to be attached to the underside of each rafter after pre-tensioning. Ties to be simply tied to the struts. 288 Ú Rajendra Desai

Figure 11C.11: Installation of In-Plane Bracing for Roof Diaphragm

Source: Author.

Figure 11C.12: Roof Diaphragm Bracings Downloaded by [University of Defence] at 01:15 24 May 2016

Source: Author. Seismic Retroﬁ tting Ú 289

References

Agarwal, Pankaj and S.K. Th akkar. 2003. ‘Seismic Evaluation of Strengthening and Retrofi tting Measures in Stone Masonry Houses Under Shock Loading’. Proceedings of Workshop on Retrofi tting of Structures, Roorkee, October. Arya, A.S. 2003. ‘Retrofi tting of Buildings — A Critical Step for Reducing Earthquake Hazard Damage’. Proceedings of Workshop on Retrofi tting of Structures, Roorkee, October. Bose, P.R. and Alok Verma. 2003. ‘Retrofi tting of Low Cost Buildings’. Proceedings of Workshop on Retrofi tting of Structures, Roorkee, October. Building Materials and Technology Promotion Council (BMTPC). 1997. Vulnerability Atlas of India: Earthquake, Windstorm and Flood Hazard Maps and Damage Risk to Housing. Ministry of Housing and Urban Poverty Alleviation, Government of India. ———. 1999. Guidelines — Improving Earthquake Resistance of Housing. Ministry of Housing and Urban Poverty Alleviation, Government of India. ———. 2000. Guidelines for Damage Assessment and Post-Earthquake Action Part II: Repair and Retrofi tting of Buildings in the Chamoli Earthquake Aff ected Areas. Ministry of Housing and Urban Poverty Alleviation, Government of India. Desai, Rajendra. 1999. Field Shake Table Program — A Sure Way of Confi dence Building in Quake Resistant Building Technology. Ahmedabad: ASAG. Desai, Rajendra and Rupal Desai. 1990. How the Poor Build Houses in Rural Gujarat. Ahmedabad: Ahmedabad Study Action Group (ASAG). Earthquake Resistant Design and Construction of Buildings — Code of Practice (Second Revision), IS:4326-1993. Improving Earthquake Resistance of Earthen Buildings-Guidelines, IS: 13827: 1993. Improving Earthquake Resistance of Low Strength Masonry Buildings-Guidelines, IS: 13828:1993. Indian Society of Earthquake Technology (ISET). 1989. A Manual of Earthquake Resistant Non-Engineered Construction. Roorkee: ISET. Repair & Seismic Strengthening of Buidlings-Guidelines, IS: 13935:1993. TARU. 1997. Rapid Assessment of Damage to Buildings Affected By Jabalpur

Downloaded by [University of Defence] at 01:15 24 May 2016 Earthquake. Prepared for BMTPC, Ministry of Housing and Urban Poverty Alleviation, Government of India. ———. 1999. Chamoli Earthquake Repair and Reconstruction Plan. Prepared for BMTPC, Ministry of Housing and Urban Poverty Alleviation, Government of India. ———. 1999. Visual Damage Identiﬁ cation Guide for Building Seismic Damage Assessment, Repair and Reconstruction Options for NW-Himalayas. Prepared for BMTPC, Ministry of Housing and Urban Poverty Alleviation, Government of India. 290 Ú Rajendra Desai

TARU. 2001. Rapid Assessment of Damage to Buildings Affected By Kachchh Earthquake of January 26, 2001. Prepared by TARU for BMTPC, Ministry of Housing and Urban Poverty Alleviation, Government of India. Visual Damage Identiﬁ cation Guide for Building Seismic Damage Assessment, Repair and Reconstruction Options for NW-Himalayas, by TARU LEADING EDGE, for BMTPC, March 1999. Downloaded by [University of Defence] at 01:15 24 May 2016 12 Medical Management of Earthquake Disasters Brig KJ Singh

A disaster is often attributed to an ‘Act of God’. Th e World Health Organisation (WHO) has defi ned a disaster as any occurrence that causes physical damage, economic disruption, loss of human life and deterioration in health services on a scale suffi cient to warrant an extraordinary response from outside the community area. An objective study of the characteristics of the physical force of an earthquake and the resultant damage has produced a change in perception regarding the issues that pertain to coping with the eff ect of disasters. Th e focus now is on public and professional participation in assessing risks and developing disaster-resistant communities. Th e northern part of India lies in an earthquake-prone zone. Approxi- mately 55 per cent of the Indian landmass has witnessed seismic activity, albeit of varying intensity. In 1993, the earthquake that rocked Latur and Osmanabad districts of Maharashtra killed almost 10,000 people and caused great misery. On 26 January 2001, an earthquake demolished the Bhuj area of Gujarat, resulting in over 30,000 deaths and tremendous material loss. Th e fi rst tremors of an earthquake on 8 October 2005 brought widespread devastation to the State of Jammu and Kashmir, houses and buildings collapsed, many lives were lost. Th e worst-hit areas were Uri, Tangdhar, Rampur, and Poonch.

Downloaded by [University of Defence] at 01:15 24 May 2016 Disaster Management

Disaster management, though a matter of concern for a long time, gathered momentum in the decade 1990–2000, which was declared as the International Decade for Natural Disaster Reduction (IDNDR) by the United Nations. Although in the beginning the IDNDR followed a strictly technocentric and scientiﬁ c approach, after the Yokohama Conference of 1994, the focus shifted to eﬀ ective disaster prevention, 292 Ú Brig KJ Singh

mitigation and preparedness. Th ese are deemed better than disaster response in achieving the goals and objectives of the IDNDR. Disaster response alone is not suﬃ cient for it yields only temporary results, that too at a very high cost. Th is has been further demonstrated by the recent focus on response to complex emergencies which, although compelling, should not divert us from pursuing a more comprehensive approach. Prevention contributes to lasting improvements in safety and therefore is essential to integrated disaster management. Once the IDNDR lapsed, these initiatives were supplemented by the United Nations International Strategy for Disaster Reduction (ISDR) which came into being in December 1999.

Disaster Management Cycle

Th e activities of each stage of the disaster management cycle are inter- linked and mutually dependent, with considerable overlap.

(i) Prevention: Measures to prevent the disaster from having harmful eff ects on the community, e.g., legislation to minimise its eff ect through land-use regulation. (ii) Mitigation: Specifi c programmes intended to reduce the eff ects of a disaster on a community, e.g., enforcement of disaster- resilient building codes. (iii) Preparedness: Measures which enable governments, organisations, communities, and individuals to respond rapidly and eff ectively to disaster situations. Th is includes the formulation of viable disaster contingency plans which can be brought into eff ect when required. (iv) Disaster Impact: Th is being the point at which the disaster event occurs or aff ects a particular area. (v) Response: Th ese measures are undertaken prior to and imme-

Downloaded by [University of Defence] at 01:15 24 May 2016 diately following an earthquake, e.g., the evacuation of personnel from threatened areas; the restoration of essential services — food, water, electricity, health services; and the physical and psychological rehabilitation of the aﬀ ected population. (vi) Recovery: Th is is the process by which the aﬀ ected communities are assisted in returning to their normal level of functioning after a disaster. Th is process can be very protracted, taking 5–10 years or even more. Major activities in this period pertain to restoration, rehabilitation and reconstruction. Medical Management of Earthquake Disasters Ú 293

(vii) Development: Th is aspect of disaster management serves as a link between disaster-related activities and national development. Its inclusion is intended to ensure that lessons learnt from the disaster are eﬀ ectively reﬂ ected in future policies and action plans.

Disaster Plan

A disaster plan is needed to be prepared for eﬀ ective performance in a disaster situation. Th e plan should be simple and concise, and constitute a part of the broader regional plan (district, State, national or group of nations). It must include all the relevant components, such as the ﬁ re brigade, government administration, health departments, police, etc. Maharashtra has a comprehensive multi-hazard disaster management plan. Information about a disaster is conveyed to control rooms through a satellite-based communication network. The standard operating procedure (SOP) of all government departments would then be set in motion, and earth-moving vehicles and rescue teams would rush in from their predetermined locations. An exhaustive database has been generated based on vulnerability analysis and logistics have been planned in support of relief activities.

Disaster Manual

A disaster manual is a written statement about the disaster plan which is to be followed during disasters. All organisations involved in disaster management activities should prepare a disaster manual which is to be derived from the disaster plan of that region/area. It describes in simple and clear terms the command, control, communication, logistics and SOPs applicable to that particular organisation. Downloaded by [University of Defence] at 01:15 24 May 2016 Health Services

Th e healthcare preparedness plan of disaster management should be an integral part of the overall national disaster preparedness plan. It must be understood that many sectors will be involved in the plan. These will include, among others, meteorology, services, national planning, public works, utilities (electricity, water, gas, and sanitation), construction, food, industry, ﬁ nance, communications, transportation, 294 Ú Brig KJ Singh

education, public health, armed forces, civil defence, police, ﬁ re department, the Red Cross Society, and private volunteer organisations, etc. Each agency will have to develop its own speciﬁ c action plan to execute the particular task assigned to it within the national plan. Th eir activities should be arranged in three distinct chronological stages:

(i) Pre-disaster preparedness (ii) Action taken during a disaster (iii) Post-disaster rehabilitation

The general or overall planning objectives regarding healthcare provision in a disaster situation are to:

z Co-ordinate the use of resources and make it more rational so as to avoid duplication of activities and functions. z Maintain an accurate inventory of human, material and institutional resources available in the medical care area. Ensure regular rest and rotation of personnel, and regular replenishment of medical supplies. z Carry out training activities for healthcare personnel, volunteers and the staﬀ of the emergency care system and the civil defense system. z Ensure the establishment of effective co-ordination and co- operation between national and regional organisations, e.g., the National Disaster Management Authority has a task at central level of providing a Rapid Response Force, whereas States and districts have their own plans; co-ordination between them is essential. z Orient community action in the proper and required directions in disaster situations.

Organisation Downloaded by [University of Defence] at 01:15 24 May 2016 Eﬀ ective and eﬃ cient disaster management requires a multi-dimensional approach. Th ere should be intra- and inter-networking of districts, States and the nation or even groups of nations. Networking may be understood as linking up for the augmentation/optimisation of available resources, which may be in the form of information, materials and/or manpower. Th e organisation of healthcare centres for a disaster situation should be planned in advance and should be based primarily on the existing Medical Management of Earthquake Disasters Ú 295

or normal health and medical care structure available in the area. It should include not only the offi cial healthcare system but also all other autonomous and private medical facilities. Th e basic structure should be maintained as far as possible, but it should be adjusted and strengthened according to the area aff ected and the magnitude and extent of the disaster. Th e activities of voluntary agencies, in a disaster situation, are adapted to the plan decided by the National Disaster Management Authority in general and the National Committee on Healthcare in particular. Most demands for healthcare and medical services occur within the fi rst 24 hours of the disaster. Injured people may continue to be brought to medical facilities during the fi rst 3–5 days after which presentation patterns return almost to normal. Patients may come in two waves — fi rst, comprising casualties from the area immediately surrounding a medical facility, and the second, of referrals as relief operations in the more distant areas get organised. Depending on the location and magnitude of the calamity, fi rst-level care is generally organised at the disaster site itself to provide quick relief and fi rst aid to the victims. Search-and-Rescue Units Th ese personnel rescue the injured from the disaster sites and administer fi rst aid to them so that they may then be transported to the closest medical care centre in the disaster area. At the same time an initial triage (the classifi cation of mass casualties based on the medical criterion of chance of survival) is performed. Th is is necessary when a large number of casualties must be cared for with limited availability of medical resources. Th e principle that is adopted is ‘maximum good to the maximum number’. Th e time and resources spent in attempting to salvage a hopeless case should not be at the risk of deterioration in the condition of many other salvageable cases. In this process colour-coded tags are used to indicate the degree of seriousness of injury and priority of evacuation: Downloaded by [University of Defence] at 01:15 24 May 2016 thus red signifi es fi rst priority for evacuation and immediate medical care, green signifi es second priority for evacuation as the injuries are not life-threatening, yellow signifi es minor lesions or fatal injuries, and black tags are placed on the dead. Hospitals All hospitals, health centres and other parts of the national health system should have an emergency plan to guide them in the event of 296 Ú Brig KJ Singh

a disaster. Th is plan would also cater to services provided outside the facility and also, therefore, in the event of the facility itself being aﬀ ected by the disaster. Th e permanent medical care facilities in the area must be reinforced to undertake referrals and to provide simple nursing care. Mobile hospitals that can provide ambulatory patients ﬁ rst aid and simpler primary medical services may be set up as required. Th ey may be built on a bus chassis or inside a cargo container transported by a prime mover. Operation Theatre Th e provision of an operation theatre (OT) is necessary for surgical treatment of the victims. It may be necessary to improvise an OT if a suitable building is not available. Th e options available for this purpose are:

(i) Tents: Th ese are cheap, easily available in a variety of sizes and can be rapidly erected. A reasonable level of asepsis can be obtained by using a laminated sheet as fl oor cover and laminated boards to buttress the walls. (ii) Prefabricated Huts: Th ese are similar to structures used in the fi eld to create temporary settlements. Th ey are readily available and cost approximately Rs 15–20 lakh. Various shapes and sizes can be constructed in modular form to suit specifi c requirements. (iii) Pneumatic Shelters: Th ese are collapsible, and can therefore be transported easily. Th ey are similar to tents in shape. Infl ating the pneumatic columns with air erects and maintains the shape of the structure. Th e cost is approximately Rs 2 crore. (iv) Container-based OT: Standard 20 × 8 feet cargo containers can be converted into a modern OT. Th ey can be transported by road, rail, sea or air. Th eir cost may vary from Rs 3–5 crore. Downloaded by [University of Defence] at 01:15 24 May 2016 (v) OT-on-wheels: Th ese are similar to container-based OTs with the important exception that they are self-propelled. A number of vehicles are used, comprising various modules, which when interconnected create a modern OT. Th e cost may vary between Rs 15–20 crore. Since each vehicle constitutes a part of the OT, the reliability of the vehicle is of critical importance, as the non-availability of even one vehicle may render the OT complex completely unserviceable. Medical Management of Earthquake Disasters Ú 297

Disposal of Dead Bodies All dead bodies should be segregated. If required a temporary morgue should be created. Details of identiﬁ cation of the bodies and photographs, if possible, should be kept and maintained as a record. Bodies should be handed over to the police for further disposal. Volunteer Services After a natural disaster a large number of volunteers may be willing to oﬀ er their services. Th e utilisation of their services requires proper co-ordination. Th ey can be assigned tasks such as escorting patients, providing telephone contact, work in stores, laboratories and other such areas, as per their competencies. Rehearsal and Revision of Plans Th e hospitals’ disaster management plan should be periodically rehearsed and updated. Suitable simulated drills, for example, add an element of realism to the training. Special emphasis must be given to the adequate provision of water, electricity, medical supplies, and communications. Ethics and Teamwork Disasters pose the dilemma of what is the best allocation of resources? Innate human virtues like honour, respect for colleagues, a desire to care for the ill and the injured come to the fore in disaster situations. People with a sense of charity and compassion come together and work as interdependent teams to help alleviate the misery of the disaster victims.

Environmental Health

Environmental health may be understood as referring to the control of

Downloaded by [University of Defence] at 01:15 24 May 2016 those factors in the environment that may have a deleterious impact on people’s physical, mental or social well-being. Earthquakes can and often do result in considerable deterioration of environmental conditions. Partial or total disruption of environmental health services is expected in the aftermath of an earthquake, particularly of services such as water supply, food production and distribution, transportation and power. When sewerage and liquid waste disposal systems are disrupted the probability of water-borne diseases increases. Th e interruption of vector control activities, similarly, can increase the incidence 298 Ú Brig KJ Singh

of diseases that are otherwise incidental. Overcrowding itself is a major cause for housing sanitation to decline. Lack of water and toilet facilities makes it diﬃ cult to maintain personal hygiene and there is a rise in the number of instances of diarrhoeal disease, typhoid, scabies, etc. Environmental health services, essential to ensuring the well-being of individuals in high-risk areas must therefore receive the highest priority. Equally important, in this endeavour the co-operation of the local population should be sought. Th e minimum level of necessary services that must be provided include:

z Adequate supply of drinking water z Protecting food supplies against contamination z Adequate shelter provisions z Facilities for excreta and liquid waste disposal z Protection against water-borne diseases through vector control activities and chemoprophylaxis z Vector Control z Personal hygiene Community Involvement and Education In this entire exercise, the unavailability of the appropriate manpower required to maintain good environmental health can be a limiting factor in an emergency intervention after a disaster. Using locally available experts should, in such a situation, be given ﬁ rst consideration because they are much more familiar with the pre-disaster socio-economic conditions of the aﬀ ected area. Community awareness and participation is in fact integral and extremely important in all relief activities. Th is is necessary to keep the community informed about its own vulnerability in the future and give it a critical role in disaster management activities. To that extent, public education is of utmost importance. Awareness of the emergency measures that may be necessary after a disaster and what

Downloaded by [University of Defence] at 01:15 24 May 2016 may be further expected of the public is a major step towards reducing operational plan problems. Water Supply Providing adequate amount of drinking water is of the utmost importance to supporting life in the aftermath of a disaster. Only after drinking water requirements have been met should supply for domestic uses be considered. Needless to say, water puriﬁ cation by suitable methods Medical Management of Earthquake Disasters Ú 299

such as chlorination should be ensured and tanks should be cleaned in order to prevent any later contamination. Environmental Sanitation Excreta disposal and vector control is necessary to prevent contamination of food and water. Solid waste disposal requires receptacles, means of transportation, and incineration and burial facilities. Functioning of the public sewage systems must, therefore, also receive attention. Shelter Th e fi rst priority is to provide shelter to displaced persons, even if it has to be in public buildings like schools, churches, hotels, etc. Sub- sequently, these people must be assisted in the construction of temporary shelters or in repairing their old dwellings. Food Sanitation Food degradation can occur because of improper storage conditions that are a result of disruption in power supply and the physical damage suff ered by warehouses. All food must be inspected for fi tness for consumption, especially at mass feeding centres. Th e public should be made aware of which foods are safe to consume. Vector Control Unsanitary conditions and debris accumulation allow insects and ro- dents to proliferate. Th e population should be informed about measures to eliminate their breeding sites as well as other protective measures. Insecticide sprays supplement these measures and help to prevent the proliferation of insects. Personal Hygiene Lack of water and bathing facilities leads to a fall in the standards of Downloaded by [University of Defence] at 01:15 24 May 2016 personal hygiene. A suffi cient quantity of water therefore should be made available in residential sites. Establishing Settlement Sites for Displaced Persons Safe water, food supplies and basic sanitation facilities must be available in all camps for displaced persons. Also the occupants of the camps should be encouraged to return to their homes by assisting them in reconstruction activities. 300 Ú Brig KJ Singh

Emotional Stress Disorder

Disaster situations may result in the loss of loved ones, or of cherished belongings. Diﬀ erent people, however, respond to this kind of stress diﬀ erently. While some initial responses may be fear, disbelief and disorientation, later responses include anger, apathy, depression, withdrawal, anxiety, etc. In fact, the individuals employed in disaster relief may themselves become victims of emotional disorders. Some alleviating measures are that have been suggested include meditation, psychotherapy and rest.

Medical Store Supply

Earthquakes cause many injuries and often result in death. Medical attention and medical supplies therefore, are urgently required after an earthquake. Medical supplies are defi ned as pharmaceuticals and other supplies required for the medical treatment of victims. Medical supplies constitute a subset of the larger group of health supplies required in times of emergency. Health supplies include, but are not limited to, hospital, clinic, laboratory, X-ray and environmental health related supplies and equipment. However, such emergencies may require the sudden use of specifi c medical items that may be in short supply. Even if available within the country, they may not be available where needed and maybe diffi cult to provide due to the damage and disruption caused by the disaster. On the other hand, local medical supplies are more likely to be salvageable after an earthquake. Moreover, even during normal times, when procedures for medical supplies management are fairly routine, it is a sizable task that requires employment of qualifi ed manpower. Th is is also true in a post-disaster

Downloaded by [University of Defence] at 01:15 24 May 2016 scenario, and becomes that much more complicated. Complex co- ordination is required between public and private bodies, and between national and foreign agencies and institutions that may be the source of emergency supplies. Th e most critical supply demands often arise within the fi rst 48 hours after impact, and external assistance probably will not be available at such short notice. Th erefore, effi cient medical supply management at the local level is essential. During the fi rst few weeks after impact, the health relief supplies needed will consist mostly of pharmaceuticals Medical Management of Earthquake Disasters Ú 301

and medical equipment required for treating casualties and preventing communicable diseases. Later needs may include health supplies and equipment necessary to restore public services. Blood Supply Being a scarce and perishable item the storage and transportation of blood requires special consideration. To make the availability of blood easier, linkages and networking with established blood banks is mandatory. Vaccines Vaccines should not be placed on the list of essential supplies or used in emergency disaster relief. Th is policy is advocated for many reasons, including the inability to control a proper and complete mass campaign in a post-disaster scenario. Mass vaccinations are not recommended. However, routine vaccination programmes should not be ignored or postponed during an emergency. Th e normal immunisation programme must be continued even during the emergency period or re-established at the earliest. Medical Supply Management Unsolicited donations of medical supplies are more of an impediment than a help. Th ey generally accrue — unsorted, unlabelled, poorly packed, used or outdated — or are not meant for emergency use. Th ey often create logistical problems that can interfere with critical disaster-relief functions. A continuous cold chain1 is not always guaranteed thereby resulting in degradation even before reaching the destination. Health relief co-ordination should in fact be designated as a part of pre-disaster planning with a single-point clear mandate to direct all health sector relief activities, including approving requests for external co-operation and accepting or rejecting oﬀ ers of assistance. Th e public and private

Downloaded by [University of Defence] at 01:15 24 May 2016 components of the health sector should be integrated with overall relief and rehabilitation eff orts. Eff ective management depends on integrating and identifi cation of problems as they arise and delivering specifi c items precisely when they are needed.

1 Th e cold chain is the process of transporting and storing vaccines within the safe temperature range of 2–8 degrees celsius from the point of manufacture to the point of administration. 302 Ú Brig KJ Singh

Th e principle areas of preparation include:

(i) Standardising availability of medical supplies

z Brick System2 z Earthquake brick z Surgical brick z Avalanche brick (ii) Ensuring availability of stocks (iii) Securing location and access (iv) Acquisition and requisition (v) Distributions and transportation (vi) Communication (vii) Training of personnel (viii) Co-ordination with agencies and government (ix) Legal aspects (x) Simulation and audit of medical supply management Standardised List of Medical Supplies A standardised list of essential medical supplies should be prepared at the national level. WHO/United Nations High Commissioner for Refugees (UNHCR) has developed a list of standard drugs and clinical equipment which can meet the requirements of 10,000 persons for three months. In 2005, following the Kashmir Earthquake (also known as the South Asian Earthquake), WHO kits containing standardised dressing materials were sent to Srinagar from Delhi for the victims. Other Stores Th e following items require special attention:

z Tentage and improvised shelter

Downloaded by [University of Defence] at 01:15 24 May 2016 z Clothing and bedding z Water tank and puriﬁ cation material z Food items z Generator, kerosene oil and emergency lights

2 Brick system is a logistics supply term. A predetermined number and quantity of each item constitutes a brick. Stores are issued as bricks or multiples of bricks. Th e aim is faster stocking and replenishment. Medical Management of Earthquake Disasters Ú 303

z Furniture — table, chair, trolleys, beds, wheel chair, stretchers z Hygiene and sanitation equipment

Case Study Management of the Gujarat Earthquake (2001): Victims at the Military Hospital, Bhuj

Kachchh is a geographically isolated border district of Gujarat — the largest in the State. With the Greater Rann (wilderness) to the north, the Little Rann to the east, it is bound by the sea towards the west and the south. Bhuj is the district headquarters and it has an airport. Kandla is a major seaport located in its south-eastern part. Even though Kachchh is located in Seismic Zone V, there was a general lack of information and awareness of the earthquake risk and its implications amongst all sections of society, and therefore a total lack of preparation to be able to deal with the disaster.

Disaster Impact

Th e earthquake struck without warning at 08:46 am IST on 26 January 2001. Th e epicentre was located approximately 30 kilometres north- east of Bhuj and measured 6.9 on the Richter scale. Th ere was massive destruction of property and infrastructure. Th e rail and road links across the Little Rann were disrupted. Electricity, water supply and telecommunications links were disrupted. Th e district hospital in Bhuj had been destroyed. Civil administration was left completely overwhelmed by the magnitude of the disaster and there was a general state of anxiety and panic in the local population, made worse by a total lack of information about the impact of the disaster, its geographical extent and the availability of counter-disaster resources. Downloaded by [University of Defence] at 01:15 24 May 2016 Local Response at the Bhuj Military Hospital

Th e immediate medical response centred around the military hospital in Bhuj. Th is hospital is meant to provide basic speciality services to local armed forces personnel and their families. Being a border hospital, one of its roles is to cater to casualties in case of hostilities. It had contingency plans and the resources needed to fulﬁ l that role. In case of natural calamities its mandate was to augment the resources of the civil health 304 Ú Brig KJ Singh

authorities. However, in the aftermath of the earthquake, the hospital had to act as the fi rst responder despite its buildings having suff ered severe structural damage. Th is unit took the lead in the medical relief operations in spite of numerous resource constraints. Th e most important task was to ensure the operational readiness of the hospital for mass casualty management. A number of concurrent actions were initiated in order to cope with the mass infl ux of injured people pouring into the hospital. A quick damage assessment of the hospital was carried out. All available paramedical personnel were rapidly deployed to set up multiple reception, triage and resuscitation stations. Other personnel were utilised to move indoor patients into the open and to salvage equipment and stores. An operation theatre zone was established in the open on a cemented platform. Makeshift operating tables were placed in these areas. Pre-operative and post- operative areas were earmarked adjacent to the operative areas. Patients’ holding areas and evacuation areas were demarcated. Th e local army formation provided generator sets, water tankers, tents and personnel for crowd control. Subsequently, plaster-of-Paris application areas, X-ray, laboratory and blood storage facilities were established. A delivery suite/labour room was set up to meet the emergency obstetrics requirements of the community. An information desk was made functional. Cooking facilities were set up to provide food and tea for hospital personnel, patients and their attendants. Paramedical personnel from the sister fi eld medical unit were sent along with the army search-and- rescue teams to provide necessary aid to the victims.

Casualty Management Protocol

A simple standardised patient management protocol, adapted to the available resources and skills, was followed. Th e aim of this protocol was to standardise treatment, save lives, prevent major secondary com-

Downloaded by [University of Defence] at 01:15 24 May 2016 plications and prepare casualties to withstand evacuation to hospitals outside the disaster zone. A protocol of graded assessment was prepared to cope with the large influx of casualties. Paramedical personnel graded the casualties into relatively minor or severe injuries, and thus carried out the initial assessment. All patients with major injuries were given treatment and resuscitation. Th e protocol consisted of: Injection Tetanus Toxoid 0.5 ml intramuscular, Injection Cefotaxime 1 gram intramuscular, Injection Diclofenac 75 mg deep intramuscular, Medical Management of Earthquake Disasters Ú 305

and Ringer lactate infusion through intravenous canula. Th e phys- ician and medical offi cers carried out airway management and placed chest drainage tubes for those with respiratory embarrassment. Th e gynaecologist, who was also the administrative leader of the team, triaged the patients into those whose injuries could be handled locally and those who needed specialised facilities. Th e fi nal decision on whether or not to salvage injured limbs was performed by the operating surgeon at the operating table itself. Th is protocol of graded assessment allowed the hospital to regulate the management of patients with some semblance of order. Some idea of the scale of the workload can be imagined by the fact that almost 300 major procedures and approximately 3,000 minor procedures were performed at the facility had been performed by midnight on 26 January 2001. Th e military hospital was helped by about 20 surgeons, orthopaedic surgeons, general practitioners and even a dentist from the town and the destroyed local district hospital. Medical relief teams arrived at the Military Hospital Bhuj only around 11 pm on 26 January 2001 after the local airfi eld had been cleared for receiving aircrafts. Th e returning aircraft was used to evacuate casualties to hospitals located outside the disaster zone. Th is redistribution allowed the hospital’s existing space to handle more patients and also provided for the management of the injured in more conducive circumstances.

Central Response

As news of the devastation caused by the earthquake fi ltered in, armed forces’ medical services rapidly mobilised resources to rush relief workers to the disaster zone. Th is massive undertaking was called ‘Operation Sahayta’. As soon as the airfi eld at Bhuj was declared functional, an immense air eff ort started ferrying in rescue, heavy engineering, medical and communications equipment and personnel. Relief,

Downloaded by [University of Defence] at 01:15 24 May 2016 medical and subsistence supplies also poured in. Th e medical relief operation consisted of deploying 10 self-contained surgical teams and six fi eld ambulances (mobile medical units). Of these fi ve surgical teams and one fi eld ambulance were able to reach Bhuj by midnight of 26 January. Th e focal point of the medical relief operations was the military hospital in Bhuj. From here the fi eld ambulances were deployed all over Kachchh district. Th ey set up mobile surgical hospitals at the worst-aff ected areas. Th e Navy dispatched a hospital ship, which 306 Ú Brig KJ Singh

was berthed at Kandla port. Mobile medical teams fanned out into the less accessible areas along with relief and rescue teams. With the arrival of medical relief teams and aid supplies at Military Hospital Bhuj the tempo of work reached a war footing.

Casualty Data

Between 26 January and 31 January 2001, 12,975 patients were treated at the military hospital. Th e surgical workload is shown in Table 12.1:

Table 12.1: Surgical Operations Conducted at Military Hospital Bhuj

No. of major No. of minor Date surgeries surgeries Total Remarks 26 January 2001 303 3,085 3,388 27 January 2001 1,222 3,329 4,551 Peak Load 28 January 2001 219 681 900 29 January 2001 227 169 396 30 January 2001 16 235 251 31 January 2001 15 29 44 Total 2,002 7,528 9,530 Source: Compiled by author from the hospital’s operations register.

During the same period, the other armed forces medical services elements in the area treated 16,610 patients, of which the number of major surgery cases were 235 and minor surgery were 1,046. Workload fi gures are clearly representative of the fact that the fi rst 48–72 hours following sudden impact disasters are crucial in the management of casualties. Emphasis was also put on the evacuation and redistribution of casualties to other hospitals outside the disaster zone. Th is was done to provide the injured facilities for defi nitive surgery under

Downloaded by [University of Defence] at 01:15 24 May 2016 optimal conditions and also to make place for the casualties awaiting treatment that were continuing to pour in. Th e aircraft returning after ferrying in aid were utilised for the purpose of evacuation. Casualties were airlifted to various armed forces hospitals at Pune, Khadki, Jamnagar, Ahmedabad and Mumbai. Subsequently, as communications systems improved, casualties were also evacuated to various civil institutions depending upon the availability of bed. A community health team was inducted to look after the environmental health aspects. Units of Medical Management of Earthquake Disasters Ú 307

the armed forces medical services also participated in the health surveillance programme of the WHO and the Gujarat government.

Difﬁ culties Encountered

(i) Collapse of Civilian Health Services: Th e civil hospital was razed to the ground by the impact of the earthquake. Th e military hospital at Bhuj, therefore, was required to act as the ﬁ rst and sole responder in the crucial period immediately following the disaster. (ii) Structural damage to the military hospital: Some wards, the radiology department, pathology department and the labour/ delivery rooms had collapsed. Th ere was extensive damage to the operation theatre, the medical stores, the physiotherapy department and the generator room. All the buildings of the hospital developed cracks. (iii) Inadequate Facilities for Surgical Treatment: Surgery had to be done in the open, under tents, due to damage to the hospital and the operation theatre. Th ere was also the danger of further aftershocks, limiting use of indoor facilities. (iv) Shortage of Operating Room Equipment, Instruments, Linen, and Disinfectants: Th is was due to the large number of casualties which had to be attended to due to the sudden impact of the disaster. Th is problem was overcome once the supply channels and communications links were re-established. To tide over the crisis, rapid chemical disinfection of instruments was resorted to. (v) Lack of Water and Electricity Supply: Following the earthquake, there was complete disruption of water and electricity supply. Th e generator room of the hospital was badly damaged and unusable. Generator sets and water tankers had to be requisitioned from the local army formation.

Downloaded by [University of Defence] at 01:15 24 May 2016 (vi) Induction of Medical Teams and Personnel: Civil medical teams and personnel reported to the disaster area without adequate briefi ng and preparation. Th ey came without the necessary equipment or even the means of personal subsistence. Th ey, in fact, had to be supported by the aff ected community. (vii) Crisis Expansion: Bhuj’s military hospital is a peripheral hospital with a limited crisis expansion capacity. Th is was inadequate to deal with the large infl ux of casualties that poured into its 308 Ú Brig KJ Singh

precincts. No tents had been issued to the hospital. Th ese had to be requisitioned from the local army formation. ( viii) Lack of Communications: All communications at the hospital were completely cut off . Th ere was a partial restoration of telephone communications after 29 January and on 30 January 2001 a satellite phone was made available. (ix) Information Management: Th ere was lack of information and no proper assessment of the extent and magnitude of the earthquake’s impact, the availability of other counter-disaster resources, and the serviceability of road links and hospitals to which casualties could be referred. Similarly information could not be passed on as to the number, nature and severity of injuries suffered by patients being evacuated. Due to a paucity of staff there were many delays in the functioning of the information cell. (x) Casualty Evacuation: Due to lack of functioning communications systems and facilities it was not possible to plan an orderly redistribution of casualties to other hospitals outside the disaster zone. As a result of the haphazard casualty evacuation that took place, family units were broken up. Th e absence of special stretchers for evacuation of patients with spinal trauma was another major problem. (xi) Documentation: There were 82 personnel available at the hospital on 26 January. Th is included fi ve medical offi cers, seven nursing offi cers and 29 paramedical staff . Th e entire staff was involved in providing medical aid to the injured; engaging in salvage operations; arranging logistical support for casualty management, hospital personnel and their families; and interacting with the local military formation and fi eld ambulances. As a result, satisfactory documentation of casualties was not possible until the arrival of reinforcements.

Downloaded by [University of Defence] at 01:15 24 May 2016 (xii) Crowd Control: Th ere was a large inﬂ ux of casualties to the hospital as it was the only functional hospital in the disaster zone. Relatives of the seriously injured patients demanded early surgery from the overworked and exhausted surgeons. Th ey were sometimes not willing to remain in prioritisation areas, and would try to rush their injured relations into the operating zones and harass the surgeons. Inadequate personnel were available to control the crowd and streamline the ﬂ ow of patients. Medical Management of Earthquake Disasters Ú 309

( xiii) Disposal of Dead Bodies: Th e task of identifi cation, photo- graphing and disposing off dead bodies was given to the civil police. (xiv) Biomedical Waste Disposal: Amputated human limbs and other biomedical waste was disposed of by incineration. (xv) Human Refuse Generation: The sewage system had been rendered non-functional by the earthquake. The capacity of the few shallow trench latrines and urinals too was soon overwhelmed by the mass of humanity (relatives and well- wishers of the casualties) who thronged the hospital. ( xvi) Management of Relief Supplies: Th is was critically aff ected by the lack of communications systems. Th ere was an initial delay in the arrival of appropriate and required supplies, which was overcome once communications improved. A signifi cant problem was the donation of unsolicited relief supplies by generally well-meaning but sometimes publicity-seeking donors. Th ese supplies were not need-based, required sorting and burdened the staff who were expected to manage them. Donors would often get upset when their help was politely turned down and they were directed to the civil authorities. Th ere was also the problem of storing the medical relief supplies that were coming in. Th ese were kept in tents where they were exposed to dust and the elements. ( xvii) Media Management: Th e visual and print media have a powerful role to play in shaping public perception and opinion. In such situations, clear-cut guidelines and detailed briefi ng of a spokesperson are required to avoid any possible misinter- pretation of the situation and misreporting of facts.

Lessons Learnt

Downloaded by [University of Defence] at 01:15 24 May 2016 Th e most important lesson, brought out in stark reality in the aftermath of the Gujarat earthquake, was that the best way to deal with such calamities is to increase public awareness and preparedness. Every facility must carry out a realistic risk assessment for likely disaster scenarios, their vulnerability to such disasters and their capability in terms of remaining functional and coping with the aftermath of such a disaster. Local counter-disaster resources need to be identiﬁ ed in the light of such assessments and appropriate institutional and community-based 310 Ú Brig KJ Singh

disaster response plans must be prepared. Another essential pre- disaster activity would include liaison with other similarly involved agencies to formulate a co-ordinated command, control and response mechanism. Proper back-up systems need to be identifi ed that can cater to requirements of communications, power, water supply, transport, shelter, and relief supplies. Caution needs to be exercised while inducting healthcare personnel into a disaster zone. Th ey need to be self-contained and adequately equipped so as not to become a burden on the already strained local resources. Similarly, the medical relief supplies required must be need-based — appropriate for the given situation, recognisable and in secure packing. Information management plays a vital part in an eff ective response strategy. Th is requires the availability of a functional communications set-up at the health facility engaged with counter-disaster measures. Th is will allow better fl ow of information on disaster impact, availability of other counter-disaster resources, projection of requirements and information on the redistribution of casualties to other facilities. In- formation management involves providing adequate information on the status of casualties to the next-of-kin. Dealing with the media is also an integral task of a counter-disaster facility and requires an informed spokesperson. It would help in providing proper coverage and avoid any misrepresentation of facts. In fact, short modules on dealing with the media should be a part of service training courses.

Organisation of Disaster Management in India

Each disaster exposes the chinks in the mechanisms of disaster preparedness and response in India. Th e Gujarat earthquake of 2001 was no diﬀ erent. We must, however, learn from the experiences acquired during this calamity, otherwise we will be condemned to repeat the same

Downloaded by [University of Defence] at 01:15 24 May 2016 mistakes, possibly with more tragic consequences. In view of the large-scale loss of life and property caused by natural calamities and the devastating potential of man-made disasters, the union government decided to institutionalise disaster management and focus on prevention, an enhanced level of preparedness, prompt and eﬀ ective response, and capacity-building. To this end, on 23 December 2005, the Disaster Management Act, 2005 was enacted. Th e Act seeks to institutionalise a mechanism for eﬀ ective disaster management at the national, State and district levels, that will allow the administration Medical Management of Earthquake Disasters Ú 311

to plan, prepare and ensure a swift response to both natural calamities and man-made disasters/accidents. At the national level, the GoI has taken the initiative to create a National Disaster Response Force to deal with natural and man-made disasters.

Salient Gaps and Recommendations On-Site Mass Casualty Management (i) Th e concept of pre-hospital care and mass casualty management in the ﬁ eld is not yet fully developed. (ii) SOPs for on-site mass casualty management of victims are not available. (iii) In the hospital setting, the concept of triage is followed informally, depending upon the clinical acumen of the attending doctor. Triage concepts are required to be developed and followed meticulously. (iv) SOPs regarding specialised care for chemical, biological, radiological, and nuclear (CBRN) casualties, including the use of antidotes, is lacking. (v) Th e number of mobile hospitals and mobile teams available is inadequate. (vi) Th ere is lack of on-site planning in district-level disaster management plans. (vii) Th e medical inventory categorised for natural, man-made and CBRN disasters is also inadequate. No stockpile of medical supplies exists for surge capacity. Proper standardised protocol for emergency procurement and relay, and the concept of supply chain management need to be further developed. ( viii) Th ere is an inadequate number of trained paramedics. Transportation

Downloaded by [University of Defence] at 01:15 24 May 2016 All modes of transport are used in a mass casualty event (MCE) setting — personal vehicles, trucks, tractors, tempos, and even bullock carts. Th e major gaps existing in this area include:

(i) Ambulances are conspicuous by their absence. Presently, basic life-support facilities are grossly inadequate. Th ere is also a lack of co-ordination between the various agencies possessing ambulances; thus, a major part of the aﬀ ected community is not beneﬁ tted. 312 Ú Brig KJ Singh

(ii) Th ere are also very few advanced life-support ambulances. In fact, there is no accreditation system for ambulance services in India. (iii) Th ere is lack of co-ordination and integration amongst the various fi rst responders, including fi re services and the police department. Communication An effi cient communications system is one of the most essential elements in managing MCEs. The disruption of infrastructure in the aftermath of a disaster can aff ect the terrestrial/wired communication system and that has a debilitating impact on all sectors including healthcare delivery. State-of-the-art communications systems are available in the country but their utilisation in the health delivery system is grossly inadequate. Th e communication network between hospitals as it exists today cannot is manage mass casualties and must be over- hauled. Also there is a lack of intra-communication facilities in most of the district/sub-district level hospitals. Training and Education Th ere has been some attempt at training hospital managers in mass casualty management from 1996 onwards through WHO projects. Th e emphasis in these training programmes is that district hospitals should have their own disaster management plans. However, these eff orts are merely a drop in the ocean considering the enormous training load. Th ere is a need to evolve a training module and education curriculum for updating of mass casualty management skills for doctors, nurses, paramedics at the district and sub-district levels, on basic and advanced life-support systems for managing MCEs. Th ere is also a need to evolve standardised training modules for diff erent medical responders/community members at the disaster site. Downloaded by [University of Defence] at 01:15 24 May 2016 Disaster Preparedness of Hospitals A critically important unit for the management of mass casualties, hospital co-ordination and preparedness in the pre-disaster phase enhances the eff ectiveness of their response during disaster situations. In India, diff erent hospitals fall under diff erent administrative set-ups. Th e availability and quality of the medical facilities they off er diff er drastically from urban to rural areas, and from private to government-run hospitals. Therefore, preparedness requires hospital disaster management Medical Management of Earthquake Disasters Ú 313

planning at the hospital level, its development and upgradation, planning at district/State levels, as well as overall regional plans for eff ective management. Community Preparedness Communities need to be trained in dealing with disasters. Th is can be done through the identifi cation of task forces constituted from diff erent community groups like resident welfare associations, vyapar mandals, PRIs, NGOs, local healthcare providers and others, for varied tasks of disaster management. Th e roles of these groups will be incorporated in disaster management plans and will be rehearsed periodically.

International Agencies

Many international agencies are involved in providing assistance to nations in the gamut of activities ranging from prevention to post- disaster programmes. Prominent amongst them are various agencies of the United Nations, World Health Organisation, Asian Development Bank, European Community Humanitarian Oﬃ ce, and Oﬃ ce of Foreign Disaster Assistance (USA).

Role of the Armed Forces

Th e armed forces are requisitioned for all types of disaster management tasks. Th eir response has always been immediate and it has played a vital role in rescue and the provision of prompt relief, especially in remote areas of the country. Th ey may be called upon to provide the following types of assistance:

(i) Communication — setting up telephone and radio communications, along with providing specialist manpower. Downloaded by [University of Defence] at 01:15 24 May 2016 (ii) Provision of medical aid, including medical teams and treatment at armed forces hospitals. (iii) Rescue and relief missions. (iv) Transportation of relief materials by aircraft, ship and land transport to aﬀ ected areas. (v) Establishment of relief camps. (vi) Repair and maintenance of essential services. (vii) Demolition of unsafe structures. 314 Ú Brig KJ Singh

Th e local armed forces authorities are also prompt to comply with the requests of the civil authorities. Formal sanction of the central government is accorded post-facto. Inter-service operations are co-ordinated at the appropriate level.

Quality Improvement

Quality improvement is a useful tool to assess the eﬃ cacy and cost- eﬀ ectiveness of disaster preparedness. A systems approach to components like personnel, facilities, equipment, communications, transportation, stores supply, processes, etc., is necessary for assurance and further improvement in the delivery of disaster relief.

Conclusion

Earthquakes occur suddenly and unexpectedly, causing widespread devastation, injury and loss of life. Disaster management entails a multi-disciplinary and multi-departmental approach to the relief and rehabilitation of the aﬀ ected community. A comprehensive disaster management plan is, in fact, essential for the co-ordinated mobilisation of all the necessary resources. Periodic reviews of the plan and the rehearsal of drills are also necessary to maintain an optimum level of preparedness and to eﬀ ectively manage a disaster situation.

References

Cariappa, M.P. and P. Kanduri. 2003. ‘Health Emergencies in Large Populations: Th e Orissa Experience’, Medical Journal Armed Forces India 59 (4): 286–89. Carter, W.N. 1992. Disaster Managment: A Disaster Managers Handbook. Manila: Asian Development Bank (ADB). Dave, P.K., Shakti Gupta, N.K. Parmar and Sunil Kant. 2001. Emergency Medical Services and Disaster Managment: A Holistic Approach. Delhi: Jaypee. Downloaded by [University of Defence] at 01:15 24 May 2016 Government of Maharashtra. 1986. A Guide to Health Management in Disaster. Mumbai: Directorate of Health Services, Government of Maharashtra. National Disaster Management Authority (NDMA). 2007. National Disaster Management Guidelines: Medical Preparedness and Mass Casualty Management. New Delhi: Magnum Books. Noji, E.K. 1997. Th e Public Health Consequences of Disasters. New York: Oxford University Press. Medical Management of Earthquake Disasters Ú 315

World health Organisation (WHO). 1983. ‘Health Services Organization in the Event of a Disaster’, Scientifi c Publication No. 443. Washington, DC: Pan- American Health Organization (PAHO) and WHO. ———. 1983. ‘Medical Supply Management after a Natural Disaster’, Scientifi c Publication No. 438. Washington, DC: PAHO and WHO. ———. 1989. ‘Guide to Emergency Health Management after National Disaster’, Scientifi c Publication No. 407. Washington, DC: PAHO and WHO. ———. 2000. ‘Natural Disaster: Protecting the Public’s Health’, Scientific Publication No. 575. Washington, DC: PAHO and WHO. Downloaded by [University of Defence] at 01:15 24 May 2016 13 Trauma after Earthquakes Mitigating the Psychosocial and Mental Effects R. Srinivasa Murthy

A ll over the world there is growing awareness of the mental health consequences of disasters. Th ere has been a major rethink on the needs of populations surviving disasters. Th e publication of the Inter-Agency Standing Committee (IASC) Guidelines in February 2007 represents a milestone in this area. Th is is refl ected in the regular inclusion of mental health components as part of relief and rehabilitation eff orts (IASC 2007). Mental health professionals have become part of the teams of professionals working with disaster-aff ected populations all over the world (Murthy 2000). Much like the IASC Guidelines for HIV/AIDS Interventions in Emer- gency Settings, the Task Force on Mental Health and Psychosocial Support in Emergency Settings Guidelines (IASC 2007) reviewed the state of knowledge in this area and identifi ed the key elements for inclusion in emergency responses. Th e fi nal guidelines include the list of recommended steps, specifi cally for mental health/psychosocial care, described in 25 action sheets:

(i) Establish co-ordination of inter-sectoral mental health and psychosocial support. (ii) Conduct co-ordinated assessments. (iii) Initiate participatory systems and processes for monitoring and Downloaded by [University of Defence] at 01:15 24 May 2016 evaluation. (iv) Identify, monitor, prevent, and respond to protection threats and failures through social protection. (v) Apply a human rights framework through mental health and psychosocial assistance. (vi) Identify and recruit suitable staﬀ and engage volunteers who have a deep understanding of local culture. (vii) Enforce staﬀ codes of conduct and ethical guidelines. Psychosocial and Mental Effects of Disasters Ú 317

(viii) Organise orientation and training of aid workers in mental health and psychosocial support. (ix) Prevent and manage mental health and psychosocial problems in staff and volunteers. (x) Facilitate conditions for community mobilisation, ownership and control of emergency response in all sectors. (xi) Facilitate community social support and self-help. (xii) Facilitate conditions for appropriate cultural and religious healing practices. (xiii) Facilitate support for young children (0–8 years) and their care- givers. (xiv) Include specifi c social considerations (safe and culturally appropriate access for all) in the provision of water and sanitation. (xv) Include specifi c social and psychological considerations (safe aid for all, that take into consideration cultural practices and household roles) in the provision of food and nutritional support. (xvi) Include specifi c social considerations (safe, dignifi ed, culturally and socially appropriate assistance) in site planning and shelter provision in a co-ordinated manner. (xvii) Include specifi c social and psychological considerations in the provision of general healthcare. (xviii) Provide access to care for people with severe mental health problems. (xix) Protect and care for people with mental disorders living in cus- todial settings. (xx) Learn about and, where appropriate, collaborate with local, indigenous and traditional healing systems. (xxi) Minimise harmful use of alcohol and other substances. (xxii) Provide access to formal and non-formal education. (xxiii) Organise psychosocial support in educational settings. (xxiv) Provide information to the aff ected population on the emer-

Downloaded by [University of Defence] at 01:15 24 May 2016 gency, relief eﬀ orts and their legal rights. (xxv) Provide access to information about constructive coping methods.

Th ere are three aspects of the guidelines that are worth emphasising. First, the approach to organising psychosocial interventions is considered in detail in the initial worksheets. Second, there are worksheets that focus on the core psychosocial and mental health interventions. Third, there are measures to include mental health 318 Ú R. Srinivasa Murthy

components in nutrition, education, shelter, media, and other activities in emergencies. Th e publication of the guidelines and related fact- sheets can be expected to bring about greater recognition of the need for better co-ordination of the eff orts of diff erent agencies, increased access to services and new knowledge to guide future interventions. Th e scope of this article will be to review the psychosocial/mental health needs in the diff erent stages of disasters, and to support these needs with the experience of some recent disasters in India.

Psychosocial Interventions in Different Phases of an Earthquake

Individuals need support in trying to minimise the eff ects of an earthquake in three respects. First, they need help to understand the normalcy of the changes they experience in their feelings, bodily functions like sleep and, more importantly, the behavioural changes. Second, they need help in mastering these changes by using corrective behaviour and coping strategies. Th ird, they need help to rebuild their lives in the least disruptive manner, to minimise the negative eff ects of the disaster. Efforts should be directed towards providing support in all three of these stages. In India, following the Marathwada and Gujarat earthquakes, eff orts have been made to try and understand the impact of an earthquake on the mental health of the aff ected population, as well as to develop interventions to mitigate these eff ects. A large number of manuals, training programmes and community-based programmes have been developed and evaluated.

Predisaster Phase

Th is is very important for earthquake-prone areas. Usually, preparation

Downloaded by [University of Defence] at 01:15 24 May 2016 in the pre-disaster phase areas includes building of earthquake-resistant houses and improving communications, etc. It is in this phase that mental health professionals should work with other officials and voluntary organisations to prepare the community and individuals for the impact of a disaster. Th is would include participating in meetings on disaster preparedness, sharing information about stress and its eﬀ ects on individuals, emphasising the need to keep the family as a unit during relocation (avoid separating men from women and children), keeping the community together when large groups are relocated temporarily, Psychosocial and Mental Effects of Disasters Ú 319

and teaching adults to maintain regular daily routines and methods of handling disruption of psychological functions by ensuring regular rest, sleep and organising group interactions. It would also be valuable to identify at this stage itself some volunteers who can be trained in providing psychosocial help. Th ese people will not only be able to provide help but also become communication agents between the affected community and the helping agencies. Ideally, there should be one key individual for every 25–30 families. Th e manuals that were prepared in response to the Orissa Cyclone (1999) are suitable for their training (Kishore Kumar et al. 2000; Srikala et al. 2000).

Initial Phases of Disaster (1–4 weeks)

Th is is a period of intense disruption in the lives of the aff ected individuals. A lot of immediate relief activities, in terms of food, clothing, shelter, security, and medical care are provided by a variety of groups — both governmental and non-governmental. Th ere will be people who would have lost relatives, experienced injuries, been separated from the family and the chief need is for crisis intervention. Often the psychological aspects are not given importance at this stage. Th e lack of sensitivity to psychological problems leads to various types of other complications like non-acceptance of help and not following directions, even expression of hostility. Mental health professionals will have to reach the people themselves, as well as through a variety of persons working with the disaster-aff ected population. A special group that should be sensitised are the medical personnel as they see most of the distressed population. Th e hospital staff often complain about the crying of patients and patients wanting to share their experiences of the earthquake again and again with them. Th e healthcare staff , especially nurses, can view this process of sharing as a waste of their time, and not understand that this as an important emotional need. An orienta-

Downloaded by [University of Defence] at 01:15 24 May 2016 tion among them to understand the need to share the experience of the disaster again and again led to a better understanding of behaviour and greater sympathy for the aff ected population. Similarly, encouraging family members to stay with the physically ill is also helpful in decreasing the anxiety of the aff ected persons and their families. During this phase all aff ected persons live in a world of ‘impending disaster’ and tend to over-react to all types of uncertainties. At this stage mental health professionals should reach out to the leaders of voluntary organisations for two reasons: one is to emphasise the needs of the staff working with 320 Ú R. Srinivasa Murthy

the disaster for emotional support, and the other is to prepare them to understand the emotional needs of the population, along with other physical needs. Translations of simple and clearly written self-help material, emphasising how individuals can take care of themselves, is useful. Some of the messages of self-help that can be shared with the general population are relaxation and meditation; sharing of feelings; resuming family routines; rituals; restarting activities; and participating in relief and rehabilitation. Th e goal should be to instill in the population a sense of empowerment (Srikala et al. 2000; Dave et al. 2002a; Dave et al. 2002b; Sekar et al. 2002; Narayana et al. 2002; Indian Red Cross Society 2003a, 2003b and 2003c).

First Few Months (1–6 months)

Th is is the most important phase for mental health interventions. Th e initial phase of intense relief eff orts usually starts receding from the agenda of the help providers by this time. Often the fi rst six weeks are described as the ‘honeymoon’ period. Soon after this, individuals have to face the reality of rebuilding their lives and coming to terms with their losses. Psychological reactions are common in this phase; studies report that nearly everyone experiences anxiety, depression, panic and other related symptoms during this phase. Mental health professionals will always be too few to meet the needs of a large population. Th is is the period, therefore, for sharing caring skills with all the help providers working in the community. It is tempting to think of a separate mental-health worker, but this is best avoided except at the supervisory level. Instead, the focus should be on identifying all the care providers and orienting them to mental health care. Some of the key persons are primary healthcare personnel, school teachers and developmental workers. Mental health professionals should become the trainers of these groups and provide regular support to them as they try to inte- Downloaded by [University of Defence] at 01:15 24 May 2016 grate mental health care with all the other activities. It is also important not to present emotional needs as deviance, as this approach would stigmatise individuals and lead to a denial of help. Th e eff ort should be to provide help as a natural need of the aff ected population and to everyone rather than to ‘ill’ persons only. Again the eff ort should be to strengthen methods of understanding the disaster at the level of family and community. An additional role of mental health professionals will be to function as referral support to those persons who require a greater Psychosocial and Mental Effects of Disasters Ú 321

degree of help and those found not responding to the initiatives of the community-level help providers (Murthy 2000; Narayana et al. 2003).

Indian Experiences in Disaster-related Mental Health Care

India has experienced a large number of disasters, natural as well as man-made, in the last 25 years. Th e fi rst disaster studied for its mental health eff ects was the fl oods in Andhra Pradesh in the late 1970s. Th e fi rst systematic and prospective study of mental health eff ects of a disaster followed the death of about 70 children in the circus fi re in Bangalore in 1981 (Narayanan et al. 1987). Th e Bhopal disaster in December 1984 was the fi rst major disaster in India to be studied prospectively and systematically. Th e fi rst assessments of the mental health eff ects were made in the fi rst week of February 1985 (two months after the disaster) by psychiatrists visiting the aff ected population at home and examining those attending the general medical facilities. Subsequently, a mental health intervention programme and a fi ve-year annual survey for mental disorders was undertaken by the Indian Council of Medical Research (ICMR), New Delhi (Murthy and Issac 1987; Murthy et al. 1987; Bharucha and Bharucha 1987; Sethi et al. 1987; Basu and Murthy 2003). Following the demolition of the disputed structure at Ayodhya on 6 December 1992, there were riots in Bombay in January 1993, followed by bomb blasts in March 1993. Approximately 1,500 people died in the riots and the blasts, and 30,000 students dropped out of school after the riots. Motivated social workers, both lay and professional, associated with institutions like the Tata Institute of Social Sciences (TISS), Mumbai, College of Social Work Nirmala Niketan, Mumbai, and NGOs carried out psychosocial counselling individually and in groups (Joseph 2000). Th e supercyclone that hit the Orissa coast in 1999, was one of the most devastating in terms of the damage suff ered by the people, and the destruction of animals and

Downloaded by [University of Defence] at 01:15 24 May 2016 the environment. Over 15 million survivors of the disaster faced this massive damage with an extremely limited mental health infrastructure to provide care (Action Aid 2002; Kanchan and Gwynn 2000). In- novative approaches to mental health care were developed by involving community resources (Kishore Kumar et al. 2000; Srikala et al. 2000). Th is was a major breakthrough in the development of psychosocial care programmes in the country. An evaluation, two years after the disaster, demonstrated the beneﬁ ts of psychosocial interventions for individuals, families and the community (Murthy et al. 2003: 103). 322 Ú R. Srinivasa Murthy

Th e approach was innovative in that it utilised community resources and demonstrated that psychosocial care is feasible even in situations of limited professional resources. Th e communal riots of Gujarat in February 2002, from a disaster-related mental health care perspective, are important for a number of reasons. Th e riots were man-made and it is well known that in such a situation the mental health eﬀ ects are greater. Th e majority of the survivors were from the minority community and there were fears and barriers to utilising all the available mental health professional help. Th e psychosocial needs were addressed innovatively by utilising community-level volunteers. Another very important development has been the priority given to psychosocial care by a number of major voluntary organisations like Action Aid, Oxfam, Cooperative for Assistance and Relief Everywhere (CARE), and Self- employed Women’s Association (SEWA) to name a few. Another major development, possibly the most important, is the initiative of the American Red Cross through the Indian Red Cross to give disaster-related mental health high priority. A number of simple, practical and culturally applicable educational and self-care materials have become available. Th ese materials simultaneously address the survivors, community-level volunteers and professionals (Indian Red Cross Society 2003a, 2003b, 2003c; Narayana et al. 2003). Training programmes and the development of human resources is another component. United Nations Development Programme (UNDP) initiatives (with USAID) to develop ‘more than 40,000 volunteers will receive information and equipment for disaster-response decision-making and more than 250,000 disaster management teams will be created’ is another important development (Th e Hindu, 2003; ICMR 2003).

Experience of Working with Earthquakes in India Downloaded by [University of Defence] at 01:15 24 May 2016 Uttarkashi Earthquake

On 20 October 1991 the Garhwal region of Uttar Pradesh (UP) was rocked by an earthquake of magnitude 7.7 on the Richter scale. Eight hundred people died, 10,000 houses collapsed and 30,000 were damaged in varying degrees. Immediate interventions were carried out by the Indian Army, the Government of UP, the local administration and NGOs. Unfortunately, there was no organised mental health intervention and Psychosocial and Mental Effects of Disasters Ú 323

no eﬀ ort on the part of institutions (governmental or non-governmental) to initiate psychological rehabilitation.

Marathwada Earthquake

Th e Marathwada earthquake was a major earthquake of magnitude 6.4 on the Richter scale which struck the area on 30 September 1993. Approximately 8,000 people died and 14,000 were injured, and over 34,000 buildings suff ered varying degrees of damage. Th e earthquake resulted in the total destruction of 67 villages, while another 886 villages experienced damage to their houses. Th e total aff ected population was around 170,000. Th e disruption of electricity, incessant rains and the aftershocks made rescue operations arduous. Th e impact of the disaster was thus multi-dimensional. Of those living close to the epicentre of the earthquake, one-third suff ered human loss (i.e., death of a family member), and in most of these families the losses were multiple. One-third of the adults were trapped in the debris, but most of them escaped in a short time. As many as 58 per cent of the adults were exposed to the death/near death of others and the sight of mutilated bodies. Nearly 36 per cent of adults were exposed to body handling. Almost 100 per cent of the sample was rendered homeless, and 40 per cent had lost all their means of livelihood (Agashe and Phadke 1994; Acharya 2000; TISS 1994a, 1994b; TARU 1995; Sharan et al. 1994; Pande 1994; Pande et al. 2000a, 2000b; Government of Maharashtra 1994; Gandevia and Chitale 1995; Gandevia 2000; Hegde 1995). Th e ICMR set up a Centre for Advanced Research on Mental Health Consequences of Earthquake Disaster with Special Reference to Mental Health for a period of fi ve years at the Maharashtra Institute of Mental Health (MIMH) at Pune. A fi ve-year prospective study was completed. The findings of this study showed that psychological morbidity in the disaster-aff ected population was at least twice that in the control

Downloaded by [University of Defence] at 01:15 24 May 2016 population ﬁ ve years after the disaster. Th e study involved the assessment of 4,096 individuals, whereas the control group consisted of 3,525 people residing at Parner (Ahmednagar), who had not experienced the earthquake. Th e ﬁ rst phase of assessment was carried out during one-

and-a-half to two-and-a-half years, post-disaster. A follow-up study was undertaken four-and-a-half to fi ve years after the disaster. Th e prevalence of psychiatric morbidity in the disaster-aff ected group was 13.9 per cent as against 6.8 per cent in the control group. In the aff ected group, 21.5 per cent of the adult males and 15 per cent of the women had received 324 Ú R. Srinivasa Murthy

psychiatric diagnosis (comparative values for the control population were 13 and 5 per cent, respectively). Alcohol-related problems accounted for the excess prevalence of psychiatric disorders in males in both the aff ected and control groups. In both genders and in both groups, the most common diagnosis was ‘other reactions to severe stress’ (16 per cent). Th is is equivalent to adjustment disorders that are caused due to the catastrophic nature of stressful events. Th e other disorders more frequently reported in the aff ected group were major depressive disorders (9 per cent) and post-traumatic stress disorder (PTSD) (1.3 per cent). In children, sleep disturbances were the predominant problem, in addition to ‘other reactions to severe stress’. Risk factors for developing psychiatric disorders were injury, occurrence of disaster deaths in the family, the experience of being trapped inside rubble, and dissatisfaction with social support in the case of adults and the occurrence of disaster deaths in the family in the case of children. A trend was noted towards an increase in psychiatric morbidity as the severity of exposure increased in both adults as well as in children. A clustering of cases was noted in the families, and this was more prominent in the worst-aff ected villages. When the sample was followed up three years later (almost fi ve years post-disaster), there was signifi cant remission in psychiatric morbidity in that 69 per cent males, 71 per cent females, 83 per cent boys and 98 per cent girls no longer had a psychiatric diagnosis. Variables promoting remission were satisfaction with social support, occurrence of desirable life events and absence of disaster injuries. Th e desirable activities include cultural and religious activities like participation in bhajan, kirtan, waari (pilgrimage), and other traditional festivals (Pande 1994; Pande et al. 2000a, 2000b). Immediate relief measures were carried out effi ciently by the Gov- ernment of Maharashtra (GoM), the Indian Army and by local NGOs. Mental health needs were in fact recognised by the State government in its proposal for the Maharashtra Earthquake Rehabilitation Programme. Downloaded by [University of Defence] at 01:15 24 May 2016 Th e document stated:

the fragile mental health of affected women and children needs to be strengthened through a programme of psychological rehabilitation. Coun- seling groups could be the basis for the entry point to gain the conﬁ dence of these people (GoM 1994: 3).

Interventions for psychosocial care were carried out by a number of groups. For the ﬁ rst time in the history of disasters in India, mental Psychosocial and Mental Effects of Disasters Ú 325

health professionals reached the disaster area to provide care on the third day of the disaster, in contrast to the Bhopal disaster where psychiatrists visited two months after the disaster (Pande 1994). Immediate assistance and the delivery of mental health services was provided by agencies in the hospitals and relief camps. Th e counselling of groups and individuals, by the government, NGOs and professionals, took place at hospitals housing the injured during the immediate post-impact phase itself. Medication and counselling were carried out at Latur and Osmanabad by psychiatrists attached to government hospitals, MIMH (Pune), the staff of mental hospitals of the State and NGO professionals, and other motivated mental health personnel. MIMH planned the delivery of services in three phases: immediate intervention (fi rst month post-disaster); outreach counselling services (2–6 months post-disaster) and creating infrastructure locally in collaboration with the government, NGOs and the private sector. A core team, comprising psychiatrists, psychiatric social workers, clinical psy- chologists, psychiatric residents and nursing counsellors was planned. Th e data collectors were trained mental health workers and they also participated in the delivery of such care. Th e core team imparted intensive training to the personnel involved at each stage, defi ned the objectives and the nature of services to be off ered and explained the theoretical orientation. Role playing, video materials, group discussions and didactic lectures were used. Th e emphasis was on developing listening and communication skills. Immediate interventions were targeted at the hospitalised injured survivors. Th e objectives of this stage were to detect the psychopathology early on itself, facilitating grief, providing support, giving an opportunity for venting and preparing for discharge. Outreach counselling services were targeted towards Zone ‘A’, the area of highest destruction. As a result of this, 500 individuals, including children, received some kind of psychological inputs. Th e mental health workers involved were debriefed to protect the people from stress and improve their performance. Th is gave the people an opportunity to Downloaded by [University of Defence] at 01:15 24 May 2016 express and ‘unload’ their feelings. In addition to providing these services, MIMH carried out sensitisation and training programmes for peripheral health workers, NGO workers and private practitioners. Services were presented as stress, crisis and counselling services rather than as psychiatric services. However, the team did not quantify the impact of the services. Th e eff ort showed two important points: that a large number of disaster-aff ected people require mental health services and that, in such a situation, these services are acceptable and feasible in such a situation (Pande et al. 2000a, 2000b). 326 Ú R. Srinivasa Murthy

Gujarat Earthquake

On 26 January 2001, at approximately 8:46 am, an earthquake measuring 7.7 on the Richter scale occurred in the State of Gujarat, in western India. Th e US Geological Survey placed the epicentre of the earthquake 69 km north-east of the city of Bhuj. According to the GoI, the earthquake and its aftershocks aff ected nearly 15.9 million people, resulting in more than 20,000 deaths, 167,000 injuries, and the destruction of over 1 million homes (Desai et al. 2002; Bhadra 2001; Mehta et al. 2001a, 2001b; Vankar and Mehta 2004; Sengupta 2004: 119–27). A survey of distress among hospitalised trauma patients at the Civil Hospital, Ahmedabad showed that 131 of the 151 patients had scored four or more (indicating signifi cant distress and ‘caseness’) on a mental health screening questionnaire. Similarly, in another study, 45.5 per cent of the patients attending primary health centres appeared to have probable psychological distress. Higher symptom scores were associated with more disability. Th is study also found psychological morbidity in 80 per cent of the people who had suff ered paraplegia and amputation, with an associated high family burden (Bhadra 2001). A study of teachers in earthquake-aff ected Kachchh, carried out 6–8 months after the earthquake, found 34 per cent of them were suff ering from PTSD (Mehta et al. 2001a). Th e common symptoms reported are presented in Table 13.1. Being a woman emerged as a risk factor for PTSD — 57.6 per cent of the PTSD patients were women, while 43.8 per cent of the teachers who did not have PTSD were women. Th e most signifi cant stress factors were the presence of a pre-disaster stress and the death of or injury to a child in school. Several known risk factors like death in the family or injury to oneself were not associated with PTSD. High-school adolescents studying in standards X and XI who were exposed to earthquake trauma were also studied. Th ey were drawn from two boys schools, one in Bhachau and the other in Rapar — the areas

Downloaded by [University of Defence] at 01:15 24 May 2016 worst aff ected in Kachchh (Mehta et al. 2001b). Out of 126 adolescents, 25 (20 per cent) had clinically signifi cant PTSD features. Th e commonest manifestations included impaired concentration, lack of pleasure, being easily startled, avoidance of trauma reminders, survivor guilt, apprehen- sion of recurrence, and regressive behaviour. Th ese manifestations were present in more than 40 per cent of the adolescents who were identifi ed as suff ering from PTSD (i.e., 20 per cent of the total sample). A lon- gitudinal study of PTSD symptoms indicated a gradual decline with the passage of time and stabilisation of symptoms of about 21 per cent of them in 10 months post-earthquake (Vankar and Mehta 2004). Psychosocial and Mental Effects of Disasters Ú 327

Table 13.1: Features of PTSD

Manifestation PTSD teachers (%) Recurrent memories, thoughts, images 35.9 Nightmares 14.1 Reliving experience 33.0 Reminders 38.9 Autonomic hyperactivity 34.1 Avoidance of talks/feelings 48.9 Avoidance of activities/situations 47.1 Loss of memories 28.2 Lack of interest 55.3 Feeling of estrangement 48.3 Emotional numbing 24.7 Foreshortened future 52.4 Sleep disturbance 42.9 Anger outbursts 38.2 Lack of concentration 41.7 Hyperalertness 72.9 Easily startled 52.3 Source: Mehta et al. (2001a).

UNICEF developed a large-scale intervention programme through schools (Sengupta 2004: 119–27). Th e strategy adopted under this project focused on integrating psychosocial support with the restoration of primary education system in Kachchh district. Th e project was carried out in Kachchh district, covering the blocks of Bhuj, Anjar, Rapar, and Bhachau. Th e key activities were capacity building among teachers to provide psychological support to students and identifying children who were at risk of developing long-term eﬀ ects of trauma. Th e service providers, in this case the teachers, had themselves borne the impact of trauma. Most of the teachers, like other members of the community, had lived through the shock of the earthquake and the repeated aftershocks, and had been witness to widespread destruction and

Downloaded by [University of Defence] at 01:15 24 May 2016 damage. Around 1,000 children had died in Kachchh alone. Moreover, the teachers were also seen as a resource by the government for many other post-earthquake activities and it was thus an uphill task for them to also be able to focus on psychosocial interventions. In Khavda region, teachers had to walk about 15 kilometres one way to reach their schools. Sengupta summarises some of the problems faced while training teachers:

Th e training process for teachers was challenged in various ways. For the ﬁ rst training programme, some teachers travelled with their families, as 328 Ú R. Srinivasa Murthy

they did not want to leave their family alone post the earthquake. Other teachers were worried that they had taken a grave risk in being present for the training as living in open areas they had to leave their belongings in the open. Some others said that it was a relief to be able to travel away from the problems of daily existence, even though for a few days (2004: 180).

A two-day orientation and sensitisation workshop held with 75 teachers at Samkhyali helped in designing the programme. It was conducted by the UNICEF team, psychiatrists and an NGO called Chetna. Th e response of the teachers reaffi rmed the need for a psychosocial programme and that the schools would be good starting points. After the master trainers received their training, suitable material for distribution was developed and training organised for about 174 primary school teachers at Ambaji. Th ere was a post-training debrief on each day to understand the diffi culties faced. Th e instructors were given the role of providing the bridge between the psychiatrists’ technical information and the teachers. Th is was done by interspersing the content with icebreakers like folk songs, metaphors and anecdotes. Similar training programmes were organised for another 100 teachers at Mandvi, and in total 538 teachers were trained at Vandhai and Mandvi. Th e process of training was participatory and started with introductions to each other and then gave space for teachers to express what they had experienced during the earthquake and their feelings around the trauma. Th is session was called anxiety sharing, and though initially resisted by the participants it became the part they most remembered about the training. Th e session, however, did bring up diffi cult experiences of personal trauma for some participants and a number of teachers broke down while talking about their personal losses. Training further focused on the immediate impact of trauma on people, especially children. Teachers were told about normal responses and the kind of responses that might need help. Th e symptoms of PTSD were talked about, listening skills were focused upon, and some classroom activities Downloaded by [University of Defence] at 01:15 24 May 2016 were defi ned. Th e listening skills that were discussed seemed to be more appropriate to help adults share their experiences. Th e group was also able to rehearse these skills through role-play. Th e processes of group intervention and working with children through activities were mentioned. Th e training of these skills was interspersed with some folklore-based songs, and some other activities that broke the tedium of too much information. Th ese became extremely popular with the group and achieved two objectives, one to get the group to work as a team, and second, to help distract them from their immediate worries. One key Psychosocial and Mental Effects of Disasters Ú 329

strategy that got reinforced through the whole process was distraction and helping people get distracted.

The Relief and Rehabilitation Phase

By this time the disaster-aﬀ ected population and their needs start to recede from the headlines of the media as well as the priority list of the help providers. Th e rebuilding of the disrupted lives of individuals and communities, however, is a continuous process. Th e attempt at this stage is to try and rebuild the community’s institutional structures such that they can take over charge of activities from the outsiders. Th e media are an important part of disaster relief activities. It is the media that brings to attention the human side of the disaster. However, it is commonly seen that the media are relatively insensitive to people’s psychological needs. Often they refer to psychological distress in derogatory terms — mental health needs are projected as ‘people going mad’ or ‘people committing suicide’. Th is happens due to a lack of correct information. Mental health professionals have to work with the media to help them understand emotional reactions, the diﬀ erent phases of such reactions and the correct way to project these issues.

Documentation

Th ere is a lack of documentation of the many innovative experiences of mental health care provided to diff erent disaster-aff ected populations in the country. It appears that everyone is busy doing something and in this process the experiences do not get written up and the core lessons are not identifi ed. An important role for mental health professionals will be to build up documentation of their ongoing activities of relief and rehabilitation. Downloaded by [University of Defence] at 01:15 24 May 2016 National-Level Initiatives

Over a quarter-century of mental health initiatives in the area of disaster care in India has seen the gradual growth of awareness and the development of speciﬁ c programmes. Th e other developments of importance are the two national conferences on Psychosocial Consequences of Disasters held in 1993 and 1997 at the National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore (see NIMHANS 1997). 330 Ú R. Srinivasa Murthy

Th e India Disaster Report was published (Parasuraman and Unnikrishnan 2000) and a special issue of the Indian Journal of Social Work was devoted to the topic (Jaswal 2000). Diﬀ erent groups of professionals have also presented their concerns relating to disasters, for instance on social work training (Murthy 2000; Nadkarni 1991; Pereira 2000; Parasuraman and Acharya 2000), role of mental health professionals (Murthy and Shetty 1995), cultural aspects (Patel 2000), special issues of women (Lohokare and Davar 2000) and community development (Verma and Bhushan 1994). Th e National Institute of Advanced Studies (NIAS), Bangalore organises annual courses on disaster management for IAS oﬃ cers and this course includes disaster mental health (Rajagopal and Chari 2003). National-level institutions like NIMHANS in Bangalore, the Tata Institute of Social Sciences (TISS) in Mumbai and Institute of Human Behavior and Allied Sciences (IHBAS) in Delhi are actively involved in services, the training of personnel and research. Th e progress of the mental health programme in India, namely, its decentralisation, integration with general healthcare and the de-professionalisation of services, are supportive of the approaches developed for disaster mental health care. An awareness of the psychosocial aspects of other forms of ‘disasters’ is slowly emerging. Investigators have focused on the impact of disasters and emergencies on gender violence (Ali 1997; Purewal and Ganesh 2000), political unrest (Ali and Jaswal 2000), displaced persons (Th ukral 1996), drought, poverty and famines (Sen 1981). However, a detailed study of mental health aspects and the development of appropriate interventions in these latter group of disasters is a task that is yet to be undertaken. Against the many positive developments, that the integration of mental health as part of total disaster care plans has still a long way to go can be seen by two examples. A recent book on disaster management (Lewin 2003; 161–70) includes no separate section on the mental health aspects, in spite of the fact that the training programme on which the

Downloaded by [University of Defence] at 01:15 24 May 2016 book is based included psychosocial aspects in its coverage. Th e chapter on health and psychosocial consequences has only one line on this subject. Th e Tsunami experience of 2004, has taken psychosocial interventions to a higher level of recognition and support. Th ere has been both an increase in the number of centres working with disaster mental health care as well as the use of more sophisticated approaches along with the evaluation of the outcome of interventions (World Health Organisa- tion [WHO] 2006, 2008). A national-level draft policy has also been Psychosocial and Mental Effects of Disasters Ú 331

developed in the last two years (National Disaster Management Authority 2007).

Lessons Learnt for Psychosocial Care

Based on these experiences, some tentative conclusions can be drawn at this stage. First, the mental health needs of a disaster-aff ected population are now an accepted part of the agenda of disaster care. Second, there is still a strong component of not acknowledging fully the personal suff ering and losses of the aff ected population. Th ird, community participation in post-disaster rebuilding seems a goal largely unattended. Larger political and administrative changes need to take root and grow for true community participation. Fourth, we do not have a framework for rehabilitation with a long-term perspective. Fifth, it is well recognised that disaster mental health care cannot occur only with outside professionals or resources. Most of the care has to come from within the community. Th e need, therefore, is for all professionals in general, and mental health professionals in particular, to develop methods of care and intervention that can be adopted by the community. Mental health professionals’ involvement in disasters in India has been active only in the last 20 years. Th e eff orts of these professionals have been in creating awareness about the impact of disasters on the mental health of the aff ected populations among the general public and policy-makers; providing direct care to the aff ected populations; training primary care doctors, health workers and volunteers; supporting the initiatives of voluntary organisations and research.

Conclusions

Psychosocial care is an essential aspect of post-disaster response in preventing the adverse eﬀ ects of such events. Th is has been rmlyﬁ

Downloaded by [University of Defence] at 01:15 24 May 2016 established over the last two decades of experience in caring for the survivors of disasters. Th e emotional reactions witnessed among the survivors of an earthquake are a normal reaction to an abnormal situation; they are directly related to the intensity of the disaster experienced. Very often it is biological changes that determine the occurrence of these symptoms. Post-earthquake events and developments are important in achieving a full recovery. Everyone who goes through an earthquake needs psychosocial care. It is every survivor’s right. Moreover, early care leads to better outcomes and can be provided by a variety of people. 332 Ú R. Srinivasa Murthy

Past experiences show that it is entirely possible to develop and implement a programme of psychosocial care. Th ere lies a big advantage in extending psychosocial care to the survivors of an earthquake — besides the improvement in mental health of the population experiencing excess stress, the general population too will become sensitive to the overall mental health needs, as apparently ‘normal’ people develop and recover from psychological symptoms resulting from the disaster, and they do this better when help is provided. Th is can also result in looking at psychological problems as a part of life. In conclusion, a better understanding of the mental health aspects of disasters is developing rapidly all over the world. Indian mental health professionals have contributed to these eff orts. Mental health professionals can bring the best of their skills and knowledge to benefi t the aff ected population.

References

Acharya, N. 2000. ‘Double victims of Latur Earthquake’, Th e Indian Journal of Social Work 61 (4): 521–26. Action Aid. 2002. A Programme for Community Care: Snehakarmis following Orissa Supercyclone. Bhubaneswar: Action Aid. Agashe, M.M. and S.S. Phadke. 1994. Proceedings of Symposium on the Health Consequences of the Marathwada Earthquake Disaster. Pune: Maharashtra Institute of Mental Health (MIMH). Ali, N. 1997. ‘Study of Political Unrest: Its relation with Mental Health in Low Income Urban Households in Srinagar (Jammu and Kashmir)’. Unpublished MA thesis, Tata Institute of Social Sciences (TISS), Mumbai. Ali, N. and S. Jaswal. 2000. ‘Political Unrest and Mental Health in Srinagar’, Th e Indian Journal of Social Work 61 (4): 598–618. Basu, A.R. and R. Srinivasa Murthy. 2003. ‘Disaster and Mental Health — Revisiting Bhopal’, Economic and Political Weekly 38 (11): 1074–82. Bhadra, S. 2001. Review of Psychosocial Support Programmes in the Rehabilitation of the Earthquake Survivor. New Delhi: Oxfam India. Bharucha, E.P. and N.E. Bharucha. 1987. ‘Neurological Manifestations among Downloaded by [University of Defence] at 01:15 24 May 2016 those Exposed to Toxic Gas at Bhopal’, Indian Journal of Medical Research 86 (supplement): 59–62. Dave, A.S., K. Sekar, S. Bhadra, G.P. Rajashekar, K. Kishore Kumar and R. Srinivasa Murthy. 2002a. Riots: Psychosocial Care for Individuals. Bangalore: Books for Change. Dave, A.S., K. Sekar, S. Bhadra, G.P. Rajashekar, K. Kishore Kumar, P. Beena and R. Srinivasa Murthy. 2002b. Riots: Psychosocial Care for Children Surviving the Riots. Bangalore: Books for Change. Psychosocial and Mental Effects of Disasters Ú 333

Desai, N.G., D.K. Gupta, P.C. Joshi, R.A. Singh, T.B. Singh, M. Lal, and A. Kumar. 2002. Mental Health Aspects of the Earthquake in Gujarat. New Delhi: Indian Council of Medical Research (ICMR). Gandevia, K. 2000. ‘Killari: Psychosocial Health of a Village Ravaged by an Earthquake (case studies)’, Th e Indian Journal of Social Work 61 (4): 652–63. Gandevia, K. and V. Chitale. 1995. ‘Psychosocial Assessment of the Marathwada Earthquake Victims’, in S. Parasuraman (ed.), Organisation and Administration of Relief and Rehabilitation following Marathwada Earthquake. Mumbai: TISS. Government of Maharashtra (GoM). 1994. Proposal for Maharashtra Earthquake Rehabilitation Programme. Mumbai: Government of Maharashtra. Hegde, D. 1995. ‘Physical and Psychological Health of the Earthquake Victims in an Agrarian Society: A case study of village Killari’. Unpublished MA Pro- ject Report, TISS, Mumbai. Th e Hindu. 2003. ‘$4 million grant for disaster management’, 5 September. Indian Council of Medical Research (ICMR). 2003. Mental Health Studies of Bhopal Disaster. New Delhi: ICMR. Indian Red Cross Society (IRCS). 2003a. Communities in Disasters: From Crisis to Recovery. New Delhi: IRCS. ———. 2003b. Psychological First Aid. IRCS: New Delhi. ———. 2003c. Share Your Feelings. IRCS: New Delhi. Inter Agency Standing Committee (IASC). 2007. IASC Guidelines for Mental and Psychosocial Support in Emergencies. Geneva: IASC. Jaswal, J. 2000. ‘Editorial: Disasters and Mental Health’, Th e Indian Journal of Social Work 61 (4): 521–26. Joseph, H. 2000. ‘Salokha: A Response to Communal Riots, a Social Disaster (case studies)’, Th e Indian Journal of Social Work 61 (4): 664–74. Kanchan, A. and I. Gwynn. 2000. ‘The 1999 Supercyclone in Orissa’, in S. Parasuraman and P.V. Unnikrishnan (eds), India Disasters Report: Towards Policy Initiative. New Delhi: Oxford University Press. Kishore Kumar, K., C.R. Chandra-Shekar, P.C. Choudhury, R. Parthasarathy, S. Girimaji, K. Sekar, and R. Srinivasa Murthy. 2000. Psychosocial Care for Community Level Helpers. Bangalore: Books For Change. Lewin, S. 2003. Health and Psychosocial Consequences of Disasters — A Bangalore

Downloaded by [University of Defence] at 01:15 24 May 2016 Response. Bangalore: National Institute of Advanced Studies (NIAS). Lohokare, L. and B. Davar. 2000. ‘Women in Disasters and Mental Health’, Th e Indian Journal of Social Work 61 (4): 565–80. Mehta K., G.K. Vankar, Navneet Johri, Amul Patel, Sudeep Shroﬀ , and Kamlesh Dave. 2001a. ‘Post-traumatic stress among teachers in earthquake aﬀ ected Kachchh’. Paper presented at Annual Conference of Indian Psychiatric Society, Kolkata. Mehta, K., G.K. Vankar, Bharat Panchal, Rajesh Kumar, Amul Patel and Kamlesh Dave. 2001b. ‘Post-traumatic Stress among Adolescents in Earthquake 334 Ú R. Srinivasa Murthy

Affected Kachchh’. Paper presented at Annual Conference of Indian Psychiatric Society, Kolkata. Murthy, R. Srinivasa. 2000. ‘Disaster and Mental Health: Responses of Mental Health Professionals’, Th e Indian Journal of Social Work 61 (4): 675–92. Murthy, R. Srinivasa and H. Shetty. 1995. ‘Mental Health Care for Disaster aff ected Population in India’, in S. Parasuraman (ed.), Organisation and Administration of Relief and Rehabilitation following Marthwada Earthquake. Mumbai: TISS. Murthy, R. Srinivasa and M.K. Issac. 1987. ‘Mental Health Needs of Bhopal Disaster Victims and Training of Medical Offi cers in Mental Health Aspects’, Indian Journal of Medical Research 86 (supplement): 51–58. Murthy, R. Srinivasa, M.K. Issac, C.R. Chandrasekar, and A.V. Bhide. 1987. Bhopal Disaster — Manual of Mental Health Care for Medical Offi cers. Bangalore: ICMR Centre for Advanced Research in Community Mental Health, NIMHANS. Murthy, R. Srinivasa, N. Kar, K. Sekar, S. Swami, V. Mishra, and U. Daniel. 2003. Evaluation Report on Psychosocial Care of Survivors of Super Cyclone in Orissa. Bhubaneswar: Action Aid. Nadkarni, V.V. 1991. ‘Developing Curriculum in the Area of Disaster Manage- ment’, in S. Bharat and M. Desai (eds), Research on Families with Problems in India: Issues and Implications (Vol. I). Mumbai: TISS. Narayana, R.L., A.S. Dave, K. Sekar, K. Kishore Kumar, and R. Srinivasa Murthy. 2002. Riots: Psychosocial Care for Women Surviving the Riots. Bangalore: Books for Change. Narayana, R.L., R. Srinivasa Murthy and P. Daz. 2003. Disaster Mental Health in India: Monograph. New Delhi: American Red Cross and Indian Red Cross. Narayanan, H.S., K. Sathyavathi, G. Nardev, and S. Th akrar. 1987. ‘Grief Reactions among Bereaved Relatives following a Fire Disaster in a Circus’, NIMHANS Journal 5(1): 13–21. National Disaster Management Authority (NDMA). 2007. National Disaster Management Guidelines — Psychosocial Support and Mental Health Initiatives. New Delhi: Government of India. National Institute of Mental Health and Neurosciences (NIMHANS). 1997. Report on National Workshop on Psychosocial Consequences of Disasters. Bangalore:

Downloaded by [University of Defence] at 01:15 24 May 2016 NIMHANS. Pande, N.N., S.S. Phadke, M.S. Dalal and M.M. Agashe. 2000a. ‘Mental Health Care in Marathwada Earthquake Disaster — 1: Organisation of Services’, Th e Indian Journal of Social Work, 61 (4): 631–39. Pande, N.N., S.S. Phadke, M.S. Dalal, P.J. Gadkari, U.S. Nagapurkar, and M.M. Agashe. 2000b. ‘Mental Health care in Marathwada Earthquake Disaster — 2: Short-term Outreach Counseling’, Th e Indian Journal of Social Work 61 (4): 640–51. Pande, N.R. 1994. ‘Role of MIMH in Management of Disaster Consequences’, in M.M. Agashe and S.S. Phadke (eds), Proceedings of Symposium on Health Psychosocial and Mental Effects of Disasters Ú 335

Consequences of the Marathwada Earthquake Disaster. Pune: Maharashtra Institute of Mental Health (MIMH). Parasuraman, S. and N. Acharya. 2000. ‘Analysing Forms for Vulnerability in a Disaster’, Th e Indian Journal of Social Work 61 (4): 581–97. Parasuraman, S. and P.V. Unnikrishnan. 2000. India Disaster Report. New Delhi: Oxford University Press. Patel, V. 2000. ‘Culture and the Mental Health Consequences of Trauma’, Th e Indian Journal of Social Work 61 (4): 619–30. Pereira, M.S. 2000. ‘Teaching Disaster Management: Addressing Complexities of Social Work Practice’, Th e Indian Journal of Social Work 61 (4): 693–701. Purewal, J. and I. Ganesh. 2000. ‘Gender Violence, Trauma and its Impact on Women’s Mental Health’, Th e Indian Journal of Social Work 61 (4): 542–57. Rajagopal, S. and S.K. Chari. 2003. Disaster Management – A Reader. Bangalore: NIAS. Sekar, K., A.S. Dave, S. Bhadra, G.P. Rajashekar, K. Kishore Kumar, and R. Srinivasa Murthy. 2002. Riots: Psychosocial Care by Community Level Helpers for Survivors. Bangalore: Books for Change. Sen, A. 1981. Poverty and Famines: An essay on Entitlement and Deprivation. New Delhi: Oxford University Press. Sengupta, S. 2004. ‘Psychosocial Assistance after an Earthquake: A Case Study of UNICEF Support in Gujarat’, in R.L. Narayana, R. Srinivasa Murthy and J.P. Diaz (eds), Disaster Mental Health in India. New Delhi: Indian Red Cross. Sethi, B.B., M. Sharma, J.K. Trivedi, and H. Singh. 1987. ‘Psychiatric Morbidity in Patients Attending Clinics in Gas Aff ected Areas in Bhopal’, Indian Journal of Medical Research 86 (supplement): 45–50. Sharan, P., G. Chaudhary, S.A. Kavathekar, and S. Saxena. 1994. ‘Preliminary Report of Psychiatric Disorders in Survivors of a Severe Earthquake’, American Journal of Psychiatry 153 (4): 556–58. Shroff , Sudeep and Kamlesh Dave. 2001. ‘Posttraumatic Stress among Ado- lescents in Earthquake Aff ected Kachchh’. Paper presented at Annual Con- ference of Indian Psychiatric Society, Kolkata. Srikala, B., C.R. Chandrasekar, K. Kishore Kumar, P. Chowdhury, R. Parthasarathy,

Downloaded by [University of Defence] at 01:15 24 May 2016 S. Girimaji, K. Sekar, and R. Srinivasa Murthy. 2000. Psychosocial Care for Individuals after the Orissa Supercyclone. Bangalore: Books for Change. TARU. 1995. Marathwada Earthquake: Rehabilitation and Reconstruction Work Two Years After. New Delhi: TARU. Tata Institute of Social Science (TISS). 1994a. Status Report on Rehabilitation of Women and Children in Latur and Osmanabad Districts. Mumbai: TISS. TISS. 1994b. Survey of People Aﬀ ected by Earthquake in Latur and Osmanabad Districts, 1993. Mumbai: TISS. Th ukral, E.G. 1996. ‘Development, Displacement and Rehabilitation: Locating Gender’, Economic and Political Weekly 31 (24): 1500–04. 336 Ú R. Srinivasa Murthy

Vankar, G.K. and K. Mehta. 2004. ‘Gujarat Earthquake’, in R.L. Narayana, R. Srinivasa Murthy and J.P. Diaz (eds), Disaster Mental Health in India. New Delhi: Indian Red Cross. Verma, B.K. and B. Bhushan. 1994. ‘Disaster Management in India: A Community Perspective’, in V.K. Sharma (ed.), Disaster Management. New Delhi: Indian Institute of Public Administration. World Health Organisation (WHO). 2006. Psychosocial Support for Tsunami aff ected Population of India. New Delhi: WHO India Country Offi ce. ———. 2006. ‘Report of the Psychosocial Support Programmes in the Tsunami aff ected Populations in India’, International Review of Psychiatry 18: 299–308. ———. 2008. Psychosocial Support for Tsunami aff ected Populations: Consolidated Report, January 2005–December 2007. New Delhi: WHO India Country Offi ce. Downloaded by [University of Defence] at 01:15 24 May 2016 14 Catastrophe Risk Assessment and Management Developing Risk Models for the Insurance Industry Haresh C. Shah and Weimin Dong

During the past 15 years, there has been spectacular growth in the use of risk analysis and management tools, developed by engineers, in the fi nancial and insurance sectors. In particular, the insurance, reinsurance and investment banking sectors have enthusiastically adopted loss estimation tools developed by engineers in evolving their business strategies and to manage their fi nancial risks. As a result, insurance/reinsurance strategies have evolved as a major risk mitigation tool in managing catastrophe risk at the individual, corporate and government levels. Th is is particularly true in developed countries such as the US, Japan and countries of western Europe. Unfortunately, it has not received adequate attention in developing countries, where such a strategy for risk management is most needed. In the earlier years of catastrophe model development, risk analysts were mainly concerned with risk reduction options based on engineering strategies, and relatively little attention was given to fi nancial and economic strategies. Th is state of aff airs still exists in many developing countries. New developments in the science and technologies of loss estimation due to natural catastrophes have made it possible for fi nancial sectors to model their business strategies such as peril and geographic diversifi cation, premium calculations, reserve strategies, reinsurance Downloaded by [University of Defence] at 01:15 24 May 2016 contracts, and other underwriting tools. Th ese developments have not only changed the way in which fi nancial sectors assess and manage their risks, but have also changed the domain of opportunities for engineers and scientists. Th is chapter describes the role catastrophe risk insurance and reinsurance have played in managing fi nancial risk due to natural catastrophes. Historical losses and the share of those losses covered by insurance will be presented. How such risk sharing can help the nation 338 Ú Shah and Dong

share the burden of losses between the tax-paying public, the ‘at risk’ property owners, insurers and reinsurers will be discussed. Th e chapter will summarise the tools that are used by insurance and reinsurance companies to estimate their future losses due to catastrophic natural events. It will also show how the results of loss estimation technologies developed by engineers are communicated to the business ﬂ ow of insurance/reinsurance companies. Recent developments in loss estimation technologies have impacted the way insurance and reinsurance industries construct their business strategies. Let us look at the role the insurance industry has played in mitigating the economic impacts of recent disasters. Table 14.1 shows the top 10 US insured property losses in the last 15 years. Table 14.2 shows long-term statistics of losses — insured and total — for great natural catastrophes1 between 1950 and 2002. All losses in this table are in US$ billion and they have been normalised to 2002 values. Th is table demonstrates some very interesting trends. Th e absolute value of economic losses is increasing with every decade and so is the value of insured losses. What is more interesting is the ratio of losses. Th e ratio of economic losses in the 1960s compared to the 1970s, 1980s and 1990s has gone up from 1.83 to 7.3. However, a similar ratio for insured

Table 14.1: Top 10 US Insured Property Losses (in US$ billion)

Event Loss in erstwhile dollars Loss in 2001 dollar value World Trade Center (2001) 40.0 40.0 Hurricane Andrew (1992) 15.5 19.6 Northridge Earthquake (1994) 12.5 14.9 Hurricane Hugo (1989) 4.2 6.0 Hurricane Georges (1998) 2.9 3.2 Tropical Storm Allison (2001) 2.5 2.5 Hurricane Opal (1995) 2.1 2.4 Hurricane Floyd (1999) 2.0 2.1 20-state winter storm (1993) 1.8 2.1 Oakland Firestorm (1991) 1.7 2.2 Downloaded by [University of Defence] at 01:15 24 May 2016 Source: Dong, personal communication.

1 Th e deﬁ nition of great natural catastrophes is taken from Munich Re (2002): Natural catastrophes are classed as great if the ability of the region to help itself is distinctly overtaxed, making interregional or international assistance necessary. Th is is usually the case when thousands of people are killed, hundreds of thousands are made homeless, or when a country suﬀ ers substantial economic losses, depending on the economic circumstances generally prevailing in that country. Risk Assessment and Management Ú 339

Table 14.2: Long-Term Statistics, 1950–2002

1950– 1960– 1970– 1980– 1990– Last 10 Decade 59 69 79 89 99 Years Number of events 20 27 47 63 91 70 Economic losses 42.1 75.5 138.4 213.9 659.9 550.9 Insured loss – 6.1 12.9 27 124 84.5 Ratio of insured to – 8.1 9.3 12.6 18.8 16.6 economic losses %

1970s/ 1980s/ 1990s/ Last Factor 1960s 1960s 1960s 10/1960s Number 1.74 2.3 3.4 2.6 Economic losses 1.83 2.8 8.7 7.3 Insured 2.11 4.4 20.4 13.9 Source: Prepared by author using data taken from Munich Re (2002).

properties has gone up from 2.11 to 13.9. Th is indicates that insurance has taken a larger burden of natural disaster losses. Not only have the absolute numbers gone up but also the burden of losses has been increasingly transferred to insurance. Th is transfer of risk is quite signifi - cant. Similar statements cannot be made for developing countries such as India. Figure 14.1 shows losses due to great natural catastrophes worldwide by year and by decade. Figure 14.2 shows similar losses for the US. It is clear that in the US, the burden of losses is shared by insurance, whereas this is not true on a worldwide basis. So what is the point of these numbers and statistics? It is that a prudent strategy of catastrophe risk management requires that besides focusing on technological fi xes to reduce losses, a nation must develop fi nancial risk management strategies which also include risk transfer strategies to insurance, reinsurance and capital markets. In the next section, we will look at the stakeholders in this fi nancial risk management pyramid.

Downloaded by [University of Defence] at 01:15 24 May 2016 Insurance could play a very important role in disaster recovery. In the 1994 Northridge earthquake in the US, two-thirds of the loss was borne by the insurance industry, which greatly reduced the government’s burden and resulted in rapid post-disaster recovery (Figure. 14.3). As opposed to this, in the 1995 Kobe earthquake, less than 5 per cent of the burden of economic losses was covered by insurance, leaving the rest to come from the taxpayers of Japan and from the local governments. In a wealthy country, this strategy may work but for the rest of the world, this is certainly not the most desirable way of managing catastrophe risk. Figure 14.1: Losses due to Great National Catastrophes Worldwide (a) by Year and (b) by Decade

$170 bn. $100 $90 $80 I $70 -fcfl- $60 8 $50 $40 $30 $20 $10 $0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 Year

• Uninsured Economic Losses • Insured Economic Losses

$700

$600

$500

o $400

$300

$200

Downloaded by [University of Defence] at 01:15 24 May 2016 $100 E2 $0 1950-1959 1960-1969 1970-1979 1980-1989 1990-1999 Decade

• Uninsured Economic Losses • Insured Economic Losses

Source: Munich Re (2002:14). Figure 14.2: Losses due to Significant in the US Natural Disasters (a) by Year and (b) by Decade

$60

$50 o 2 $40 -fae-

$30 s 1$20 $10

$0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 200( Year

• Uninsured Economic Losses • Insured Economic Losses

$250

$200 |

$150 GO

$100

E3 $50 Downloaded by [University of Defence] at 01:15 24 May 2016

$0 1950-1959 1960-1969 1970-1979 1980-1989 1990-1999 Decade

• Uninsured Economic Losses • Insured Economic Losses

Source: American Re (2002:12). 342 Shah and Dong

Figure 14.3: Distribution of Northridge Recovery and Reconstruction Funds by Sources Government/Transportation Other Insurance US$ 7,157 _____ US$ 1,042

Commercial Residential insurance insurance US$ 4,854 US$ 12,651

Funds in million (Total = US$ 25.7 billion) Source: Craig Comartin, personal communication.

Basic Issues of Catastrophe Insurance

The main stakeholders in managing financial risk are shown in Figure 14.4. Property owners are the ones who can potentially bear losses due to natural catastrophes. To protect themselves, they may transfer some of their risk to insurance companies. When insurance companies accumulate a large amount of risk in the above process, they may want to transfer some of their risk to reinsurance companies. The capital markets at the top of the risk pyramid provide capital to the insurance and reinsurance markets through financial instruments such as catas- tronbp bonds.

Figure 14.4: Stakeholders in the Management of Risk

Capital SEC

Downloaded by [University of Defence] at 01:15 24 May 2016 markets Rating agencies Reinsurers State insurance regulators Insurers

Property owners

Source: Dong, personal communication. Risk Assessment and Management Ú 343

It is the insurance rating agencies and state insurance commission- ers generally who regulate the functioning of the insurance and the reinsurance companies, and the Security and Exchange Commission regulates the capital markets. Th us, in the above pyramid of stakeholders, the government, private industries, capital markets, and society at large are the interested parties. Th ough all the above stakeholders may have an interest in developing a workable mechanism for risk transfer, due to the nature of catastrophe risk (rare events with huge consequences), insurance companies in general may be reluctant to issue insurance policies and could eventually claim that such a risk is uninsurable. So, what does the insurer need to know in order to make catastrophe insurance work? Simply put, an insurer needs to know the following:

(i) What is the expected annual (or annualised) loss? Is it, for instance, $100 per policy or $200 per policy? Th is will make a difference in determining the premium rate charged to customers. (ii) Of course, every insured property is diﬀ erent. As an example for earthquake risk, some houses may be located on very soft soils, like near the river banks, while some may be on rock. Th eir expected annual loss would be very diﬀ erent. Th erefore, the premium needs to be adjusted so as to account for various site conditions, building types, building performance under earthquake loads, and construction quality. (iii) For a company portfolio, what is the probable maximum loss? Th e probable maximum loss is often expressed as loss for a selected return period (e.g., 250 or 500 years). How much surplus does the company need in order to cover losses of such magnitude? (iv) How to control the over-concentration of exposure by limiting underwriting in overexposed regions? Downloaded by [University of Defence] at 01:15 24 May 2016 It has been a common practice in the insurance industry to seek answers to the above questions by resorting to accumulated loss data, such as those collected by the Property Claims Services (PCS) for earthquakes and hurricanes in the US. However, because catastrophes are rare events, these loss databases are small and sporadic (recent loss records for the US date back only to 1949). Th e most important reason that dissuades the use of historical data alone as the basis for catastrophe risk management is that the data window is simply too narrow to 344 Ú Shah and Dong

catch the full impact of rare, adverse events. Th is window is very small compared to the return period of catastrophic events. Typically, major earthquakes have an average return period in the order of hundreds of years. Th e length of historical loss data is simply too short to allow its use as the empirical basis for reliable loss estimates. Missing one extreme loss (event) will result in a much lower estimate of the average annual loss. Conversely, including a big loss in a short time interval will overestimate the annual loss. Table 14.3 is a tabulation of the premium income/loss histories in California with regard to earthquakes since 1970. From this table, it

Table 14.3: Underwriting Experience (1970–94, including Northridge) — Earthquake Insurance Premiums and Losses (in US$ million)

Earthquake Premium collected Losses paid Year event in each year in each year 1970 — 5.9 0 1971 San Fernando (6.6) 4.6 .8 1972 — 9.0 2.1 1973 — 10.9 .1 1974 — 13.0 .4 1975 Oroville 13.8 0 1976 — 17.1 .1 1977 — 19.8 .1 1978 Santa Barbara 23.2 .4 1979 Imperial Valley (6.6) 29.0 .6 1980 — 38.5 3.5 1981 — 50.2 .5 1982 — 58.9 0 1983 Coalinga (6.7) 70.4 2.0 1984 Morgan Hill (6.2) 79.5 4.0 1985 — 132.9 1.7 1986 Southern California 180.0 16.7 1987 Whittier (5.9) 208.4 47.6 1988 — 277.8 31.8

Downloaded by [University of Defence] at 01:15 24 May 2016 1989 Loma Prieta (7.1) 333.6 433.0 1990 Southern California 384.6 180.9 1991 Northern California 427.4 73.7 1992 — 479.9 87.7 1993 — 521.0 13.2 1994 Northridge (6.9) 619.4 7,414.1 Total 4,008.7 8,295.0 Average loss ratio (excl. 1994) 0.26 Average loss ratio (incl. 1994) 2.07 Source: Roth and Van (1993–94). Risk Assessment and Management Ú 345

is obvious that a single modern event can alter historical estimates signiﬁ cantly. In particular, according to the historical data, the average loss ratio for the period 1970–93 is 0.26, but increases to 2.07 when 1994 is included. Th e Northridge earthquake was a moderate earthquake. Th e jump would be much greater in case of an event on the Newport Inglewood fault, or a repeat of the 1906 San Francisco earthquake (Roth and Van 1993–94). Based on the short historical data, it is impossible to obtain the probable maximum loss for a much longer term. From Table 14.3, the maximum loss is US$ 433 million (1989) before the Northridge earthquake, but US$ 7,414 million (1994) after this event. Inclusion of the Northridge event boosts the maximum loss by about 18 times even for short return periods (25 years or less). For these reasons, the use of historical data alone to forecast catastrophe losses is unsatisfactory; such estimates contain large uncertainties. However, these uncertainties can be quantiﬁ ed and understood through the use of physically-based engineering models.

Engineering Modelling and Loss Estimation Technologies

Recent advances in loss estimation technologies, based on mechanistic and scientifi c principles, have made it possible to quantify losses from catastrophes with reasonable accuracy for insurance purposes where this type of information is needed at the portfolio level rather than at the micro level of a specifi c building. Phenomenological models have been developed and used to estimate potential losses to insurance exposure under any conceivable scenario. While modelling does not eliminate uncertainty, loss estimates based on engineering models can complement the information distilled from historical data because they can be made to refl ect current and future exposure, vulnerability and

Downloaded by [University of Defence] at 01:15 24 May 2016 valuation changes. At present, almost all major property and casualty insurance and reinsurance companies use engineering based models to help them in underwriting, risk control, price setting, reinsurance purchase and optimal capital allocation. In general, catastrophe impact models have four components or modules: a stochastic event module, an attenuation module (for earthquakes or for hurricanes), a vulnerability module, and a ﬁ nancial analysis module. Technical and scientiﬁ c professionals mainly provide the 346 Ú Shah and Dong

input for the fi rst three modules, whereas insurance professionals provide the input for the fi nancial analysis module. Th is last module converts damage information into dollar loss estimates, and propagates this loss through the entire insurance fi nancial structure. In general, a complete engineering model for catastrophe impacts has three components or modules (Dong et al. 1988; Grossi et al. 2005):

(i) A hazard module, which characterises a hazard in terms of a series of events with various sizes, locations and frequencies. (ii) An attenuation module, which generates the local hazard severity (ground shaking intensity or peak wind gust) from the source, given the size and location of the event. (iii) A vulnerability module, which estimates the loss given the local hazard severity.

Th ese various modules are discussed in the following sections.

Hazard Module

Th e starting point in the engineering modelling approach is the energy source of the disturbance. For example, for earthquake hazards, the source is the rupture of a particular fault (Figure 14.5). A credible rupture scenario is postulated, which is consistent with the seismological and geological database on frequency, magnitude and rupture length. Hurricanes are associated spatially with tracks rather than with faults, and historical data are used to establish their credible tra- jectories in a given region (Figure 14.6). Th e source for a hurricane is, therefore, a probabilistic event in a general ocean area, with a certain level of severity, as indicated, for example, by the Saﬃ r-Simpson category. For each region and peril, a stochastic event set is created based on the best knowledge as well as historical data. Each event in the sto- Downloaded by [University of Defence] at 01:15 24 May 2016 chastic event set will have geographic details, magnitude and occurrence rate, and the attenuation module will then generate a local intensity (or footprint) as described next.

Attenuation Module

Th e energy released at the source manifests itself in the form of physical hazards such as ground-shaking or increased wind velocity. Th ese hazards Risk Assessment and Management 347

Figure 14.5: Major Faults in California

LEGEND • >7.0 % 6.5-7.0 O 6.0-6.5 A 5.5-6.0

0 50 100 Miles

Source: Dong, private communication.

are propagated from the source region, and attenuate and disperse as

Downloaded by [University of Defence] at 01:15 24 May 2016 they propagate (Figure 14.7). The degree of attenuation is important as it affects the severity of hazards felt at the site of interest and the extent of influence of the event. The governing physical laws are summarised in the form of attenuation functions, which rely mostly on empirical data for calibration. The corresponding key hazard parameter for hurricanes is wind speed, which is also attenuated in an analogous fashion. Secondary or collateral hazards (e.g., liquefaction and landslide for earthquakes, and wave surge and rain for hurricanes) are generated based on the key hazards and local Figure 14.6: Historical Hurricane Tracks

Key Statistics • Cat 5 2 Category 5 • Cat 4 15 Category 4 • Cat 3 47 Category 3 37 Category 2 • Cat 2 57 Category 1 • Catl 1,6 events per year

Source: Dong, private communication.

Figure 14.7: Attenuation of Ground-Shaking with Distance

Local Site Effects

Source Model Downloaded by [University of Defence] at 01:15 24 May 2016

Attenuation of Seismic Energy

Source: Authors. Risk Assessment and Management 349

site conditions (e.g., soil type and site slope for earthquakes; topological conditions and site roughness for hurricanes).

Vulnerability Module

The damage caused to an asset exposed to a hazard depends on its vulnerability. While understanding the damage process entails complex physical phenomenology and engineering development, the concept can be illustrated with reference to the loss ratio curve shown in Figure 14.8, defined as repair cost divided by replacement cost. The figure gives the extent of damage to a particular class of structures when they are subjected to various levels of ground shaking (generally in terms of Modified Mercalli Intensity [MMI]). Uncertainties in the estimate of damage, due to inherent variability in the details of the structure as well as the state of knowledge, are also noted. Vulnerability to wind damage is quantified and represented in a similar fashion. These curves have since been greatly extended and refined, based on damage and insurance claim data collected in recent events as shown in Figure 14.9. Vulnerability and loss information have become increasingly

Figure 14.8: Loss Ratio of Building Classes 80

50 Damage 40 ratio (%) 30

10 Downloaded by [University of Defence] at 01:15 24 May 2016 0 VI VII VIII IX X XI XII MMI

• Low-rise wood frame • High-rise un-reinforced masonry

Source: Dong, personal communication. 350 Shah and Dong

Figure 14.9: Historical Data for Wind Damage 100.00%

10.00%

1.00%

0.10%

0.01% Wind speed

•ANDREW HUGO «OPAL • ERIN

Source: Dong, personal communication.

detailed and comprehensive in order to satisfy the needs of modern insurance applications. In summary, the engineering model provides a series of potential scenarios for a region with different magnitudes and locations. All those scenarios have an annual occurrence rate, based on scientific knowledge and historical data. Then, given a scenario, the engineering model will estimate all local impacts, which includes monetary loss to buildings; casualty to the occupants; and potential down time for the operation. In the next section, we will discuss how those losses are shared by various stakeholders through the insurance function, i.e., to distribute losses among the property owners, insurer/reinsurers, and the potential capital market. Downloaded by [University of Defence] at 01:15 24 May 2016 Financial Module for Catastrophe Insurance

In general, insurers attempt to protect themselves from enormous loss by a variety of means such as deductible limit and by buying reinsurance. Allocating losses among various participants is actually a very complicated task, dependent on the insurance portfolio composition and on insurance structure. In this section, we will first describe the Risk Assessment and Management 351

composition of an insurance portfolio, and then discuss how to quantify an insurer's risk.

Insurance Portfolio Composition

Insurance companies who issue catastrophe policies (for earthquakes and hurricanes) are concerned with their probable maximum loss (PML). Their portfolios consist of many individual policies. Assets to be insured can be situated in a single location, or distributed in many locations (e.g., cities, counties or states). An insurance policy stipulates how the assets located at a group of locations are to be protected. Most commercial policies cover more than one location, whereas residential policies usually cover only a single location. A collection of policies constitutes a portfolio. This hierarchy is illustrated in Figure 14.10.

Figure 14.10: Depiction of an Insurance Portfolio

Portfolio

Policy 1 Policy m Policy 2 Policy 3

Loci Locn Loc2 Loc3

Source: Grossi and Kunreuther (2005:121). Downloaded by [University of Defence] at 01:15 24 May 2016

Residential Portfolio Most residential policies have a simple insurance structure: one policy, one location, with deductible and coverage limit for building, contents, and additional living expense. So, a residential portfolio for a primary insurance company could have many policies with one location for each policy, the number of policies for a portfolio could be as large as 352 Ú Shah and Dong

hundreds of thousands. Most primary insurance companies collect relatively detailed data for their policies, including building type, year built, street address (or ZIP code), market values, deductibles, and limits. Some companies may even record building-specifi c information, such as the number of stories, presence of cripple walls, past retrofi ts, etc., as well as site-specifi c geological conditions. When reinsurers assume risk from primary insurers, however, there is usually a marked decrease in the data quality. It is much more common for reinsurers to get aggregate data, such as the total value of the insured risk in a ZIP code or county with an average deductible and coverage limit. For instance, US$ 125.3 million of exposed value with US$ 3.6 million deductible and US$ 110 million of coverage limit in ZIP code 94306. Th e total number of policies in a ZIP code or county may not be provided. Detailed information regarding the building features is no longer present. A distribution of building types must then be inferred for any analysis of the portfolio risk. Due to the relatively low value of the individual residential properties, reinsurance treaties are based on whole portfolio and are generally not attached to any single policy. Commercial Portfolios For a commercial portfolio, a given insurance policy may cover a franchise and thus involve many locations. Since the insured value of a commercial building is much greater, detailed information about the building is usually required by the insurer to underwrite these policies. An engineer may therefore need to inspect construction plans or be sent to the site for particularly valuable risks. Some policies may cover a schedule of nearly uniform buildings in diff erent parts of the country, e.g., a fast-food outlet chain. In most cases, there will be a location deductible and limit for building, contents and business interruption. Also, there will be a policy-level deductible and coverage limit to protect excessive insurance losses

Downloaded by [University of Defence] at 01:15 24 May 2016 for that policy. For instance, a policy may cover 100 locations, each with a $1 million coverage limit. In case all locations suﬀ ered total damage up to the limit ($1 million), the insurer would have to pay $100 million. In order to avoid excessive loss, the insurer may impose a policy coverage limit for this policy, say $20 million. In that case, if the sum of losses for all 100 locations is greater than $20 million, based on a policy limit, the loss for the insurer would be capped at $20 million. For a commercial portfolio, the quality of data provided to a reinsurer is likely to be better than for a residential portfolio, particularly if the commercial portfolio Risk Assessment and Management Ú 353

comprises a small number of high-value locations. Reinsurance may be applied at a location level, since the exposure for a commercial policy is much larger. Reinsurance Structures Reinsurance plans come in various forms and complexities, and an in- depth discussion on these is outside the scope of this chapter. For the present purpose, one needs to distinguish between treaty reinsurance and facultative reinsurance. Treaties generally apply to all policies in a portfolio, while facultative reinsurance addresses individual risks. Th ere are four types of treaties (Dong, private communication): Quota Share, Surplus Share, Working Excess, and Catastrophe.

(i) Quota Share (QS): reinsurance is a form of proportional (pro- rata) reinsurance in which the reinsurer assumes an agreed percentage of each policy being insured and shares all premiums and losses accordingly with the cedant. (ii) Surplus Share (SS): reinsurance is a form of proportional (pro- rata) reinsurance that indemnifi es the ceding company against loss for the surplus liability ceded. Surplus shares can be viewed as a Quota Share, wherein the re-insurer’s pro rata share of insurance varies from policy to policy. (iii) Working Excess (XE): reinsurance is a generic term describing reinsurance, which, subject to a specifi ed limit, indemnifi es the ceding company against all or a portion of the amount in excess of a specifi ed retention. Working excess is an area of excess reinsurance in which loss frequency is anticipated (i.e., working), as opposed to loss severity. Th us, a working cover would usually have a low indemnity and would attach above a relatively low retention. (iv) Catastrophe (CAT): reinsurance is a form of excess of loss re- Downloaded by [University of Defence] at 01:15 24 May 2016 insurance, which, subject to a specifi ed limit, indemnifi es the ceding company for the amount of loss in excess of a specifi ed retention with respect to an accumulation of losses resulting from a catastrophic event or series of events. Th e actual reinsurance document is referred to as ‘a catastrophe cover’.

Facultative reinsurance is similar to treaty reinsurance in structure and feature, except that each facultative treaty (FAC), is associated with 354 Ú Shah and Dong

a particular policy. A policy may have more than one FAC associated with it. For primary insurers, the deductible and coverage limit, plus all kinds of reinsurance, can be used to manage portfolio risk. A reinsurer’s portfolio is diﬀ erent. As noted previously, they usually do not have detailed information regarding individual location and policy. Th eir portfolios consist of all treaties and facultative reinsurance from many ceding companies. In turn, a given reinsurer may choose to cede some of its portfolio risk to another reinsurer.

Portfolio Risk Quantiﬁ cation

The loss estimated in the engineering model described in the previous section is total loss, sometimes called ‘ground-up loss’, for the locations and an insurance portfolio. Th e total loss has to be allocated among diff erent participants (client, insurer and/or reinsurer). Th e task of portfolio risk quantifi cation is to do exactly that. Th e following sections will focus on how to quantify fi nancial risk and discuss several important factors involved in this process. The Bottom-Up Approach Th e bottom-up approach provides the most robust means to quantify portfolio risk. Th at is, losses are fi rst calculated for the insured and the insurer at the location level, based on deductible and coverage limit. If there is reinsurance attached at the location level, then the loss to the reinsurer is estimated. Th e next step is to aggregate all location losses in a policy to fi nd the gross loss to the insurer for this policy. All policy-level reinsurance treaties will be applied to further distribute losses between the insurer and reinsurers for that policy. Th e third step is to aggregate losses over all policies in the portfolio for all parties. Since QS, SS and XE usually have a per risk limit (for each policy) and per occurrence limit Downloaded by [University of Defence] at 01:15 24 May 2016 (for all policies), the aggregation of QS, SS and XE losses must be capped at their corresponding per occurrence limit. Th e excess parts will be passed back to the insurer. For instance, if the sum of all policy losses for QS is $13.5 million and the occurrence limit for QS is $10 million, then the excess part of $3.5 million will be passed back to insurer’s net loss. Finally, all CAT treaties will be applied to the insurer’s net loss before CAT to fi nd out the insurer’s net loss after CAT as well as all CAT losses (see Figure 14.11). Downloaded by [University of Defence] at 01:15 24 May 2016

Figure 14.11: Depiction of the Four Types of Treaties and the Use of Facultative Reinsurance

— Per-Occurrence Limit CAT-01 Portfolio Applied To Total Portfolio Loss NetofQS,SS,FACAndXE

Unallocated losses. — Per-Risk Limit Applied To The Loss Net Of QS, SS And FAC In Each Policy

XE-01 XE-01 XE-01 — Per-Occurrence Limit Applied To The XE Loss Summed Over All Policies Unallocated loss Unallocated loss Unallocated loss — Per-Risk Limit QS-01 'QS-01 Applied To The Gross QS-01 SS-O1 SS-01 Loss In Each Policy — Coverage % of FAC-02 FAC-01 FAC-ON, SS May Vary From — Per-Occurrence Limit Policy to Policy Applied To The Sum Of Policy #1 Policy #2 Policy #N Treaty Losses Over All Policies

Source: Dong (2001: 93). 356 Shah and Dong

In allocating losses, QSs, SSs and FACs are first applied to each policy, and the reinsurer's loss under each type of agreement is limited by the per-risk limit, if appropriate. The remainder or excess nominally becomes the responsibility of the insurers. However, if there are any XEs in effect, a portion of the excess is reallocated to the reinsurer according to the terms of the XE, taking into consideration any per-risk limits that may exist. The allocation of loss at the policy level is then complete. A valuable graphic tool in describing the roles of the various treaties and their interactions is the Loss Diagram. We proceed to give a brief introduction of this tool before discussing the treaties. A Useful Tool for Financial Loss Calculation — Loss Diagram The loss diagram is a two-dimensional representation of loss (see Figure 14.12) that is the backbone of the insurance/reinsurance models. The Y-axis of a loss diagram measures the loss, and the X-axis runs from

Figure 14.12: Loss Diagram Displaying Sample Loss as an Area Loss

TEXP

Deductible + Limit

A sample loss Downloaded by [University of Defence] at 01:15 24 May 2016

(A sample loss)

Deductible

Pro rata

0% 100% Source: Author. Risk Assessment and Management 357

0 to 1, which correspond to 0 per cent to 100 per cent pro-rata, respectively. With reference to the sampled loss displayed in the diagram, often called the loss line, three thresholds are of fundamental importance in loss modelling. They are, in increasing order of magnitude: (z) Deductible, (zz) Deductible + Limit, and (zzz) Total Exposure. These are illustrated in the loss diagram of Figure 14.12. The Deductible is the first portion of the loss absorbed by the client (the insured). The insurer will absorb the loss above deductible up to the coverage limit. If the loss is greater than deductible + limit, then the excess part will go back to the insured. Usually, we refer to the total loss as 'ground-up loss' and loss to the insurer before application of reinsurance as 'gross loss'. Figure 14.13 shows a more complex loss diagram with four treaties: two facultative reinsurances, one quota share treaty and one surplus treaty. The policy has deductible of 0.5 million and a limit of 9.5 million. FAC1 is 20 per cent of 1.5 million above deductible. FAC2 is 20 per cent

Figure 14.13: Insurance Structure Used in Example 10

SS 5.5 5.0

FAC2

2.5 QS Loss line

Downloaded by [University of Defence] at 01:15 24 May 2016 2.0

FAC1

0.5 Deductible 0.0

20% 20% 10% Source: Author. 358 Ú Shah and Dong

of 3 million excess 1.5 million. QS is 20 per cent of 5 million above deductible, and SS is 10 per cent of 8 million excess of 1.5 million. Note that attachment is based on gross loss (i.e., above deductible). If the ground-up loss is 2.5 million (loss line), the gross loss will be 2 million. Th en we will have:

FAC1 loss = 0.2∗1.5 million FAC2 loss = 0.2∗(2–1.5) million QS loss = 0.2∗2 million And, SS loss = 0.1∗(2–1.5) million

In general, all treaties have a similar structure: attachment, limit and participation. In conventional shorthand, this is written as Participa- tion% Limit × Attachment. For example, FAC 2 in the previous case would be described as 20% 3M × 1.5 M. Th e loss can be calculated as the following general formula:

Reinsurer Loss = min (max (gross loss – attachment, 0), limit) ∗participation

Given an event, we can use the hazard and vulnerability model to calculate ground-up loss for each location in the portfolio. Stepping through the levels of the fi nancial model above allows us to allocate this ground-up loss to each of the participating parties: insured, insurer and reinsurer(s). Since the event is random, we can associate an annual rate of occurrence with it and, by extension, with the calculated losses. For all possible events with their occurrence rates, we can calculate all losses associated with each event. We can then compile an event-loss table, which will be discussed later. Importance of Recognising Uncertainty in Loss (Distributed Loss) Downloaded by [University of Defence] at 01:15 24 May 2016 Since reinsurance structures usually provide a non-linear fi lter to a ground-up loss, the uncertainty surrounding a loss estimate is of paramount importance. One cannot properly allocate losses to diff erent parties simply by using the expected or mean value of the potential ground-up loss. Th e entire loss distribution needs to be considered in the process. For example, a review of empirical data on building performance reveals that while structures of the same type do tend to perform similarly, a great deal of variability does still exist. Figure 14.14 Risk Assessment and Management x 359

Figure 14.14: Dwelling Loss Exnerienee for the 1971 San Francisco FarthnnaUp 25%

[ 20%

sfl 15% I 10% U-i O I 5% 0% w s* D5* In o Oo LT;, O S o oO O o CM esi -•-• in g £ R 0> o m § B s s o !3' W C rH A 5 c-i 5 £ a SK S£ -V •3" i 9 ,= s fS S s Damage ratio Source: Grossi and Kunreuther (2005: 129),

highlights the variability in building performance for wood-frame dwellings, normalised to similar levels of ground motion and soil conditions (Grossi and Kunreuther 2005:129). To illustrate the importance of variability in a loss estimate: lets assume that the estimated damage of a building has a probability density distribution as depicted in Figure 14.15, with an expected (mean) damage ratio of 7 per cent and the coverage deductible is 10 per cent. We compare the loss allocations made based on the expected loss or

Figure 14.15: Difference between Expected and Distributed Loss Allocations tt 1 Deductible = 10% Downloaded by [University of Defence] at 01:15 24 May 2016 Loss Distribution r

Loss

Expected Loss = 7% Source: Grossi and Kunreuther (2005: 130). 360 Ú Shah and Dong

mean damage, and based on the potential distribution of the damage ratio. In the former, because the mean damage (at 7 per cent) is less than the deductible, it would appear as if the client incurs all of the loss. In the latter, a random damage ratio is sampled from the distribution curve, and the loss allocated is based on the sampled loss and the deductible. Of the sampled losses, 100 per cent of those that are below the deductible go to the client. Th e client accepts the deductible, and the excess goes to the insurer for sampled losses that are above the deductible. If all losses are then weighted by the appropriate probability, the resulting net mean loss allocations are 5.2 per cent for the client, and 1.8 per cent for the insurer. Th e results are summarised in Table 14.4.

Table 14.4: Loss Allocation

Allocation method Client loss Insurer loss Point-estimate 7% 0% Distributed 5.2% 1.8% Source: Grossi and Kunreuther (2005: 130).

Hence, mainly because of the presence of deductibles, loss allocation can be warped only if the mean damage estimate is used. Deduct- ibles inﬂ uence allocation at the low end and coverage limits aﬀ ect allocation at the high end. On the reinsurance side, attachments and limits have similar impacts.

Insured Losses When Considering Loss Uncertainty

Damage to assets is converted to monetary losses by means of the repair/ rebuild cost. However, depending on the terms of the policy, the loss is shared by more than one party. Th is part of the modelling refl ects the workings of a policy or treaty; although not engineering per se, it is Downloaded by [University of Defence] at 01:15 24 May 2016 included as an integral part of the overall modelling approach because of the intimate interaction between the insurance structure and damage to the physical structures or assets. Th e loss diagram of Figure 14.16 is a basic but much simplifi ed view of this interaction. When the loss line is above the deductible, as is the case shown in the fi gure, it is paid by the insured, insurer and reinsurer as indicated. Given a loss line, the loss for any participant (insured, insurer and reinsurer) is called the allocated loss, which is simply: Risk Assessment and Management 361

Figure 14.16: Allocation of Loss to Participating Parties — Simplified Loss Diagram

Reinsurance Participation Probability ...distribution

Loss line

Deductible Participation %

Source: Prepared by author.

Allocated Loss = min (max (Loss Line - Att, 0), Limit) X Participation% (1) where Att denotes the attachment point (or deductible, if appropriate) of the agreement.

The loss line represents an idealisation, and is used for simplicity. The actual loss estimate is probabilistic due to the various uncertainties mentioned previously, and is underscored by the distributed loss curve shown as a dashed line in Figure 14.16. In particular, given that an event has occurred, the building loss incurred is not deterministic,

Downloaded by [University of Defence] at 01:15 24 May 2016 but is distributed with non-trivial variance. Indeed, the variability and distribution of the estimated loss can significantly affect loss allocation to the parties. Denote loss as a random variable, x, then the expected allocated loss would be:

E (AllocatedLoss) = j min (max (x-Att, 0), Limit). Participation % . f(x)dx Where E stands for Expectation and f (x) is the density function of loss distribution. 362 Ú Shah and Dong

Event Loss Table (ELT) and Exceedance Probability (EP) Curve

Based on source modelling, we can generate thousands of events, either from earthquake faults or from hurricane tracks, with particular location, size and frequency. For each of these events, the loss to the portfolio can be estimated. All these results can be summarised in a table, called an event loss table (ELT). An example of an ELT taken from actual case studies is shown in Table 14.5. Due to inherent uncertainties in engineering modelling, given an event, its loss and occurrence rate are not constant, but are random with the corresponding coeﬃ cient of variation (CV), as listed in the table. In the next section, we will discuss how to use ELT, either using simulation or other various numerical approaches, to generate a loss exceedance probability (EP) curve and other metrics.

Insurance Risk Metrics

It has always been a challenge to measure a potential catastrophe for a region or an insurance portfolio due to the uncertainties involved. Th ere are thousands of possible events, and one does not know when they will occur. Even given that an event has occurred, due to the uncertainties involved in each step of risk assessment, loss too could be a randomly distributed variable. Early attempts to measure insurance risk were based on calculating the Probable Maximum Loss (PML). For earthquakes in USA, insurers

Table 14.5: Event Loss Table

Event ID Loss (US$), L Rate, λ CV of Loss CV of Rate 801–1 4,127,951,873 0.00445 0.2396 0.4 801–2 3,577,857,992 0.00600 0.2592 0.4 Downloaded by [University of Defence] at 01:15 24 May 2016 801–3 2,023,851,254 0.00353 0.3213 0.4 802–1 1,825,150,272 0.00353 0.3142 0.4 301–1 1,231,113,404 0.00343 0.3379 0.333 2501–1 1,042,975,125 0.00019 0.3409 0.308 803–1 964,474,985 0.00328 0.3406 0.4 301–1 899,090,537 0.00153 0.3396 0.333 2501–2 839,946,698 0.00075 0.3431 0.308 802–2 797,984,184 0.00328 0.3431 0.4 901–1 696,486,271 0.00346 0.3391 0.4 Source: Dong (2001: 56). Risk Assessment and Management Ú 363

used specifi c scenarios and accumulated losses from all buildings as PML for a portfolio. Th e PML for each building was defi ned as a loss which would not be exceeded for nine out of ten buildings in a given building class. As an example, in California they used San Andreas M 8.3 as the scenario for northern California and Newport-Inglewood M 7.5 for southern California. A.M. Best is the premier rating agency in the US for insurance companies and its position is that to qualify for a secure Best’s Rating (B+ to A++), they defi ne a ‘reasonable’ catastrophic loss as the PML from either an earthquake using a 250-year return period or a hurricane using a 100-year return period. In the case of Florida this is often a 250-year return period, refl ecting A.M. Best’s conservative approach. While these measures have played an important role in insurance risk management, they could not provide a holistic view of potential catastrophe. Th e loss estimation technology provides comprehensive measures to the insurance professionals in the following four formats:

(i) Event Loss Table (ELT) (ii) Average Annual Loss and Standard Deviation (iii) Occurrence Loss Exceeding Probability (OEP) (iv) Aggregate Loss Exceeding Probability (AEP)

Table 14.6 shows a simplifi ed ELT, discussed in the previous section. Contents of the ELT are considered the basic interface of loss estimation technology with insurance applications because its content drives important fi nancial parameters of the insurance business. Each row of the ELT corresponds to a catastrophic event taken from a group of credible scenarios (e.g., earthquakes from nearby faults that are judged to have substantial eff ects on the assets should they occur). Th e event scenario is given an identifi cation (Event ID = j, for example), and the annual rate of occurrence of the event is also noted in the table. Downloaded by [University of Defence] at 01:15 24 May 2016 Table 14.6: Simplifi ed Event Loss Table

Event ID Annual rate Loss

1 λ1 L1

2 λ2 L2 :::

j λj Lj :::

J λJ LJ Source: Author. 364 Ú Shah and Dong

Each scenario event is simulated using the engineering models described, and the losses sustained by the assets are computed and entered into the table. Th e ELT also contains measures of variation (uncertainty) of the parameters (event rates and losses), which are not elaborated in this chapter. Information in the ELT on losses due to individual events lends itself readily to the calculation of the average annual loss and the standard deviation of the loss due to all events. It can also be used to quantify the concentration of exposure, as well as solvency positions. Given below are some applications. Th e average Annual Loss E(L) and the Standard Deviation of the loss σ are then given as: J E(L) = ∑ λj Lj (1) j=1 And,

J = 2 (2) σ ∑Ljjλ j=1 Th e summation index J corresponds to the total number of independent events in the ELT, i.e., the number of rows in the table. It is well known that E(L) and σ are the two key variables that govern insurance pricing. A common pricing formula is:

P = E(L) + α.σ + e (3)

P is the premium while e covers the expense and profi t margin. Th e parameter α is also called the risk load factor. Hence, the basic cost of the policy is the average annual loss E(L) plus expenses and target profi t, with a premium added to protect against any instability eff ects due to the variability in the loss estimate. Th e term α.σ is often referred to as the ‘risk premium’.

Downloaded by [University of Defence] at 01:15 24 May 2016 Entries in the ELT are based on events occurring individually within the year. However, within a particular year, many such events may occur and the losses can be compounded. In the worst-case situation, the compounded loss may be so severe that the company becomes insolvent. To get an approximate measure of the compound losses, the multiple events could be treated as a compound Poisson process with a rate that is based on the rates of the member events:

λ λ = ∑ j (4) j Risk Assessment and Management Ú 365

Th e summation is carried over all member events. It is quite straightforward to demonstrate that the probability of loss being greater than or equal to a particular value of l is:

−∑λj e 1= j P(L ≥ lj) = 1 – P(L < lj) = 1 – (5)

Equation 5 defi nes the Occurrence Loss Exceeding Probability (OEP) curve when secondary uncertainties, i.e., coeffi cient of variation of the loss given an event, are ignored. For the OEP, each loss corresponds to a particular event. For each event one could defi ne the loss return period as the reciprocal of the exceeding probability:

1 RP(L ≥ lj) = (6) PL()≥ lj For a compound process involving n events, the probability that the

aggregate loss L is less than a particular value of lj is:

n∗ F(L < lj|n) = F(L1 + L2, ∙∙∙∙ + Ln < lj) = F (L < lj) (7)

n∗ th In the equation above F (L < lj) is the n convolution of F evaluated at

the loss lj. Hence, the probability that the aggregated loss L is less than a particular value of lj is:

∞ e−λ .λ n –λ n∗ P(L < lj) = e + ∑ F (L < lj) (8) n=1 n! To evaluate Equation 8 a very large number of calculations are required in order to generate the various convolutions of the severity distribution given that a certain number of events have occurred. To reduce the computational eﬀ ort required, one could use Panjer’s (1980) recursive

Downloaded by [University of Defence] at 01:15 24 May 2016 approach (= coeﬃ cient of variation of 1.565 times the mean of 6.5). Based on this curve, if the company wants to maintain solvency at the 1 per cent probability level, the surplus allocated to this line of business must be at least $50 million. Th is example is typical of AEP curves for catastrophe events in that it exhibits a highly skewed distribution. In particular, recall that the normal distribution requires a surplus equal to the mean loss plus 2.3 times the standard deviation in order to maintain a 1 per cent exceeding probability. Th e surplus required here to maintain the same 1 per cent exceeding probability corresponds to the mean loss 366 Shah and Dong

(at $6.5 million) plus 4.3 times the standard deviation (at $10.1 million). The example underscores the fact that caution is required when determining the marginal surplus necessary to maintain a solvency criterion based on conventional wisdom (Kreps 1990). Figure 14.17 shows an AEP curve (enlarged) for a particular portfolio, with a corresponding mean loss of $6.5 million and a standard deviation of $10.1 million. It could also be shown that the AEP is always greater than or equal to the OEP for any loss level. This is illustrated with an example shown in Figure 14.18. For an expected loss of $50 million, the AEP is 4.14 per cent and the OEP is 0.8 per cent as indicated in the inset box. The box also indicates the option of computing AEP and OEP with secondary uncertainty, i.e., uncertainty associated with the engineering models rather than the rate of occurrence only. Due to space limitations this chapter does not elaborate on OEP and AEP derivations with secondary uncertainty.

Portfolio Risk Management

A portfolio manager faces two critical questions with regard to catastrophe risks: what is the expected annual loss and what is the likelihood

Figure 14.17: Aggregate Loss Exceeding Probability (AEP) Curve Probability (%) 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8

Downloaded by [University of Defence] at 01:15 24 May 2016 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Loss 40 48 56 64 72 80 88 96 104 112 120 Source: Dong, personal communication. Note: Mean loss = $6,471,313, and Mean COV = $1.565780 Risk Assessment and Management 367

Figure 14.18: Comparison of Aggregate Loss Exceeding Probability (AEP) and Occurrence Exceeding Probability (OEP) Curves Probability (%) 30 27 24 21 18 15 12 9 6 3 Loss

10 20 30 40 50 60 70 80 90 100 Source: Dong, personal communication. Note: Expected loss = $50 million; AEP without 2nd uncertainty = 4.14; OEP without 2nd uncertainty = 0.80

that the company may become insolvent? The first question is linked to premium rates. A proper rate enables a company to operate smoothly while making reasonable profits for its shareholders. The second question is linked to the company's ability to survive and ensures that the risk of insolvency remains acceptable even though the catastrophe exposure is a component of its exposure. For the solvency constraint, it is critical to adequately model the tail part of the EP curve and this is subject to a significant amount of uncertainty. Efficient management of portfolio risk is key to achieving the goal of remaining solvent while making an acceptable level of profit. In broad terms, there are two levels of portfolio risk management — micro and macro. Micro management addresses individual policies or even Downloaded by [University of Defence] at 01:15 24 May 2016 locations, while macro management considers the portfolio as a whole. The following discussion describes the several commonly-used approaches.

Understanding Risk

The first step to managing the risk of a portfolio is to quantify it. Computer-based modelling and loss estimation provides an important 368 Ú Shah and Dong

tool in this process. As explained before, it is essential to gain an understanding of the input data content and quality in order to be able to understand and interpret the results. Based on models currently available, a probabilistic loss analysis can identify the key drivers of risk by business unit, by peril, by geographic region, or by account. Th is can be used to manage the level of risk of a portfolio. We illustrate this important concept by using an example. Consider a company that wants to limit their 250-year loss to be less than $100 million; that is, the annual probability of exceeding $100 million should be less than 0.4 per cent. Based on the current book of business, losses from various events (HU – Hurricane; EQ – earthquake) with their annual occurrence rates were calculated and listed in des- cending order, as in Table 14.7. Th e Exceed Rate is the cumulative total of the rate on the current line and all lines above. From the table, it is found that the probability of exceeding $100 million is 0.557 per cent. If we look at all events whose losses are greater than $100 million, there are three earthquake events (Event ID EQ-3107, EQ-7103, EQ- 1965) whose loss barely exceeds $100 million. In the aggregate, however, their annual rates amount to 0.213 per cent (0.0586 per cent + 0.0659 per cent + 0.0888 per cent), or about 38 per cent of 0.557 per cent. As a potential strategy, if the earthquake exposure is reduced such that the aggregate loss for each EQ scenario is reduced by 6 per cent, then losses of all three earthquake events will be less than $100 million. Th e EQ events will then fall out of the Table. Th is will result in a portfolio with

Table 14.7: Event Loss Table for the Company’s Current Business

Event ID Loss ($) Rate (%) Exceed rate (%) HU-05276 279,707,730 0.000079 0.008 HU- — — — HU-01152 106,945,669 0.000098 0.232 EQ-3107 105,964,573 0.000586 0.290 Downloaded by [University of Defence] at 01:15 24 May 2016 HU-00512 105,821,572 0.000127 0.303 HU-12743 103,944,373 0.000068 0.310 HU-06031 103,428,541 0.000079 0.318 HU-04407 102,631,772 0.000267 0.344 EQ-7103 102,438,481 0.000659 0.410 HU-03101 101,664,120 0.000529 0.463 EQ-1965 101,056,232 0.000888 0.552 HU-05079 100,329,263 0.000052 0.557 HU-04302 99,526,987 0.000220 0.579 Source: Grossi and Kunreuther (2005: 126). Risk Assessment and Management Ú 369

an acceptable level of risk with the probability of exceedance at 0.344 per cent for a loss threshold of $100 million (Table 14.8).

Table 14.8: Event Loss Table for Company’s Revised Portfolio

Event ID Loss ($) Rate (%) Exceed rate(%) HU-05276 279,707,730 0.000079 0.008 HU- — — — HU-01152 106,945,669 0.000098 0.232 HU-00512 105,821,572 0.000127 0.245 HU-12743 103,944,373 0.000068 0.251 HU-06031 103,428,541 0.000079 0.259 HU-04407 102,631,772 0.000267 0.286 HU-03101 101,664,120 0.000529 0.339 HU-05079 100,329,263 0.000052 0.344 HU-04302 99,526,987 0.000220 0.366 Source: Grossi and Kunreuther (2005: 127).

Reinsurance and Solvency

Th e fi rst step in determining what risk fi nancing strategy to pursue is to undertake a risk assessment based on the company’s current portfolio. As demonstrated previously, catastrophe models can provide a full catastrophe risk profi le, including the probability of exceeding any given level of loss, as well as measures of uncertainty. Th e goal is to optimise the shape of this risk profi le under cost constraints and constraints of the existing portfolio, and to determine whether the current level of risk is acceptable. In Figure 14.19, for example, there is a 4 per cent probability per year that gross losses (losses before any risk transfer) will exceed $50 million. One of the constraints faced by the company is that, in order to write business, the company must hold capital. While minimum capital requirements are set by regulators, the true amount of capital required depends

Downloaded by [University of Defence] at 01:15 24 May 2016 on the risk profi le of the underlying business. Catastrophe models, which depict the full distribution of potential losses, can be helpful in this regard. Th ere are costs associated with holding capital; the more capital required the less leverage is available to the company. In the example above, consider the strategy to transfer the risk of losses above $50 million rather than to hold capital to fi nance them. Th is strategy results in the risk profi le illustrated by the ‘Net of 50 excess (XS) of 50’ case shown in Figure 14.19. Of course, in the extreme tail of the distribution, there is a point where the cost associated with transferring the risk is prohibitive. Downloaded by [University of Defence] at 01:15 24 May 2016

Figure 14.19: Exceedance Probability Curve Exceedance Probability (EP) Curve — Occurrence 10% 350 o 300 9% H Gross 250 8% D Net of 50 xs 50 200

7% O 150 100 6% o O 50 i-3 5% U 0 C 10 20 50 100 250 500 1,000 4% QJ Estimated Return Period w 3%

0% 0 50 100 150 200 250 300 350 400 Loss amount ($ million) Source: Grossi and Kunreuther (2005:138). Risk Assessment and Management Ú 371

In this case, the company will be forced to retain this remote risk. For ease of analysis, the risk profi le illustrated by the EP curve above can be segmented into losses that the company can handle through normal operations, losses that require some level of risk transfer/fi nancing and losses which become economically unfeasible to fi nance. Purchase of catastrophic reinsurance can also be used to keep the risk of insolvency at an acceptable level. However, one needs to determine the reinsurance layer that needs to be purchased and what would be an acceptable price. Suppose a company has premiums of $20 million and a surplus of $20 million. A loss analysis for the company portfolio shows that in order to keep the loss exceedance probability at less than 1 per cent, the company would need to buy an additional reinsurance layer of $10 million, such that when combined with its own surplus and premium, the total amount available to cover a potential loss adds up to $50 million. However, purchase of excess loss reinsurance requires an outlay of (reinsurance) premium, which is an expense that must be factored into the reinsurance strategy. As the reinsurance premium is based on the attachment of the reinsurance layer and the reinsurance limit, fi guring out exactly what reinsurance to purchase is an iterative process. To simplify, suppose the reinsurance premium is $2 million. Th e amount of reinsurance needed is calculated by adding the surplus and premium reserves ($20 million + $20 million = $40 million). Th e reinsurance premium is subtracted from this total ($40 million – $2 million = $38 million). Th erefore, to cover potential losses of $50 million, $12 million in reinsurance coverage will be needed. In reinsurance parlance, the contract is ‘100 per cent of 12 million excess 38 million’. Th at means the reinsurance layer has an attachment at $38 million and a limit of $12 million based on 100 per cent participation from the reinsurer. A lesser participation percentage, as is the common practice, will have a correspondingly higher limit. It should be mentioned that Downloaded by [University of Defence] at 01:15 24 May 2016 the reinsurance purchase is also greatly dependent on market conditions, such as the availability of capital, amount of capacity needed and reinsurance premium.

Underwriting and Risk Selection

Risk modelling is also a valuable tool for underwriting and pricing. By quantifying risk, the impact of adding an account becomes transparent 372 Ú Shah and Dong

and if the potential loss is too large a decision can be made to avoid it. Th is will limit writing certain types of structures like unreinforced masonry buildings and to avoid bad soil locations with a high potential for landslide or liquefaction, etc. In summary, prompt risk quantiﬁ cation enables the agents to select better risk, resulting in a more desirable portfolio. By estimating potential losses and their variability, it also provides a means to determine the appropriate technical premium for a given policy form, account or geographic area, to design insurance structures (deductible, limit and reinsurance venues), and to estimate the proper premium to charge. In general, premium is determined by the expected loss for the layer insured and its variation, measured by standard deviation or variance of loss. Th e former is referred to as pure premium or loss cost, and the latter is used to determine risk load. One such decision module implemented on a desktop computer is shown in Figure 14.20. Th e left-most column denotes the loss line, divided arbitrarily into uniform layers of $9 million each. Th e layers are represented by graphic ‘buttons’ so that the risk manager can select any or all of them for a CAT treaty by clicking the appropriate buttons. Th e second column depicts the AEP curve, which gives the exceedance probability for various loss thresholds. Th e expected loss for each layer is given in the third column, while the standard deviation of the loss calculated is given by the fourth column. It can be seen that higher attachment points will have lower loss cost (the expected loss) and the coeﬃ cient of variation is larger. Upon selection of the layers to insure (by the risk manager), the consequences of such a decision are output in the box on the upper right-hand corner of the interface. In this particular example, Figure 14.20 shows that the fourth and sixth layers (or CAT treaties) have been chosen, corresponding to loss segments of 9 × 27 million (9 million excess of 27 million) and 9 × 45 million (9 million excess of 45 million), Downloaded by [University of Defence] at 01:15 24 May 2016 respectively. Th e expected loss E(L) and standard deviation σ for these two layers are found to be approximately $0.5 million and $2.7 million, respectively. E(L) and σ are the two key variables that govern insurance pricing. A common pricing formula is:

P = E(L) + α.σ + e Figure 14.20 shows the expected loss and standard deviation for diﬀ erent layers (the upper right-hand corner indicates the expected Risk Assessment and Management 373

Figure 14.20: Expected Loss and Standard Deviation for Different Layers

File Analysis Option Help

99 0.03 3,930 1.80,228 Expected Loss 90 0.06 476,150 7,625 §46,950 Standard Deviation 81 0.11 2,724,051 14,595 ^37,376 72 0.24 28,934 63 0.45 58,482 54 0.85 104,562 1911,^24 45 1.59 204,578 1,266,620 36 3.08 371,588 1,710,182 27 5.33 639,060 2,199,671 18 9.33 1,193,962 2,854,567 9 19.18 3,892,257 3,377,047 n 100.00 Layer (Million) Probability (%) Expected Loss Standard Deviation

Source: Dong (2001:147).

loss and standard deviation of loss for two indented layers) where P is the premium and e encompasses the expense and profit margin. The parameter a is the so-called risk load factor (e.g., 20 per cent). Hence, the basic cost of the policy for these two compound treaties is the average annual loss E(L) plus expenses and desired profit, with a load added to reflect variability in the loss estimate. The term 'a-a' is often referred to as the 'risk premium' or 'risk load'. Downloaded by [University of Defence] at 01:15 24 May 2016

Catastrophe Bonds

The engineering modelling approach, which is outlined above, provides a quantitative basis for analysing the natural peril risk associated with any catastrophe bond. The knowledge and expertise required to model a catastrophe bond varies according to its structure. There are three main types, which are reviewed in increasing order of complexity. 374 Shah and Dong

Structure of the Bonds

Parametric Bonds A loss to a parametric catastrophe bond would be triggered simply by the physical parameters of a hazard event, as measured and reported by a specified national or international scientific agency. The trigger, of course, has to be defined in such a way as to match, as closely as possible, the loss to the sponsor of the bond. Indeed, in Ireland, where some special purpose vehicles are established, it is a legal requirement that the trigger is defined with this intention. For a windstorm, the measured parameters would typically be peak windspeeds at a range of locations. A composite trigger index can be constructed as a simple function of the individual windspeed measurements. In order for this index to constitute an approximate surro- gate for loss, the set of windspeed measurements should span the entire region of portfolio exposure. Thus, if the portfolio covers the whole of France, then the windspeed data should be obtained from all available French wind stations. Given a set of peak windspeed values spanning the portfolio region, the question arises how to define an optimal index? The formal procedure is as follows. First, a statistical smoothing procedure has to be applied to interpolate the measured peak wind speed values, so that an entire map of peak wind speeds is obtained. As indicated in Figure 14.21, the peak wind speed at a given location may be expressed as a

Figure 14.21: Spatial Interpolation of Measured Hazard Values

W[l]

W[2] Downloaded by [University of Defence] at 01:15 24 May 2016 W[3]

W[4] W[5]

Source: Woo and Dong (2003:131). Risk Assessment and Management Ú 375

combination of neighbouring peak wind speeds. Various smoothing procedures exist, diff ering in the distance dependence of the interpolation weights. Whichever procedure is favoured, it has to be robust against data acquisition problems such as instrument malfunction. Once the interpolation has been achieved, local damage levels can be estimated in terms of a simple function of local interpolated peak wind speed values. (Typically, a cubic function of peak wind speed is a good indicator of damage.) Summation of this damage function over all local areas then yields a viable regional windstorm index. Th e reliability of a windstorm index may be checked by comparing index values with both modelled loss estimates for the stochastic dataset of windstorms, as well as actual historical windstorm loss values. For an earthquake, the situation is simpler in that a global measure of event size exists in terms of its magnitude, and its geometry is also simply defi ned by its epicentre and focal depth. However, there may be a discrepancy between the actual pattern of ground shaking, and that which is theoretically modelled just from knowing these source parameters, and inputting them into an attenuation relation. In principle, the basis risk might be reduced in future by following the windstorm example, and using spatially distributed data to defi ne a trigger in terms of peak ground-motion values recorded across a network of seismic strong-motion instruments. Such networks already exist in Japan (K-Net) and southern California (TriNet). Whatever the trigger defi nition, the risk analysis for a parametric bond requires an assessment of the frequency and severity of hazard events. An investor is exposed to risks associated with these variables, but not to any measure of damage caused by an event. However, the payout, as a function of the trigger parameters, would be devised so as to minimise the basis risk for the sponsor. Th is is achieved by calculating the expected value of the defi cit between the payout from the bond and the estimated loss to the sponsor. Being dependent solely on physical

Downloaded by [University of Defence] at 01:15 24 May 2016 hazard measurements, the payout can be determined without ambiguity or signifi cant delay. Th e fi rst parametric transaction was appropriately named Para- metric Re, and involved the construction of a simple inner and outer rectangular box geometry around Tokyo. As investors have become accustomed over the past few years to the principles of parametric deals, so the geometries have become more elaborate, both in shape and in the number of boxes. Th e portfolio-specifi c tailoring of triggers allows the basis risk to be lowered to the sponsor’s advantage. 376 Shah and Dong

In parametric transactions, risk analysis focuses on hazard sources. Since portfolio loss distributions are not derived (which can be very computer-intensive), it is computationally feasible to implement a logic-tree representing the uncertainty in modelling hazard sources. Instead of a single best-estimate hazard source model, an ensemble of alternatives are considered, each weighted according to plausibility. Thus, in a California earthquake hazard source model, logic-tree branches might be introduced for the slip-rate on a fault, and for its characteristic magnitude. Figure 14.22 illustrates a simple case with two alternative values S [1] and S [2] for slip-rate, and three alternative values M [1], M [2], and M[3] for characteristic magnitude.

Figure 14.22: Logic-Tree Branches for Modelling the Hazard Contribution of a Fault

M[l]

Slip rate M[2] S[l] M[3] Characteristic magnitude values M[l] S[2] M[2]

M[3] Source: Woo and Dong (2003:136).

Modelled Loss Bonds The basis risk to the sponsor is clearly reduced if the payout is aligned more closely with the actual loss incurred. To avoid investors having to Downloaded by [University of Defence] at 01:15 24 May 2016 bear the moral hazard associated with a changing portfolio, and possible unfavourable post-event loss adjustment, the payout may be defined at the outset as the modelled loss for the hazard event, as computed by a risk modelling firm. On the occurrence of a potentially significant event, the modelling firm would obtain the event parameters, and use these as input into an escrowed catastrophe model to calculate a loss value. The principal reduction, i.e., the payout, would be based on the model output, rather than on any attempt at evaluation of the actual loss. Risk Assessment and Management 377

As shown schematically in Figure 14.23, a continuous modelled loss curve can be approximated by a step-function. Clearly, the more steps there are the closer the approximation will be. There is, however, a limit to investor tolerance of the complexity of a parametric trigger; beyond a certain number of steps in the payout function, a modelled loss curve may be more acceptable.

Figure 14.23: Approximation of a Modelled Loss Curve by a Parametric Trigger Step-Function

Payout (%) Model loss curve

100

Measured hazard event parameter Source: Author.

Indemnity-based Bonds Due to a potential discrepancy between the actual loss to the issuer and the modelled loss, there remains some basis risk in a modelled loss transaction. This may be eliminated if the payout is the actual loss Downloaded by [University of Defence] at 01:15 24 May 2016 experienced by the issuer, as best as it is known. Risk analysis for an indemnity-based catastrophe bond requires detailed information on the issuer's portfolio geography and property characteristics, and is sensitive to the estimation of building vulnerability. The intense scrutiny required of the underlying portfolio to address moral hazard concerns among investors can greatly prolong the time scale for a rigorous risk analysis. This is one of the practical reasons why indemnity-based bonds have become less popular. 378 Ú Shah and Dong

Numerical Risk Simulation

Single peril catastrophe bonds are favoured amongst some investors for their simplicity, which makes the risk analysis easier to grasp. Furthermore, the loss correlation with other catastrophe bonds may be gauged more readily without the need to disentangle the contribution from the individual component perils, as is the case with multi-peril bonds. Even if investors are informed as to the risk contribution of each peril, investors may prefer to mix and match their own bonds. But, however much sponsors are prepared to accommodate the preferences of investors, their own fi nancial requirements must remain paramount, and the issuance of multi-peril bonds may provide precisely the cover that is really needed to fi t with their own reinsurance or retrocession programmes. Apart from investors, risk analysts themselves may be saddled with the challenge of tackling a problem of greater mathematical complexity. With multi-peril bonds, the payout structure may interweave events from diff erent perils in a rich variety of ways. Such structures are well beyond the bounds of direct closed-form computation of probabilities; the only practical way of handling this complexity is via a Monte Carlo stochastic simulation, involving a multitude of alternative hypothetical realisations of possible loss outcomes. A typical form of a Monte Carlo simulation involves the following sequence of steps:

(i) For each of the years of the bond term, the number of trigger events for each peril are sampled. Th e annual count distribution is typically Poisson for the earthquake perils. For European windstorms, which may exhibit temporal clustering behaviour, the Negative Binomial distribution may be used. (ii) For each trigger event, the payout is sampled from the condi- tional peril payout distribution. Downloaded by [University of Defence] at 01:15 24 May 2016 (iii) Th e impact that the sampled events would have on depleting each tranche is evaluated.

From ranking the output payout values of the set of simulations, a probability of exceedance payout curve is generated for each tranche. With any Monte Carlo simulation, especially one probing very low probability levels, the numerical convergence of the results needs to be veriﬁ ed. A million simulations are usually adequate, and are computationally viable. Risk Assessment and Management Ú 379

Pricing

Irrespective of the type of catastrophe bond, risk analysts will provide investors with information on the probability of diﬀ erent degrees of principal reduction. Th e minimum information provided for each tranche is the probability of some loss (the attachment probability); the probability of complete loss (the exhaustion probability); and the expected loss. Th ese are the three key ﬁ gures that govern bond rating and pricing. Lower tranches are exposed to the more frequent hazard events; upper tranches are exposed to very rare (super-CAT) hazard events, or to second or subsequent hazard events. Risk analysts can also calculate the mean and standard deviation of bond return, and so obtain a Sharpe ratio value. An indicator of the increasing familiarity with risk-linked secur- ities as an asset class is the assignment of an A rating to the Class A notes of Atlas Re II. Holders of these notes would have at least two warnings before any reduction of principal. General information about this and other recent issues is provided in Table 14.9.

Table 14.9: Pricing and Risk Exposure Information for Some Recent Catastrophe Bonds, with Ratings by Standard & Poors/Moodys/Fitch

Attachment Exhaustion Expected Issuer Rating: {S&P/M/F} probability probability loss Spread Mediterranean Re: Class A 0.28% 0.17% 0.22% 2.6% {BBB+/Ba3/BBB} Mediterranean Re: Class B 1.47% 0.93% 1.16% 5.85% {BB+/Ba3/BB+} PRIME Capital Calquake & 1.69% 1.07% 1.33% 7.5% Eurowind {BB/Ba3/BB–} PRIME Capital Hurricane 1.46% 1.08% 1.27% 6.5% {BB/Ba3/BB–} Trinom Ltd.: Class A-1 2.42% 0.39% 1.11% 8.0% {BB/Ba2/BB–}

Downloaded by [University of Defence] at 01:15 24 May 2016 Trinom Ltd.: Class A-2 1.01% 0.43% 0.67% 4.0% {BB+/Ba1/BB} ATLAS II: Class A {A/A3} 0.07% 0.03% 0.05% 2.375% ATLAS II: Class B {BB+/Ba2} 1.32% 0.55% 0.90% 6.75% Source: Woo and Dong (2003: 136). Present Value For managers of a fund of catastrophe bonds especially, but also for investors with a minor holding of catastrophe bonds, ﬁ nancial valuation 380 Ú Shah and Dong

should include an assessment of present value, allowing for the cash fl ow. Th e present value of a catastrophe bond is the discounted value of the quarterly coupon payments and the outstanding principal at the end of the term. Included as part of the coupon is the current London Interbank Off ered Rate (LIBOR), which should be at least as large as the discount rate, which is tied to US treasury rates. If there were no natural peril risk, the present value of a bond should thus be just above 100, i.e., at par. If the discount rate were actually the same as LIBOR, the present value of a bond would be exactly 100. Of course, there is a natural peril risk, and investors in catastrophe bonds, therefore, receive a spread above LIBOR, which typically is a multiple of the actuarial risk, as calculated by natural peril analysts. Furthermore, the spread payment (i.e., spread original principal) is guaranteed for the fi rst year of the term, even if it may be lowered in subsequent years with the possible reduction in principal associated with the occurrence of a catastrophe event. Th is guarantee is equivalent to the receipt of an annual reinsurance premium. Historically, reinsurers have formed a major section of the catastrophe bond investor community. Compared with reinsurers, whose core business is natural peril risk, other investors may have felt dis- advantaged by their comparative lack of technical expertise in risk assessment. With a broadening of the investor base, and the provision of specialist technical information to capital markets investors to give them greater understanding and confi dence in the risk analysis, the sizeable spreads off ered by catastrophe bonds may approach the equilibrium values expected by financial economists. In this idealised equilibrium, investors would not be rewarded with a higher spread for accepting risk over which there is a degree of uncertainty. Provided risk analyses are unbiased, the eff ect of such uncertainty on a portfolio should even out. In reality, the spreads are well above the expected loss level. For a

Downloaded by [University of Defence] at 01:15 24 May 2016 BB-rated bond with a 1 per cent annual expected loss, the spread may be around 6 per cent or 7 per cent. Th e present value of a catastrophe bond is a random variable, being dependent on the timing as well as the number and severity characteristics of events that occur. Th rough numerical simulation, a probability distribution for the present value of a catastrophe bond can be readily constructed. Th ere is a small chance that the present value may fall below par, but most likely the present value will be above par. If the above hypothetical BB bond had a three- year term, the expected present value would then be of the order of Risk Assessment and Management Ú 381

118. Stress tests of a risk analysis often involve hypothesising extreme percentiles (e.g. 90, 95 or 99 per cent) for event frequency and severity. But even if extremely pessimistic changes were made to a risk analysis, and even if these pessimisms were compounded, the expected present value would still be within positive territory, well above 100. Bond Portfolio Management For a portfolio of bonds, the task of calculating the probability distribution for present value is best undertaken through event simulation. It is not possible to use marginal distributions for individual bond losses and the bond loss correlation matrix, because these, in themselves, do not determine the joint bond loss distribution. Although comparatively straightforward in principle, in practice event simulation will demand event loss estimation, which may be problematic without access to detailed risk modelling information. Th e lack of correlation of catastrophe bonds with other fi nancial investments has been a persistent selling point. To the extent that natural catastrophes happen without regard to interest rates or stock performance, the diversifi cation potential is very attractive. But holders of catastrophe bonds with overlapping perils will be mindful of the risk of multiple bond losses being incurred in a single event. Th is prospect might reduce the investor risk appetite for California and Japan earthquakes, East coast hurricanes and European windstorms. Innovative types of structures will doubtless off er new challenges in the modelling of catastrophe bonds.

Summary

In this chapter, the various kinds of information that an integrated engineering and ﬁ nancial modelling system can provide have been presented. Th e main aim of this chapter is to show how the currently Downloaded by [University of Defence] at 01:15 24 May 2016 available technologies from loss estimation modelling can be readily utilised to provide business decision models for insurance and reinsurance strategies (for further details see Stojanovski, Shah and Dong 2003). It is important to recognise the role played by risk transfer strategies in meeting the risk mitigation needs of developing societies. It would be a shame not to consider insurance and reinsurance as one of the strategies in a portfolio of many technical, social and ﬁ nancial strategies that are currently available to manage catastrophe risk. 382 Ú Shah and Dong

An overview of how the insurance sector manages catastrophe risks from natural hazards has been presented here. A portfolio manager needs to balance pricing with exposure. Pricing is related to the expected annual loss and its uncertainty. Likewise exposure is related to the loss exceedance probability: the likelihood of a crippling loss must be kept at an acceptably low level. Th e framework based on engineering modelling addresses these and related requirements. Several points are worth repeating. Adequate portfolio risk quanti- ﬁ cation involves not only the expected level of loss but also the associated uncertainty and correlation. Major sources of loss correlation are geographic concentration, site condition, attenuation and vulnerability. In the quantiﬁ cation process, it is also important to recognise that data quality is important; uncertainty or lack of information on the property, site and exposure must be incorporated and accounted for in the risk estimate. Reinsurance is an important tool for portfolio risk management. Th e risk modelling framework discussed in the chapter is essential to use this tool in an optimal manner.

References

American Re. 2002. Annual Review of North American Natural Catastrophes. Princeton, NJ: American Re Corporation. Boissonnade, A. and S. Gunturi. 1994. ‘A Knowledge-Based Computer System for Financial Wind risk Management’. Proceedings of the First Congress, Computing in Civil Engineering, Vol. 2, Washington. Dong, W. 2001. Building a More Proﬁ table Portfolio — Modern Portfolio Th eory with Application to Catastrophe Insurance. London: Reactions Publishing Group. Dong, W., F. Wong, J. Kim and H. Shah. 1988. A Knowledge-Based Seismic Risk Evaluation System For the Insurance and Investment Industries (IRAS). Proceedings 9WCEE, 8, 1083–90, Tokyo, Japan. Grossi, P. and H. Kunreuther. 2005. Catastrophe Modeling: A New Approach to Managing Risk. Boston, MA: Springer. Kreps, R. 1990. ‘Reinsurer Risk Loads from Marginal Surplus Requirements’. Downloaded by [University of Defence] at 01:15 24 May 2016 Proceedings of the Casualty Actuarial Society, lxxvii: 196–203. Kunreuther, H. and R.J. Roth Jr. 1998. Paying the Price — Th e Status and Role of Insurance Against Natural Disasters in the United States. Washington, DC: National Academy of Sciences and Joseph Henry Press. Munich Re. 2002. Topics 2002: Natural Catastrophes. Munich: Munich Re. Panjer, H.H. 1980. ‘Th e Aggregate Claims Distribution and Stop-Loss Reinsurance’. Proceedings of the Casualty Actuarial Society, xxxii: 523–45. Risk Management Solutions (RMS). 1995. What if the 1906 Earthquake Strikes Again? A San Francisco Bay Area Scenario. Topical Issue Series, May. Newark, CA: RMS. Risk Assessment and Management Ú 383

RMS. 1995. What if a Major Earthquake Strikes the Los Angeles Area? Topical Issue Series, September. Newark, CA: RMS. Roth Jr., R.J. and T.Q. Van. 1993–94. California Earthquake Zoning and Probable Maximum Loss Estimation Program. Los Angeles, CA: California Department of Insurance. Stojanovski, P., H.C. Shah and W. Dong. 2003. ‘Bridging the Gap between Earthquake Engineering and Insurance’. Paper presented at the SE-40EEE Skopje Conference, 26–29 August, Skopje, Republic of Macedonia. Woo, G. 1999. Th e Mathematics of Natural Catastrophes. London: Imperial College Press. Woo, G. and W. Dong. 2003. ‘Practical Aspects of Modeling and Pricing Catastrophe Bonds’, in Mike Halls (ed.), Insurance Risk Securitization. London: Reaction Publishing Group. Downloaded by [University of Defence] at 01:15 24 May 2016 Afterword Aromar Revi

Earthquakes, related landslides and tsunamis are among South Asia’s most intense natural hazards. Th is region will be home to close to 2 billion people by mid-century — the largest regional concentration of people in all of human history. Over 1.5 billion of this population will live in India, where approximately two-thirds of the landmass is exposed to geo-hazards, placing a larger population at risk than the Paciﬁ c ‘ring of ﬁ re’ (Building Materials and Technology Promotion Council [BMTPC] 2006). Th e losses caused by a future great Indian earthquake to urban concentrations and densely populated rural areas like the northern Gangetic and eastern Brahamaputra plains, the western coastal plains or in the Himalayas could be devastating in terms of casualties and economic loss (Khattri 1987, 1992). Estimates of 19–50 million people and 5 million buildings at risk from a great Himalayan earthquake are not unreasonable (Bilham, Gaur and Molnar 2001; Khattri 2006) with scenarios of over 0.1 million casualties in a single urban event (Arya 1992; Sinha and Adarsh 1999). Th e Great Aceh earthquake and induced Indian Ocean tsunami (2005) and the more recent Sichuan (2008) earthquake in China demonstrate that large exposed population concentrations, coupled with high vulnerability, can lead to massive casualties and destruction (ADB, UN and World Bank 2005; Telford et al., 2006, World Bank, 2005). In a world where creeping and potentially civilisation-altering climate change has captured the global political and economic imagination, it is important not to lose sight of the challenge posed by the rapid

Downloaded by [University of Defence] at 01:15 24 May 2016 onset of disasters like earthquakes. An alertness to natural hazards like earthquakes, which has permeated cultures like Japan and to a lesser extent Iran, are prerequisites to establishing an institutional and cultural basis for successful climate change adaptation (Revi 2008). Th e capacity to mitigate earthquake risks will also contribute to multi-hazard risk reduction, including that of climate change. India has a long way to go in this direction. A ﬁ tful start has been made in some states, following major earthquake casualties and losses in the 1990s and the early 2000s. Equally important is the slow Afterword Ú 385

institutionalisation of Disaster Risk Reduction (DRR) across multiple levels of government and the attempt to integrate the national development, poverty reduction and disaster risk reduction agendas (Planning Commission 2002; International Strategy for Disaster Reduction [ISDR] 2009). Early earthquake recovery and rehabilitation interventions following the Latur (1993), Jabalpur (1997) and Chamoli (1998) earthquakes have given way to a new generation of initiatives that focus more substantially on preparedness and mitigation following the Kachchh earthquake (2001) and the Indian Ocean tsunami (2004). Th e creation of the National Disaster Management Authority (NDMA) and a few exemplary State Disaster Management Authorities (e.g., in Orissa and Gujarat), a clutch of ﬁ scal instruments and a dedicated stream of resources within India’s planning framework have made an impact in the more receptive states. Nevertheless, progress continues to be slow in many regions due to hazard exposure and vulnerability. Th e opportunity to leverage DRR to mitigate income and asset shocks that drive households into poverty and dependence has yet to permeate into the priorities of the political class, bureaucracy, private entreprenuers and most communities who continue to be ﬁ xated on the more immediate aspects.

Global Earthquake Impact

Earthquake risk is important to address due to its disproportionate share in global risk given its relatively low probability. Earthquakes made up nearly half (11 out of 24) of the global mega-disasters in the period 1975–2008 (EM-DAT 2008). If the 2005 Indian Ocean tsunami is added to this list, earthquake-induced casualties worldwide are in excess of 0.8 million (45 per cent of the 1.8 million killed in these events). Th e most devastating recent earthquake in Tangshan, China (1976) killed

Downloaded by [University of Defence] at 01:15 24 May 2016 over 0.24 million people and destroyed much of a city. Th e Great Aceh (2004) earthquake and induced Indian Ocean tsunami killed 0.23 million people across nearly a dozen countries, including India’s Andaman & Nicobar Islands, and the Tamil Nadu, Andhra Pradesh and Kerala coasts. Death is only one of the dramatic impacts of earthquakes. Between 80 and 130 million people are aﬀ ected annually by them worldwide, compared to 83 million by cyclonic storms and 53 million by ﬂ oods (ISDR 2009). Th is makes earthquakes a leading cause of economic disruption caused by natural disasters globally. 386 Ú Aromar Revi

Earthquake-induced losses can also be high, especially in OECD countries because of the large monetary value of damaged physical capital and induced economic dislocation. Th ree of the top ﬁve mega-loss events globally were earthquakes: Kobe, Japan (1995) is at number three at US$ 100 billion; Sichuan, China (2008) at number four at US$ 30 billion and Chuetsu, Japan (2004) at US$ 30 billion at number ﬁ ve (EM-DAT 2008). Th e large value of economic loss in China is indicative of the potential damage by a major future event in India.

Earthquake Impact in South Asia

Th e Himalayan region has experienced four major earthquakes since the 1890s. Th ey include Shillong (M8.7 in 1897), Kangra (M8 in 1905), Bihar–Nepal (M8.3 in 1934), and Assam–Tibet (M8.6 in 1950). Yet, the impact of earthquakes in South Asia and India has been relatively low over much of the twentieth century compared to other areas in the world. Th is is largely due to the Himalayan ‘seismic gap’ — the absence of a great earthquake in the region for close to 75 years since the great Bihar–Nepal event of 1934, in spite of the continuing strain build-up along the Indian plate boundary (Khattri 1987, 1992; Bilham, Gaur and Molnar 2001; Bilham 2004). Th e bulk of the Indian peninsular shield was classiﬁ ed as seismically stable till the Koyna (1967) earthquake and the Latur (1993) event (Bapat et al. 1983). Th is has led to a reconsideration of the dynamics of intra-plate events and a subsequent seismic rezonation of western and central India (Bilham, Bendick and Wallace 2003). Over 36,000 people have died in India in 10 major earthquakes since the 1990s (EM-DAT 2008). India’s most intense event — the Kachchh (2001) earthquake, with over 20,000 casualties is ranked eighteenth among global mega-disasters of the last 30 years (EM-DAT 2008). Th e most devastating recent earthquake in South Asia was in Kashmir

Downloaded by [University of Defence] at 01:15 24 May 2016 (2007). It killed over 1,400 Indians, which was a small number compared to the 79,000 casualties reported on the Pakistani side of the Line of Control. A large proportion of the dead were children who had been victims of the collapse of highly vulnerable, RCC school buildings. South Asia has an anomalous temporal earthquake signature that poses a practical challenge to eﬀ ective DRR in India. Th e region, unlike many others, has a relatively high number of high intensity earthquakes spread over long lengths of time, along with a relatively high frequency of moderate intensity events. Since the moderate intensity earthquakes Afterword Ú 387

cause exponentially lower damage and destruction than major events — earthquakes typically slip from public memory and priority. Th is is rather diff erent from a drought, for instance, which has had a well developed institutional response mechanism since the late nineteenth century because of near annual frequency. As a consequence, earthquake DRR has largely been a matter of post-event fi re fi ghting, with few systematic institutional responses till the late 1990s. A 75 to 150-year return period for a major earthquake event almost certainly ensures that post-disaster social and institutional resilience degrades or becomes irrelevant because of technological, economic and demographic change. In the relatively stable peninsular and coastal regions with multi-century earthquake return periods and no recall of tsunami events, the challenge of almost no institutional or cultural memory is even more severe. Th is was experienced during the Latur (1993) earthquake when traditional buildings that had provided some residual seismic resistance and saved a number of lives were demolished and replaced by modern structures that were not liveable, and became risk-prone when traditional technologies were used to extend them (Revi 1994).

Differential Earthquake Impact and Risk Sharing

Earthquakes are intensive rapid-onset events that primarily impact capital stocks in the form of buildings and lifeline infrastructure, often requiring large post-event replacement investments, unlike drought that has almost no impact on physical capital. Th ey can also cause intense loss of life, and accompanying trauma injuries, some of which could lead to permanent disability. Earthquakes also impact economic output via damage to buildings, plant and machinery and infrastructure services, often leading to interruption of manufacturing, trade and commerce.

Downloaded by [University of Defence] at 01:15 24 May 2016 While mitigation investments are slowly being accepted as an economically and fi nancially eff ective way of addressing earthquake risk, this is poorly linked to current anti-poverty programmes and infrastructure and asset development plans. Th is convergence, if eff ected, could reduce the incremental costs of seismic DRR considerably. Since much of the Indian population and unorganised enterprises are uninsured, the brunt of earthquake loss has historically been borne by the state in the form of formal fi nancial ex gratia payments or grants. Enterprises and households absorb the balance of the income 388 Ú Aromar Revi

and capital shocks via coping responses (de Fuente, Felipe and Revi 2008). Severe earthquakes can lead to economic disruption and potential regional macro-economic output compression, especially in the secondary sector, due to infrastructure and facilities damage (TARU 2005). Major South Asian earthquakes typically recur in particular locations at relatively long intervals. Since they are low probability events, stochastic estimates of earthquake intensity are necessary to establish a rational and equitable sharing of risk between households, enterprises and the state. Only rudimentary steps have been taken in this direction in India. If further enabled, this could facilitate the transfer of earthquake- induced risks from vulnerable households, establishments and fi rms to appropriately hedged insurance portfolios. A fundamental challenge is that most vulnerable Indian households do not have access to formal fi nancial channels, and even those that do have little to off er in the form of collateral. Many vulnerable households often land in debt — dragging them into poverty — even after receiving government ex-gratia grants and liquidating whatever little they own by way of assets. Th e role of earthquakes in forcing people into poverty has not been strongly established in India (ISDR 2009). It has however been established as an important factor in increasing household vulnerability (Jigyasu 2002).

Seismic Risk Reduction Priorities

Exposure and vulnerability are more important drivers of earthquake risk to life, capital and economic activity than hazard risk in India (Revi 2008). Th is is slowly being appreciated by Indian development planners and disaster mitigation specialists. Th e current Government of India DRR thrust in the multi-hazard risk districts of India is a practical way to take this perspective forward. It also seeks to ensure cross-cutting contingency planning, preparedness and mitigation which will reduce

Downloaded by [University of Defence] at 01:15 24 May 2016 earthquake risk. Seismic risk reduction in India will also be enabled by rising per capita incomes and the resulting transformation of building stock. Th e bulk of this risk reduction however, will probably happen via a mix of greater public awareness; creation of a series of market incentives and public investment programmes; building a cadre of trained professionals and artisans to enable implementation; and the deepening of school and university education around DRR to mobilise support on the demand side. Afterword Ú 389

Earthquakes are virtually impossible to predict. Hence, early warning is not an eff ective mitigation measure. Th e core seismic risk mitigation strategy should therefore be focused on vulnerability reduction. Th is could be achieved by improving building and lifeline infrastructure performance, establishing an eff ectively enforced techno-legal regime that enables decentralised risk reduction; creating a response and recovery support system that is eff ective and planned using fi ne-grained composite risk assessment; and developing a multi-level preparedness programme.

Hazard Risk Zonation

Th e fi rst step in implementing an eff ective seismic DRR strategy is to conduct a fi ne-grained assessment of hazard risk across India. Con- siderable improvements can be made in India’s current seismic zonation by shifting from an experience-based to a probabilistic hazard risk assessment framework as demonstrated by Global Seismic Hazard Assessment Program (GSHAP) at low resolution and the Gujarat Risk and Vulnerability Atlas at high resolution (Bhatia et al. 1999; TARU 2005). A probabilistic seismic risk zonation that links the earthquake return period to the mean service life of elements at risk is the most economically effi cient mechanism of assessment. Th is would enable a much better fi t between estimated hazard risk and the service life of buildings and infrastructure, reducing the opportunity of underdesign and overdesign, which are both common in India. For buildings, this would match seismic intensity estimates for 50-, 100- and 200-year return periods with structures of varying importance and service lives, from semi-engineered buildings to critical infrastructure like power stations and major bridges.

Earthquake Preparedness Downloaded by [University of Defence] at 01:15 24 May 2016 Since earthquakes provide little advance warning, response systems in many of the poorer and more remote parts of India can take days if not weeks to swing into action. Hence, community-led preparedness is the ﬁ rst line of defence for most of the poor and vulnerable. Th is will need to start at household and community level, and proceed via settlement, block/mandal and district to the state and national levels. A fair degree of experience has been gained in community-led preparedness especially after the Kachchh earthquake, in which civil 390 Ú Aromar Revi

society organisations (CSOs) and people’s organisation (POs) led the initial phase of relief and response, making it one of the better managed reconstruction programmes in the country. Th e Indian Ocean tsunami (2004) and the Kashmir earthquake (2007) rehabilitation programmes have led to a widening of the geographic coverage of such innovative initiatives. Unfortunately, the regional depth and breadth of these initiatives are dwarfed by India’s sheer scale and diversity. A number of positive school safety and preparedness initiatives have been taken by State governments and local authorities. Th ese may be possible to scale-up across the country at low incremental cost, via the Sarva Shiksha Abhiyan (SSA) programme, but the linkage between education and DRR at the state level will need to be strengthened to enable this. Much of the DRR institutional response capacity is located at national and state levels, concentrated in major cities. Th e civil administration has typically depended on paramilitary and armed forces back-up when the scale of a disaster gets out of hand. Th e establishment of the NDMA altered this situation, with the creation of rapid response teams located at strategic positions across the country. Yet, to be eﬀ ective, preparedness needs to be embedded in local institutions and community practices. Th is implies building a close linkage between DRR processes and ongoing Panchayati Raj and ULB-led decentralisation processes. Preparedness and contingency planning for urban disasters is a crucial area that needs to be developed. Th ere are some indications that Delhi may break new ground in this area in preparation for the 2010 Commonwealth Games.

Strengthening Emergency Response Systems

Major earthquakes (of magnitude M6 and above) usually result in a

Downloaded by [University of Defence] at 01:15 24 May 2016 large number of casualties. Most are due to the collapse of buildings, with a typical death to injury ratio of 1:3 to 1:5, depending on the location and type of structure (TARU 2005). Th e leading causes of injury are impact and crushing by moving debris. Much of the mortality and disability could be reduced if adequate search and early medical response at the disaster site were available. Th e NDMA is making positive eﬀ orts in this direction, by establishing rapid response systems and designated trauma care centres. Some states like Gujarat and Andhra Pradesh have established an elaborate multi-level response infrastructure Afterword Ú 391

based on stochastic risk assessments all the way to the settlement level (ibid. 2007). Th e current level of DRR-related medical awareness is abysmally low across much of India’s high seismic risk regions. An eﬀ ective and responsive trauma care system needs to be established in all tertiary healthcare facilities, particularly those located in Seismic Zones III to V. Most tertiary hospitals and critical care units in seismic hotspots are themselves at risk of earthquake damage. Th is must be addressed expeditiously. Given the dominant share of non-engineered buildings in the rural and urban areas of India, it is not surprising that casualty levels are sometimes of an order of magnitude higher than in similar magnitude events in countries like Japan. A recent complication is the rapid growth of partially engineered and poorly detailed multi-storey and RCC frame- and-slab buildings in urban and peri-urban areas, which are often subject to catastrophic failure. Search and rescue operations in such destroyed or damaged buildings are a serious challenge in terms of equipment and personnel as observed during the Kachchh and Kashmir earthquakes. Th e lack of enforceable regulation and the absence of liability of engineers and builders makes the repetition of serious structural errors common, as observed in areas aﬀ ected by both the Uttarkashi (1991) and Chamoli (1998) earthquakes (TARU-BMTPC 1998).

Mitigating Structural Vulnerability

India’s stock of over 240 million buildings is slowly transforming itself in response to rising per capita incomes and savings, especially in urban areas. Over one-third of these buildings are constructed in materials that are highly vulnerable to earthquake damage being built of earth, stone and sometimes biomass walls which provide little lateral restraint to seismic forces. In metropolitan centres, where a third or more of the Downloaded by [University of Defence] at 01:15 24 May 2016 population lives in informal settlements in cities like Mumbai or Delhi, millions are at risk to ground shaking, liquefaction and the ﬁ res that often accompany a major earthquake in hyper-dense settlements. Many of the multistorey framed RCC buildings in so-called modern materials perform not very diﬀ erently from non-engineered, artisan built buildings because of poor material quality, workmanship and seismic detailing, as demonstrated by the severe damage to high-rise buildings in Ahmedabad, over 250 km from the epicentre of the Kachchh (2001) earthquake. 392 Ú Aromar Revi

The upgradation of existing buildings using well-established strengthening and retrofi tting measures has been demonstrated to be technically, fi nancially and economically viable at scale, fi rst in a rural context in Latur (1993), then at a small scale in Jabalpur city (1997) and fi nally at a massive scale following the Kachchh (2001) earthquake. A number of lessons have been learnt from these events, ranging from an emphasis on the use of local building materials, skills and details; the introduction of catalytic amounts of reinforced concrete and steel in strengthening and retrofi tting; simplifi ed details and specifi cations that can be implemented by semi-skilled labour; establishment of training and quality assurance systems for large numbers of building artisans and trades people to confi dence building measures that demonstrate the eff ectiveness of seismic strengthening via fi eld shake-table tests (National Centre for Peoples’ Action in Disaster Preparedness [NCPDP] 2008). Th e challenge, however, in such a large country is that these experiences are diffi cult to institutionalise and capture at scale without a signifi cant ongoing national programme of seismic strengthening and education. Large-scale public housing programmes are an area in which considerable progress can be made, not only to reduce seismic risk but to simultaneously rationalise material use and improve quality via better design at a small increment of the current unit costs (Arya 2000). Retrofi tting of a few public buildings was taken up in Gujarat and more recently in Delhi. Th is includes a few heritage buildings. But these eff orts have typically been pilot tests, as the priority given to such long-term DRR actions is low in the absence of the impetus of a major disaster. Retrofi tting and strengthening need to be scaled-up to a programme enforced by a tight techno-legal regime as in California. Modifi cation and upgradation of existing Bureau of Indian Standards (BIS) Codes over the last decade is a major step forward. So also is the training of an increasing number of engineers and a lesser number of architects in earthquake resistant design (Jain and Nigam 2000). Con- Downloaded by [University of Defence] at 01:15 24 May 2016 ditionalities imposed by building insurance and housing fi nance agencies have also been useful in promoting the design of seismically resistant buildings, but have little or no impact on mass housing. A primary challenge, however, is infl uencing the design, detailing and quality of construction of 60 to 75 million new buildings that will be added to the national stock over the next 30-odd years, primarily in urban and larger rural settlements. An improvement in their disaster resistance in general and seismic resistance in particular would be an important step forward in building a safer India. Afterword Ú 393

Th e introduction of disaster management into the school curriculum via a Supreme Court directive has increased awareness across the country, but very few details of earthquake and disaster resistant construction are actually provided in these textbooks. Greater emphasis on public education and awareness building could help provide a ﬁ llip to future demand-side interventions.

Strengthening Lifeline Infrastructure

Most recent Indian earthquakes have not caused catastrophic damage to infrastructure except in the core epicentral tract where serious damage to bridges, roads, power, airport structures and water systems have been reported (TARU 1993, 1998, 2001b; World Bank 2002). Th is will probably not be true for a future great earthquake strike in the Himalayas or a region of high population concentration. Hence, a national seismic vulnerability reduction mission should take on the task of addressing this seismic defi cit, especially in Seismic Zones IV and V. Th e Indian Railways system is probably at signifi cant risk because a high proportion of its bridges and culverts are well beyond their service life and are not designed according to current seismic standards. Th is is a largely unexplored challenge that will have huge capital investment implications, but also possible national security impacts if signifi cant periods of network interruption take place. A portion of bridge and culvert systems of most State highway systems, district and village roads are highly vulnerable in Seismic Zones IV and V. Th e government’s large-scale village road building programmes are reported to have lower seismic standards to expand coverage (Indian Roads Congress [IRC] 2002). Although they have not been tested by a major event except by the Kachchh earthquake (that led to some short- term bottlenecks) the national highways, the Golden Quadrilateral (GQ) and North South East West (NSEW) corridor expressway systems

Downloaded by [University of Defence] at 01:15 24 May 2016 will hopefully perform adequately after a major earthquake. Landslide disruptions, especially in the Himalayas, are a common challenge that needs to be addressed. It would therefore be ideal if a systematic seismic and structural audit of critical national road infrastructure were undertaken on a rolling basis every ﬁ ve years, to develop investment and strengthening plans that could be linked to the ﬁ ve-year planning cycle. Airports have become the cutting edge of India’s rapid disaster response system via the airlifting of paramilitary and armed forces 394 Ú Aromar Revi

response teams, airlifting of casualties to specialised trauma care centres and the establishment of emergency command and control systems. Yet, most regional airports may not survive without severe damage following a major earthquake. Th e recent upgradation of airport infrastructure in larger metropolitan centres may be a step in the right direction, but a public hazard audit of these facilities is in order. A number of new generation large dams in the Himalayas are reported to be subject to dam-break and seismic stability analysis, yet there is considerable controversy about their safety (Gusain et al. 2007; Dharmadhikari 2009). Many older structures and a large number of medium and small dams, micro-watershed structures and their associated irrigation networks that have proliferated across the country are not adequately designed to meet the upgraded seismic risks in many parts of the country (TARU 2005). Th e catastrophic failure of these systems would not only threaten the livelihoods of large numbers of people dependent on irrigation, but potentially disrupt water supply to towns and villages in many semi-arid regions. Much of India’s extensive cultural heritage has not been assessed for seismic risk. As a result few mitigation measures have been put into place. With the expected development of the tourism and hospitality industry, this requires focused sectoral attention. Power transmission and distribution systems suﬀ ered considerable damage during the Latur and Kachchh earthquakes, even though not much damage was sustained by the generation stations. Th is led to substantial revenue losses, which were probably larger than the cost of replacing the damaged capital assets. Th e raising of standards and alteration of design details for power systems in Seismic Zones IV and V may be in order (TARU 2001b, 2005). Overhead telecommunication infrastructure is often damaged by earthquakes. The shift to fibre optic networks and wireless transmission is not only cost-eﬀ ective but also reduces seismic vulnerability and reduces the down-time to recovery. Downloaded by [University of Defence] at 01:15 24 May 2016 Urban environmental and energy infrastructures, especially water, sewage and waste-water trunks systems, gas pipelines and power transmission systems are at high seismic risk in many locations. Th is is partly because of haphazard planning, lack of asset inventories, their extreme age and disuse of networks in many cities and the lack of appropriate seismic standards. India has had an impressive record of developing building standards since the 1970s, but little has been done with regard to critical infrastructure, in spite of considerable international experience from Afterword Ú 395

major earthquakes in the US, Japan and China. A major earthquake striking a large metropolitan centre could lead to the scale of infrastructure damage and economic disruption that was experienced within the epicentral tract of the Kobe earthquake at much lower ground-shaking intensity, due to the high vulnerability of large segments of these urban infrastructure networks.

Enforcing a Techno-Legal and Techno-Financial Mitigation Regime

Disaster risk reduction is based on a set of premises that society and its lead institutions (i.e., the state system, communities, markets and private enterprise) will act in order to reduce direct, indirect and external social and economic costs as much as possible. Th is includes the reduction of composite risk to both economic output and capital assets within the prevailing risk preference set. Th is is insuffi ciently realised because of many factors including: structural and political constraints; institutionalised inequity, diff erential incomes and asset holdings; entitlement and information asymmetries; multiple and diverging time horizons and risk preferences; and the lack of institutional capacity and knowledge to act on risk information, if it were available. A structured techno-legal and techno-fi nancial regime is a high-level attempt at codifying principles and instruments that could address some of these constraints and establish a hierarchy of processes, incentives and standards to mitigate risk in a multi-stakeholder environment. Parts of a comprehensive regime have been put into place in Andhra Pradesh, Maharashtra and Gujarat, with varying levels of success. Th e design and enforcement of an appropriate techno-legal regime that addresses India’s multiple challenges of aff ordability, implement- ability and transparency is a challenging proposition. Issues that need to be addressed include: planning guidelines and processes; sustainable

Downloaded by [University of Defence] at 01:15 24 May 2016 environmental service provision; land-use and land-cover controls; infrastructure and building performance specifi cations and strengthening measures; and most important simple means to incentivise and enforce their implementation. Building enough professional skills and the institutional capacity to design, regulate and be able to implement this regime will not be easy. Defi ning the responsibility and liability of designers and builders to deliver appropriate performance specifi cations in an aff ordable manner is another signifi cant challenge. 396 Ú Aromar Revi

Building and infrastructure fi nance and insurance agencies have an important role in providing appropriate incentives to the application of cost-eff ective seismic resistant construction technology to current and future buildings and projects. Appropriate conditionalities and incentives for loans for new earthquake-resistant houses, and extensions may provide a fi llip to the wider application of earthquake-resistant technology at an aff ordable cost.

Community-Led Disaster Risk Reduction

Eff ective post-disaster response and recovery services are available in only a few States of India and with the NDMA central rapid response pool. Community-led DRR is therefore the only eff ective response option for many remote and inaccessible high seismic risk areas. Systematically building up these local capacities and providing enabling support via public programmes is probably the best way for the more remote areas of Jammu and Kashmir, Himachal Pradesh, Sikkim, Arunachal Pradesh, Assam and other north-eastern states, and even some border areas of Gujarat. PRI’s capacity to support community-led DRR around preparedness and mitigation has been eff ectively demonstrated after the Kachchh earthquake and the Indian Ocean tsunami. In urban areas, however, considerable eff ort will have to be made to bridge the gap between ULBs, community mobilisation and access to adequate information and technical assistance to enable seismic risk reduction.

Educating and Training a new Generation of Practitioners

A serious challenge to implementing earthquake risk reduction in India is the paucity of trained practitioners — all the way from building artisans to earthquake engineering specialists. Th is is especially true in some Downloaded by [University of Defence] at 01:15 24 May 2016 of the less developed and more remote regions of the country, like the Himalayas, where seismic risk is very high. Some progress was made on establishing and extending earthquake engineering education and training following the Kachchh earthquake. Th e NDMA has catalysed a major professional training initiative around seismic strengthening. Earthquake engineering, training, education and documentation have been taken forward as a mission by a mix of quasi-governmental (e.g., BMTPC), academic institutions (e.g., IIT- Roorkee, IIT-Kanpur, IIT-Bombay and the National Information Centre Afterword Ú 397

of Earthquake Engineering [NICEE]) and professional institutions (e.g., Indian Society of Engineer’s and Technicians [ISET]). Nevertheless, the scale of the outreach required to impact the practice of the majority of engineering, architecture and contracting practitioners is well beyond their capacity. Th e involvement of key trade and business associations and their regional chapters will be necessary to fund and enable this. Central and State governments will need to develop a massive programme of training across their large engineering and design cadres. India’s technical education and distance education system will also need to develop a mixture of post graduate, short-term refresher and distance courses to reach out to new and in-service learners. On-the-job training of masons and building artisans is currently the weakest link in the earthquake risk mitigation chain, as they are the largest producers of buildings in the country. Th is is a challenge because of the economic compulsions that constrain them from formally upgrading their skills, along with the seasonal and migratory nature of their occupation. Nevertheless, eﬀ ective training programmes have been developed and demonstrated to scale after the Latur (1993), Chamoli (1999), Kachchh (2001), Indian Ocean tsunami (2004) and Kashmir (2007) earthquakes (NCPDP 2008). A systematic programme of training and sensitisation needs to be designed for India’s multi-hazard prone districts.

Urban Earthquake Risk Mitigation

India has little experience in urban disaster management and mitigation. Th is was evident from the weak response and rehabilitation intervention following the Jabalpur (1997) earthquake. Th e considerable early confusion that surrounded the devastation of Bhuj and Anjar and the chaos created by widespread damage to multistoreyed buildings in Downloaded by [University of Defence] at 01:15 24 May 2016 Ahmedabad following the Kachchh earthquake were also reﬂ ective of this. After a hard struggle, new standards were subsequently set for urban rehabilitation and reconstruction in Gujarat. Th is experience needs to be documented and disseminated much more widely if cities like Delhi or Mumbai which are at considerable seismic risk are to learn and move forward. India would do well to learn from the successful urban DRR experience of Iran, China and Nepal. 398 Ú Aromar Revi

Economic Risk Mitigation

Earthquakes rank between the second and third most destructive long-term natural hazards in India, after droughts and cyclonic storms depending on the region of interest (BMTPC 2006). Unlike other natural hazards that the Indian state has learnt to respond to since the mid- nineteenth century (i.e., droughts, fl oods and more recently cyclonic storms and surges), earthquakes strike with little or no warning. Th ey can have a devastating impact on buildings and infrastructure, decimating in minutes the entire investments and savings that many generations of poor people and small and medium enterprises have put together. Th is is not widely recognised within India’s planning apparatus. Only Gujarat has undertaken a systematic and probabilistic assessment of seismic risk to its economy and started making appropriate mitigation investments (TARU 2005). Lessons from this experience could be extended to other high-risk states, but will require a systematic scientifi c and fi eld-based vulnerability assessment, before clear policy conclusions and programmed interventions can be developed.

Developing Market-Based Risk Sharing Instruments

India’s size, population and level of assets exposure to seismic risk makes it prudent to prepare for a signiﬁ cant economic shock if a great earthquake were to strike a large population concentration. Th e current sets of risk transfer instruments that can enable this are rather blunt. Th ey range from a pool of contingency funds created by the Finance Commission, and large, concessional and multilateral credit that is typically sought as humanitarian reconstruction assistance. While this may have been adequate in the past when exposure levels and vulnerability were less extreme, this could be ﬁ scally untenable in the future.

Downloaded by [University of Defence] at 01:15 24 May 2016 Hence, for both public assets and private enterprises, an attempt should be made to assess and transfer risks away from the public exchequer to market-based mechanisms. Th is may be more effi cient than relatively arbitrary bureaucratic decision-making. Th e challenge, however, is to reach poor and vulnerable households who are still largely outside the formal fi nancial system. An emphasis on fi nancial inclusion (e.g., via self-help groups (SHGs) and incentivisation of earthquake resistant construction should be an important policy priority in multi-hazard districts, especially those with high seismic risk. Afterword Ú 399

Institutional Development

Th e institutionalisation of policies and earthquake disaster reduction measures in India has proceeded in ﬁ ts and starts, punctuated by major earthquakes. Th e Latur earthquake led to the creation of a range of grass-roots alliances, and district- and State-level DRR institutions in Maharashtra. Th e Kachchh earthquake marked a watershed in the institutionalisation of DRR in India with the creation of a state authority with overarching powers and responsibilities in Gujarat with the GSDMA and a unique public–NGO–PO partnership in the NGO Abhiyan which other states are considering emulating. Th e creation of the NDMA notionally created a seamless DRR hierarchy from the district, through the state to the national level. Th e Indian Ocean tsunami led to the development of an innovative public–private–international partnership to deliver a decentralised programme over a large area. Th e primary locus of DRR institutional development is the States, potentially based on prior experience and institutional innovations and ‘best practices’ from other parts of the country. Each State would need to establish a nodal DRR agency, preferably a Disaster Management Authority (DMA), with the mandate, power and institutional capacity to develop, coordinate and manage a comprehensive DRR programme across multiple sectors and line departments. Th e DMA would need to develop an operational framework that would enable the convergence of ongoing development and anti-poverty programmes, State schemes and additional central assistance to meet its DRR objectives. A State-level research and training institution will also be needed to address regional technical challenges, create locally relevant content in the key State languages, and help monitor the implementation of the techno-legal regime. A hierarchy of response and recovery institutions will need to be created at the district level and linked to the PRI/ULB system to help support community-led DRR activities through training,

Downloaded by [University of Defence] at 01:15 24 May 2016 sensitisation, information sharing and partnerships with NGOs and POs. A National Seismic Safety Initiative that is focused on the highest seismic risk districts of the country needs to be given an additional ﬁ llip and some serious funding. Th is could help institutionalise the process of risk and vulnerability assessment and contingency planning, including micro-zonation where necessary; build capacity to design and manage building strengthening and retroﬁ tting and infrastructure; promote community-based DRR; and enforce an established techno-legal and 400 Ú Aromar Revi

techno-ﬁ nancial regime in collaboration with locally situated technical, banking, insurance institutions and local planning agencies.

Learning from Others

Earthquake DRR in India has historically lurched from one event to the next, with a hiatus in which most of the innovation and learning from one earthquake is lost before the next one strikes. Th e most striking example was the repetition of many of the construction and programming errors made following the Uttarkashi earthquake in the Chamoli earthquake in the same State of Uttar Pradesh. Th is is partly because of the declining quality of governance in many States and the limited institutional capacity of State agencies that is often focused on immediate ﬁ re-ﬁ ghting rather than systemic risk reduction. India has also long considered itself a leader in earthquake engineering and research and has therefore tended to be insular in learning from the considerable experience of other countries that has built up internationally since the Skopje (1963) and Tangshan (1976) earthquakes. Even in its neighbourhood, India has a lot to learn in terms of institutional development, capacity building, implementation of large-scale DRR programmes, innovative public–private–community partnerships and building a culture of preparedness and mitigation that is backed by both resources and political interest. China, Iran, Pakistan and Nepal have interesting experiences that may be worth adapting to the Indian context and frame of governance.

Joining up DRR with Development Planning

India’s Tenth Five-year Development Plan made a signiﬁ cant departure from the past by including a special chapter on disaster management (Planning Commission 2002). Th is attempted to join up the imple-

Downloaded by [University of Defence] at 01:15 24 May 2016 mentation of DRR with development planning and poverty reduction programmes. Th is was prodded by the spiralling of public expenditure on post-disaster recovery and reconstruction and the recognition that major natural disasters could cause considerable human, economic and social losses. Th is process was unfortunately weakly reﬂ ected in the State Plans which are the frontline instruments to operationalise policy and ring fence resource ﬂ ows. Th e Eleventh Plan document made little progress in mainstreaming DRR, which continues to be an additionality, largely funded by multilateral assistance programmes. Afterword Ú 401

Th e convergence of a large number of public programmes on buildings, infrastructure and decentralised institutional capacity building at settlement, block and district levels may be necessary to meet the goals of earthquake DRR. A necessary condition for this is the preparation of high-resolution block and the state-level risk and vulnerability assessments and action plans on which states like Gujarat have taken a lead (TARU 2005, 2007). Using these tools, more eﬀ ective planning can be facilitated in close consultation with PRIs, ULBs and potential local partners. Building on this at the state level, planning for risk mitigation could be taken into consideration during macroeconomic planning and the preparation of the Annual and Five-year Plans. A series of strategic instruments to enable risk sharing and transfer between households, private enterprises and the state sector need to be developed. Th is would help unload the burden of making up capital losses from public revenues and slowly shift it towards market-based instruments.

Conclusion

Earthquakes are a highly underrated risk in India. Th is is largely because we have been living in a ‘seismic gap’ with no great earthquake for 75 years. Such an event could impact millions of people if its epicentre were located near a large and vulnerable population concentration. It could also cause serious regional macroeconomic shock and possibly push a large proportion of local and aff ected, poor and vulnerable people into poverty, in spite of the considerable growth in DRR recovery and rehabilitation capacity within the Indian state system. Seismic risk reduction can be successfully enabled by a mix of: building vulnerability reduction, strengthening lifeline infrastructure, developing and enforcing an eff ective techno-legal and techno-fi nancial regime, strengthening community-led risk mitigation, developing Downloaded by [University of Defence] at 01:15 24 May 2016 a new generation of appropriately skilled practitioners and a range of institutions to support the implementation of a multi-dimensional process. A core strategy of making this politically and economically viable is to enable the convergence of ongoing anti-poverty, infrastructure and development programmes to meet DRR goals and thereby reduce the volume of human and fi nancial resources required for their implementation. 402 Ú Aromar Revi

References

Arya, A.S. 1992. Damage Scenario of M 8.0 Hypothetical Earthquake Occurrence in Himachal Pradesh. ISET Bulletin. ———. 2000. ‘Recent Development towards Earthquake Risk Reduction in India’, Current Science 79 (9 & 10): 1270–77. Asian Development bank (ADB), United Nations (UN) and World Bank. 2005. India Post Tsunami Recovery Program Preliminary Damage and Needs Assessment. New Delhi. Bapat, R. C., K. Kulkarni and S. K. Guha. 1983. Catalog of Earthquakes in India and Neighborhood from Historical Period up to 1979. Roorkee: Indian Society of Earthquake Technology. Bhatia, S.C., M.R. Kumar and H.K. Gupta. 1999. ‘A Probabilistic Seismic Hazard Map of India and Adjoining Regions’, Annali di Geoﬁ sica 42(6): 1153–64. Bilham, R. 2004. ‘Earthquakes in India and the Himalaya: Tectonics, Geodesy and History’, Annals of Geophysics 47(2): 839–58. Bilham, R., R. Bendick and K. Wallace. 2003. ‘Flexure of the Indian Plate and Intraplate Earthquakes’, Proceedings of the Indian Academy of Science (Earth and Planetary Sciences) 112(3): 1–14. Bilham, R., V. K. Gaur and P. Molnar. 2001. ‘Himalayan Seismic Hazard’, Science 293: 1442–44. Bilham. R., F. Blume, R. Bendick and V.K. Gaur. 1998. ‘Geodetic Constraints on the Translation and Deformation of India: Implications for Future Great Himalayan Earthquakes’, Current Science 74(3): 213–29. Biswas, S.K. 1987. ‘Regional Tectonic Framework, Structure and Evolution of the Western Marginal Basins of India’, Technophysics 5(135): 302–27. Building Materials and Technology Promotion Council (BMTPC). 2006. Vulnerability Atlas of India. New Delhi: BMTPC. Dasgupta, S., P. Pande, D. Ganguly and Z. Iqbal. 2000. Seismotectonic Atlas of India and its Environs. Calcutta: Geological Survey India. de La Fuente., A., L.C. Felipe and A. Revi. 2008. Assessing the Relationship between Natural Hazards and Poverty: A Conceptual and Methodological Proposal. Geneva. Dharamadhikari, S. 2009. Mountains of Concrete: Dam Building in the Himalayas.

Downloaded by [University of Defence] at 01:15 24 May 2016 Manthan. Dunbar, P.K., R.G. Bilham, and M. J. Laituri. 2002. ‘Earthquake Loss Estimation for India based on Macro-economic Indicators’, in T. Beer and Alik Ismail- Zadeh (eds), Risk Science and Sustainability: Science for Reduction of Risk and Sustainable Development of Society. NATO: Springer. Th e Internatinonal Emergency Events Database (EM-DAT). 2008. For online access see http://www.emdat.be/Database/terms.html Gaur, V. K. 1994. ‘Evaluation of Seismic Hazard in India: towards Minimizing Earthquake Risk’, Current Science 67(5): 324–35. Afterword Ú 403

Gusain, M.S., D. Sarawal, A.K. Garg, B.K. Garg and N.S. Rana. 2007. ‘Development of Tehri Hydro Project — Inception to Commissioning’, Water and Energy International 64(1): Indian Roads Congress (IRC). 2002. SP 20-2002 Rural Roads Manual. New Delhi. International Strategy for Disaster Reduction (ISDR). 2009. Global Assessment of Risk: Reducing Risk Reducing Poverty. New York: ISDR. Jain, S.K. 2003. ‘Review of Indian Seismic Code, IS 1893 (Part 1): 2002’, Th e Indian Concrete Journal, November. ———. 2004. ‘Indian Earthquakes: An Overview’, in Sudhir K. Jain (ed.), Earthquake Engineering — An ICJ Compilation. Th ane: Research & Consultancy Directorate, Th e ACC Ltd. Jain, S.K. and N.C. Nigam. 2000. ‘Historical Developments and Current Status of Earthquake Engineering in India.’ Proceedings of the 12th World Con- ference on Earthquake Engineering, 30 January–4 February, Auckland, New Zealand. Jigyasu, R. 2002. ‘Reducing Disaster Vulnerability through Local Knowledge and Capacity’. PhD dissertation, Norwegian University of Science and Technology, Trodheim. Khattri, K.N. 1987. ‘Great Earthquakes, Seismicity Gaps and Potential for Earthquake Disaster along the Himalaya Plate Boundary’, Tectonophysics 138: 79–92. ———. 1992. ‘Seismic Hazard in Indian Region, in Seismology in India’, Current Science 62: 109–116 ———. 2006. ‘A Perspective in Seismic Hazard Risk in Northern India’, in K.S. Valdiya (ed.), Coping with Natural Hazards. Poona: Orient Longman. Khattri, K.N., A.M. Rogers, D.M. Perkins and S.T. Algermissen. 1984. ‘A Seismic Hazard Map of India and Adjacent Areas’, Tectonophysics 108: 93–134. National Centre for Peoples’-Action in Disaster Preparedness (NCPDP). 2008. Case Studies of Seismic Retroﬁ tting — Latur to Kashmir. Ahmedabad: NCPDP. Planning Commission. 2002. Tenth Five-Year Plan. New Delhi: Government of India.

Downloaded by [University of Defence] at 01:15 24 May 2016 Revi, A. 1994. Maharashtra Earthquake: Reconstruction Programme Needs Restructuring. Hindu Survey of the Environment. Chennai. ———. 2008. ‘Climate Change Risk: An Adaptation and Mitigation Agenda for Indian Cities’, Environment and Urbanisation 20(1): 207–29. Sinha, R. and N. Adarsh. 1999. ‘A Postulated Earthquake Damage Scenario for Mumbai’, ISET Journal of Earthquake Technology 36(2–3): 169–76. TARU. 2005. Gujarat Hazard Risk and Vulnerability Atlas. Gandhinagar: TARU. ———. 2007. National Cyclone Risk Mitigation Programme Report. Gandhinagar: TARU. 404 Ú Aromar Revi

TARU/BMTPC. 1993. Action Plan for the Reconstruction of Earthquake Aff ected Marathwada. New Delhi: TARU/BMTPC. ———. 1997. Building Damage Zonation and Repair and Reconstruction Options for the Jabalpur Earthquake. New Delhi: TARU/BMTPC. ———. 1998. Chamoli Earthquake Repair and Reconstruction Action Plan. New Delhi: TARU/BMTPC. TARU/BMTPC. 2001a. Action Plan for Earthquake Aff ected Garhwal. New Delhi: TARU/BMTPC. ———. 2001b. Rapid Assessment of Damage to Buildings aff ected by the Kachchh Earthquake of 26th January 2001. New Delhi: TARU/BMTPC. Telford, J., J. Cosgrave and R. Houghton. 2006. Joint Evaluation of the International Response to the Indian Ocean Tsunami: Synthesis Report. London: Tsunami Evaluation Coalition. World Bank. 2002. India Gujarat Earthquake Recovery Programme Assessment Report. Washington, DC: World Bank. ———. 2005. India Emergency Tsunami Reconstruction Assessment Report. Washington, DC: World Bank. Downloaded by [University of Defence] at 01:15 24 May 2016 About the Editors

Shirish B Patel is a civil engineer with a wide range of interests. Th e fi rm of consulting engineers he founded in 1960, Shirish Patel & Associates Consultants Private Limited, is one of India’s leading civil and structural engineering consulting fi rms, which has worked extensively in prefabrication (large panels and box-unit construction), the design of factories (especially for better natural lighting, ventilation and economical design), marine works, hotels, tall buildings, railway stations, bridges, and elevated railway tracks. In 1994, they became the fi rst and so far remain the only Indian fi rm to have won an Award from the Fédération Internationale de la Précontrainte (FIP) for excellence in design of what was judged to be one of the fi ve best structures built in the preceding four years, world-wide. Mr Patel has personally worked in applying civil engineering skills for improving agricultural yields (lift irrigation and land shaping), energy conservation (working on development of a fuel-effi cient cooker which saves 75 per cent of the energy conventionally used, given that nearly 40 per cent of India’s daily energy consumption goes into cooking), and most extensively in urban planning and urban aff airs. He was one of the three original authors of the New Bombay project; when it was fi rst set up he was in charge of planning and design for the fi rst fi ve years. He is also one of the Founder Directors of the Indian Institute for Human Settlements, a higher educational institution being set up to address the needs of urban practice.

Aromar Revi is an international researcher, practitioner and consultant with 25 years of interdisciplinary experience in public policy and governance, the political economy of reform, development, technology, Downloaded by [University of Defence] at 01:15 24 May 2016 human settlements, and sustainability. He is the Director of the Indian Institute of Human Settlements (IIHS), which will create India’s ﬁ rst independent national university to address the challenges of urbanisation through an integrated programme of education, research and knowledge generation, and consulting and advisory services. He has been a senior advisor to various ministries of the Government of India, including the national Planning Commission; consulted with a wide range of multilateral and bilateral development institutions and national and 406 Ú Recovering from Earthquakes

transnational fi rms, having led over a hundred major research and consulting assignments in India and abroad. Mr Revi is one of South Asia’s leading disaster mitigation and management experts and has led emergency teams to assess, plan and execute recovery and rehabilitation programmes for 10 major earthquake, cyclone, surge and fl ood events aff ecting over fi ve million people. He is also one of India’s leading experts on Climate Change adaptation and a lead advisor to the UN ISDR Global Assessment of Risk Report (2009). Downloaded by [University of Defence] at 01:15 24 May 2016 Notes on Contributors

B.R. Balachandran is an architect and planner with 18 years of professional and academic experience. He is best known in professional planning circles for his contribution (as Executive Director, Environ- mental Planning Collaborative [EPC], Ahmedabad) to planning the reconstruction of the city of Bhuj in Kutch District of Gujarat after the earthquake of 2001. After spending nearly a decade with EPC, he shifted base to Bangalore and established Alchemy Urban Systems Private Limited, with like-minded professionals. Engaged in a wide range of activities, Alchemy’s work ranges from planning for decentralised waste water management, to post-ﬂ ood reconstruction in Bihar and evolving a regional development strategy for conservation areas around the Ellora caves. Mr Balachandran has been involved in a number of international projects, in Sri Lanka, Nepal, Bangladesh, Mauritius, and Indonesia, and is currently the land use and physical planning expert in an international team commissioned by the World Bank to prepare a Post Disaster Housing Reconstruction Handbook.

Rajendra Desai is Managing Trustee and Honourary Joint Director, National Centre for Peoles’-Action in Disaster Preparedness (NCPDP). He obtained a Master’s degree in Structural Engineering from Rutgers University in 1972, following which he gained 13 years of diversifi ed civil engineering experience in the United States. A damage assessment assignment after the 1993 Latur earthquake made disaster mitigation his principal area of work. He spent the next six years in Latur, focusing on building peoples’ awareness about disaster safety and training artisans in earthquake resistant construction, particularly by retrofi t- ting existing buildings. For this work, he and his wife, Rupal Desai, won Downloaded by [University of Defence] at 01:15 24 May 2016 the fi rst A.S. Arya Award from IIT Roorkee. He has also been involved in similar work in Uttarakhand, Jabalpur (MP), Kachchh (Gujarat), and Kashmir. He and Rupal Desai have co-authored several publications, including a Manual for Restoration and Retrofi tting of Rural Struc- tures in Kashmir (UNESCO) and a Manual for Hazard Resistant Con- struction in India (UNDP).

Weimin Dong has over 30 years of industrial, teaching and research experience, and has specialised in seismic hazard evaluation and risk 408 Ú Recovering from Earthquakes

assessment. In 1986 he received a doctorate from Stanford University in Structural and Earthquake Engineering. His research interests include seismic design of buildings, structural responsibility, and the application of neural networks, pattern recognition and artifi cial intelligence techniques for earthquake hazard mitigation and risk assessment. He has worked extensively with the insurance and investment industries, with structural engineering fi rms, as well as with the State of California and local and city government bodies in the transfer of earthquake technology. He has also participated in many joint venture projects (US– China, US–Japan) to share knowledge in hazard mitigation with other countries with similar concerns. His publications include Building a More Profi table Portfolio-Modern Portfolio Th eory with Application to Catastrophe Insurance (2001) and Earthquake Engineering (co-authored, 1996).

Prema Gopalan is the Founder and Chairperson of Swayam Shikshan Prayog (SSP), a learning and development organisation based in Mumbai. Set up 1994, SSP is engaged in building networks of rural ‘social businesses’ that are co-created by private corporations and women survivors of disasters such as the 2004 Asian Tsunami and the Latur and Gujarat earthquakes of 1993 and 2001, respectively. In the Maharashtra Emergency Earthquake Rehabilitation Programme, SSP, as community-participation consultant to the Government of Maharashtra, implemented this model for the community-driven reconstruction programme across 2,000,000 households in 1,300 villages. Ms Gopalan is a steering committee member of GROOTS International and Huairou Commission—a network of autonomous grassroots women’s organisations across 40 countries. She facilitates Disaster Watch—a Global Working Group of the Huairou Commission. She received the Mary Fran Myers Gender and Disaster Award in 2007 and the Changemakers Award in 2008.

Downloaded by [University of Defence] at 01:15 24 May 2016 Debarati Guha-Sapir is Director of the WHO Collaborating Centre for Research on the Epidemiology of Disasters (CRED) and Professor at University of Louvain School of Public Health, in Brussels, Belgium. She holds an Adjunct Professorship at Tulane University Medical Centre, New Orleans, for Health and Humanitarian Aid. Trained in Calcutta University, Johns Hopkins University and University of Louvain she holds a Doctorate in epidemiology. Since 1984, she has been involved in field research and training in emergency and humanitarian aid issues, working closely with WHO, UNHCR, UNDP, and the European Notes on Contributors Ú 409

Commission in various disaster-prone regions of Asia, Africa and South America. She is particularly interested in health systems research, epidemiology in unstable situations and international policy related to relief and post conﬂ ict transition. Founder of the now internationationa EMDAT, that she is currently setting up a similar global system for data from civil conﬂ icts.

Philippe Hoyois is a sociologist trained in the University of Bruxelles and Louvain in Belgium. For over 30 years he has been involved in medical and epidemiological studies in both academic and non-academic contexts, developing an extensive experience in mental health issues. During the 1990s, he supported and evaluated the setting up of psychiatric emergency and crisis units in the academic hospitals of the Université Libre de Bruxelles and the Université Catholique de Louvain and worked on the development of a network of ambulatory crisis teams. He has specialised in analyses of social statistics and in particular statistics related to health. Dr Hoyois has a long experience in global analyses of disasters using the EMDAT database and his knowledge in data analysis led to an involvement in CRED’s researches as Senior Research Fellow while continuing speciﬁ c works on mental health services organisation. He has published in several English and French journals in this area.

Rohit Jigyasu is a conservation architect and risk management consultant, currently working as invited professor at Ritsumeikan University in Kyoto, Japan. He was awarded a Doctorate in Engineering by the Norwegian University of Science and Technology, Trondheim for his thesis titled ‘Reducing Disaster Vulnerability through Local Knowledge and Capacity — the Case of Earthquake Prone Rural Communities in India and Nepal’. Dr Jigyasu has taught in a number of academic institutions in India and abroad, has several publications to his credit and has been consultant to several national and international organisations

Downloaded by [University of Defence] at 01:15 24 May 2016 like Archaeological Survey of India, National Institute of Disaster Management, UNESCO, ICCROM and the Getty Conservation Institute. He also brings with him practical experience of working at the world heritage sites of Khajuraho, Hampi, Konarak, and Ajanta and Ellora.

M.N. Joglekar is Professor and Director of the Vastukala Academy, College of Architecture, New Delhi. After graduating in Architecture from M.S. University, Vadodara, Gujarat, he completed a Master’s degree in Urban and Regional Planning from IIT, Kharagpur, followed by a diploma 410 Ú Recovering from Earthquakes

with distinction in Housing, Planning and Building from Baucetrum, Rotterdam. He spent the formative years of his professional career, 1964–70, in London, working for the Urban Design division of a Local Authority. Soon after returning to India he joined Housing and Urban Development Corporation (HUDCO), a Government of India statutory institution dealing with housing ﬁ nance and design. From 1972 to 1998 he was in charge of the design development wing, specialising in low- income housing, townships as well as disaster rehabilitation projects. His work, in fact, helped HUDCO win several awards for its projects. In addition to a number of articles published in professional journals, the National Institute of Rural Development (Hyderabad) has published his book Blending the Traditional with Modernity, and along with architect S.K. DAS he has edited a book titled Contemporary Indian Architecture, published by HUDCO.

Rajeev Kapoor is an IAS offi cer from the Utter Pradesh cadre, and was District Magistrate of Uttarkashi during the post-earthquake relief and rehabilitation phase. He has served in many capacities with State and union governments, and also with a bilateral aid agencies. His acquaintance with disaster management is in the capacity of a fi eld offi cer, having handled emergencies resulting from fl oods, droughts and earthquakes in districts under his jurisdiction. His areas of interest are governance, capacity building and infrastructure fi nancing.

Sudhir Kumar is a Project Oﬃ cer working with the Asian Disaster Preparedness Center, Bangkok, and is currently based in Myanmar, where he has been involved in drafting the National Action Plan on Disaster Risk Reduction (DRR) as well as studying the DRR institutional set-up. He worked with UNDP-India for three years and more on the Disaster Risk Management Project and was a Senior Executive with the Gujarat State Disaster Management Authority (GSDMA) in the post-earthquake

Downloaded by [University of Defence] at 01:15 24 May 2016 R&R programme. He also developed Guidelines for the Preparation of Provincial Disaster Management Plan in Afghanistan. He has been involved in designing and developing academic courses on disaster management and his papers on DRR have been published in national journals. Mr Kumar holds a Master’s degree in Rural Management from Institute of Rural Management, Anand (IRMA), India. His areas of interest include policy and institutional set-ups for disaster management, community-based disaster preparedness and reconstruction and rehabilitation. Notes on Contributors Ú 411

R. Srinivasa Murthy is former Professor of Psychiatry, National Institute of Mental Health and Neurosciences, Bangalore. He has an MD in Psychiatry from the Postgraduate Institute of Medical Education and Research, Chandigarh, prior to which he was trained in medicine and surgery (MBBS) at the Christian Medical College, Vellore. In 2000–01, he was Editor-in-Chief of the World Health Report 2001, which focused on mental health. Following retirement in 2004, he worked with the WHO at the Eastern Mediterranean Regional Oﬃ ces of Cairo and Amman, and in 2006–07 he was the mental health oﬃ cer of WHO-Iraq. Dr Murthy is currently working in Bangalore with the Association for Mentally Challenged, a voluntary agency working to develop a community-based care programme for such persons and their families. He has also worked on developing psychosocial interventions for survivors of the Bhopal gas tragedy. His research contributions have been mainly in areas related to community mental health, and he has published extensively in over 200 national and international journals.

Alex Salazar is a practicing architect and principal of Salazar Archi- tects Studio, a design fi rm based in San Francisco that specialises in multi-family aff ordable housing, urban design, community planning and sustainable building practices. Mr Salazar has been recognised by the American Institute of Architects (AIA) for his contribution to community planning, winning both AIA San Francisco’s 2009 Young Architect Award and AIA California’s 2008 Community Housing Assistance Honor Award. His fi eld research on post-disaster housing in India, funded by the Graham Foundation in 1993, was carried out over a year spent living and working with several NGOs, including ASAG and SPARC in Latur, Maharashtra. Writing about the experience earned him a Master’s degree in Architecture in 1998 from the University of California, Berkeley. He now occasionally teaches architecture design studios at University of San Francisco and University of Oregon. His article, ‘Designing a Socially Just

Downloaded by [University of Defence] at 01:15 24 May 2016 Downtown’, was published in Shelterforce, 145, Spring, 2006.

Dr Satendra is a member of the Indian Forest Service (IFS), batch of 1986, Bihar cadre, and at present is Special Secretary, Department of Disaster Management, Government of Bihar, India. He holds a Doctorate in Himalayan Geo-Tectonics from Delhi University, and an MBA from Hull University, UK. During his service tenure of more than 23 years he has been associated with the Union Ministeries of Environment and Forest, Rural Development and Home. Dr Satendra has been deeply 412 Ú Recovering from Earthquakes

involved in the ﬁ eld of disaster risk management and climate change adaptation, and has had wide exposure working in these fields at national and multinational levels with government and UN agencies. He was International Lead Technical Consultant/Team Leader of the FAO–UN Bangladesh Project on Climate Change Adaptation and Disaster Risk Reduction. He has also published many research papers and articles, and is the author of three books.

Haresh C. Shah is Obayashi Professor of Engineering, Emeritus, Stanford University, California. A pioneer in the fi elds of risk analysis, earthquake engineering and probabilistic methods, he was Chairman of the Department of Civil Engineering and Founding Director of the John A. Blume Earthquake Engineering Center at Stanford. He is also Founder and Senior Advisor of Risk Management Solutions, Inc., a leading soft- ware company that specialises in assisting insurance and reinsurance companies as well as banking and other fi nancial institutions in managing fi nancial risks associated with catastrophic events. In addition to these he is Founding Director and Chairman, World Seismic Safety Initiative, and Member of the Board of Trustees and Chairman of the Academic Advisory Committee at the Nanyang Technological University (NTU) of Singapore. In April 2006, on the hundredth anniversary of the Great 1906 San Francisco earthquake, Dr Shah was given a unique award as the ‘Top Seismic Engineer of the 20th Century’ by the Applied Technology Council/Engineering New Record, USA.

Brig KJ Singh has over 37 years of service in the Army Medical Corps, at various regimental, staff, command and instructional appoint- ments, in India and abroad. Presently, he is serving as Head of Medical Services of an army formation deployed in north-west India. He was Professor of Hospital Administration at the Armed Forces Medical, Pune and post-graduate examiner for MD Hospital Administration examin-

Downloaded by [University of Defence] at 01:15 24 May 2016 ations conducted by the University of Pune and Manipal Academy of Higher Education. His areas of interest include medical management of mass casualties, combat medical support, and planning and designing healthcare facilities.

Anil K. Sinha is Co-chairman, Global Forum for Disaster Reduction, Mumbai. He is a retired Indian Administrative Service (IAS) Oﬃ cer and was formerly Executive Director, National Center for Disaster Man- agement, Government of India. With an academic background that Notes on Contributors Ú 413

covers physics, the social sciences and public administration, he has spent 30 years in these ﬁ elds, being involved in management, policy analysis and programming, and national-level capacity-building eﬀ orts for environmental and climate change issues including disaster risk reduction. He has worked on institutional and legislative systems for disaster reduction and recovery, post-disaster response and recovery planning, training and other development activities, disaster risk management and mitigation with enhancement of sustainable development policies, inter-agency co-ordination, and early warning systems. He has written extensively on these subjects, and was involved in disaster management work after the Orissa Super Cyclone, and the Bhuj and Chamoli earthquakes, among others.

Shikha Srivastava is a development professional based in Orissa. She has lived and worked in remote areas of Orissa, Andhra Pradesh and Rajasthan. She is currently working with UNICEF-Orissa as a Consultant in the Child Environment Programme. Her areas of work are disaster management, water and sanitation and sustainable development, in which she has been associated with governmental, non-governmental and international organisations for more than a decade. A graduate from Lady Shriram College for Women, Delhi University and a postgraduate from Jawaharlal Nehru University, New Delhi, she also holds an additional graduate degree in law and a diploma in journalism. She has written several papers on disaster management and has edited a book on corporate participation in disaster management for the construction federation of India.

V. Thiruppugazh is a senior officer of the IAS, batch of 2001, of the rank of Secretary to Government of Gujarat. After obtaining a Master’s in physics and Bachelor’s in teaching, he worked as a teacher in J. Krishnamurti’s school and various other educational institutions. In Gujarat he served in various district assignments (Assistant Collector

Downloaded by [University of Defence] at 01:15 24 May 2016 Morvi, District Development Offi cer- Kheda, Bhavnagar and Sabarkanta, and Collector Sabarkanata), before he was posted as the Joint Chief Executive Offi cer of the Gujarat State Disaster Management Authority, which was set up as the nodal agency for reconstruction following the Gujarat earthquake in 2001. In addition to his contribution in Gujarat to post-earthquake reconstruction, he also has vast hands-on experience of handing various disasters such as fl oods, cyclones, and droughts. He is currently doing his PhD on Natural Disasters, at the Australian National University. 414 Ú Recovering from Earthquakes

Krishna S. Vatsa is an IAS offi cer, batch of 1987, and has worked in the area of post-disaster reconstruction and disaster risk management since 1995. He implemented the Maharashtra Emergency Earthquake Rehabilitation Programme from 1995 to 1999. He has also served as Secretary to the Government of Maharashtra, Relief and Rehabilitation from 2003 to 2006, and then as Secretary, Rural Development and Panchayati Raj Department from 2006 to 2007. He implemented a post- disaster reconstruction programme in tsunami-aff ected Pondicherry, followed by another reconstruction programme in the wake of the fl oods in Maharashtra in 2005 and 2006. In 2007, Dr Vatsa joined UNDP- Philippines as the Early Recovery Co-ordinator, where he worked to set up a recovery programme following typhoon Reming. In 2008, he moved to New Delhi as the Regional Disaster Reduction Advisor with the Bureau for Crisis Prevention and Recovery (BCPR), UNDP, where he is working now.

Sandeep Virmani is qualiﬁ ed as an architect, but work with a German social services organisation soon took him into rural housing and best practices in Gujarat. He moved to Kachchh and worked on water harvesting, organic agriculture, renewable energy, nomads and their habitats, the built environment, biodiversity, and industrialisation and developed models of how we can have a more sustainable, equitable and gender sensitive society. To achieve this he learnt how to set up and manage institutions. Over the years he has helped establish several organisations: Sahjeevan, an organisation that helps incubate a topic as a knowledge centre for a period of time so that it may then be registered as an independent organisation; Arid Communities and Technologies (ACT) was the ﬁ rst of these, focusing on water in all its dimensions; Kutch Sajiv Kheti Manch (KSM) networks the initiatives of the organic farmers of Kachchh; Hunnarshala Foundation, after the earthquake in 2001 in Gujarat, which works on construction technologies; and Kutch

Downloaded by [University of Defence] at 01:15 24 May 2016 Navnirman Abhiyan (KNNA), a network of 27 NGOs working in Kachchh, where the focus is on policy and the capacity building of the district. Index

aakarni register 146 Bhuj Development Council 189, 190, Abhiyan 146, 151, 152, 155, 174, 399 193 Action Aid 322 Bochasanwasi Akshar Purushottam Aggregate Loss Exceeding Probability Swaminarayan (BAPS) 149, 150 (AEP) 363, 366, 367, 372 Bombay Public Trust Act 225 Ahmedabad Study Action Group building codes, revision of 228 (ASAG) 62 buildings, classiﬁ cation of damage to American Red Cross 322 218 Asian Development Bank (ADB) 50, Bureau of Indian Standards (BIS) 103, 163, 313 228, 247, 392 asset building and income generation 63–67 calamity insurance 12 casualty insurance 345 basti 216, 217, 218, 219, 221 catastrophe bonds 373; numerical bathrooms and toilets, provision of risk simulation for 378; pricing of: 58–59 bond portfolio management 381; ‘best practice’ villages 131 present value 379–81; structure BHADA see Bhuj Area Development of: indemnity-based bonds 377; Authority (BHADA) modelled loss bonds 376–77; Bhuj, see also Walled City of Bhuj: parametric bonds 374–76 post-disaster reconstruction 159; catastrophe impact models 345 institutional framework for catastrophe insurance: basic issues of 162–64; legal framework for 342–45; engineering modelling 164–66; proposed land-use plan and loss estimation technologies for 176; proposed road network for 345–46; attenuation module for 175; reconstruction planning 346–49; hazard module 346; 159–60; rising from rubble vulnerability module 349–50; 200–202; stages in reconstruction event loss table (ELT) and process: city-level development exceedance probability (EP)

Downloaded by [University of Defence] at 01:15 24 May 2016 plan, preparation and sanction curve 362; fi nancial module of 166–86; reconstruction strategy 350–51; insurance portfolio 160–66; urban reconstruction composition 351–54; portfolio risk package 162; water bodies in 177 quantifi cation 354–56; recognising Bhuj Area Development Authority uncertainty in loss 358–60; tool for (BHADA) 164, 180, 182, 186; calculating fi nancial loss 356–58; staffi ng and infrastructure of 181; insured losses when considering town panning schemes 189 loss uncertainty 360–61 Bhuj Commercial Co-operative Bank catastrophe models: development of 156 337; and distribution of potential 416 Ú Recovering from Earthquakes

losses 369; modelled loss bonds decision tree 74 and 376–77; reinsurance and dev bhumi 217 solvency 369–71 ‘Development Plans’ for rehabilitation catastrophe risk management 9, 339, of towns 143; planning of 343 relocation sites 181–83; proposals Centre for Research on the for heritage conservation 178; Epidemiology of Disasters (CRED) publication and sanction of 14 179–80; stages of 165 Chamoli earthquake 5, 204, 207, 215, disaster: criteria for 14; disease and 385, 391, 400 injuries after 28–30; immediate chemical, biological, radiological, and impact of 25–28; long-term impact nuclear (CBRN) casualties 311 of 30–31; plan 293; secondary Chetna (NGO) 328 eff ects of 31 Chief Minister’s Relief Fund 54 disaster cycle, phases of 18 city-level development plan, in Bhuj ‘disaster housing’ 88 166–67; development approaches Disaster Management Act 227–28, for 174–79; mapping, studies 235, 239, 242, 244, 245–47, 310 and public consultations 167–74; disaster management cycle 292–93 planning of relocation sites 181–83; disaster management, in India publication and sanction of 238–39, 291–92; acts and policies 179–80; staffi ng and infrastructure 245–47; and collaboration with of BHADA 181; timeline for 167; international community 252–53; Walled City see Walled City of Bhuj emotional stress disorder 300; civic amenities, provision of 60–61 environmental health services for commercial portfolio 352–53 297–98; community involvement community-based organisations and education 298; sanitation and (CBOs) 79, 122, 128, 223, 244 hygiene 299; settlement sites for community-led disaster risk reduction displaced persons 299; shelter 396 299; vector control 299; water Community Participation Consultants supply 298–99; future course of (CPC) 95, 104, 134 241; healthcare preparedness confi nement reinforcement, of plan of 293–94; disposal of dead structures 274 bodies 297; ethics and teamwork Cooperative for Assistance and Relief 297; hospitals 295–96; operation

Downloaded by [University of Defence] at 01:15 24 May 2016 Everywhere (CARE) 322 theatre (OT) 296; organisation corporate social responsibility (CSR) 12 294–95; rehearsal and revision cottage industries, for livelihood of plans 297; search-and-rescue restoration and sustainability units 295; volunteer services 297; 230–31 human resource development cyclones 14, 22; types of injuries and capacity building 250–52; caused by 29 manual 293; medical store supply 300–303; organisation of 310–11; debris removal programme 151–53; international agencies 313; quality cases of corruption 153 improvement 314; role of armed Index Ú 417

forces 313–14; salient gaps and District Rural Development Agency recommendations 311–13; plan for (DRDA) 154 293; policies 3; recommendations droughts, 14, 19; number of people of HPC for 239–41; district level aﬀ ected by 20, 27 244–45; local level 245; national level 242–44; state level 244; earthquake mitigation project 248 standard operating procedure Earthquake Rehabilitation Policy 50 (SOP) for 293 earthquake-resistant construction 8, Disaster Management Policy 236, 242; 104, 125, 127, 130, 132, 133, 134, objectives of 227 217 disaster manual 293 Earthquake Resistant (EQR) features disaster mitigation and management 132 in India 247–49; community earthquake-resistant technology, involvement in handling 6; application of 61, 73, 130 government’s view on 5–7; earthquakes: Chamoli (29 March policy implications for 33–35 1999) 5; damage and losses caused disaster preparedness, and post- in Gujarat 224; global impact of earthquake rehabilitation 138–39 385–86; impact and risk sharing disaster-related disease, outbreak 387–88; impact in South Asia patterns of 30 386–87; impact on non-engineered disaster-related mental health care, structures 255–58; insurance Indian experiences in 321–22 premiums and losses 344; Kachchh disaster relief 3, 33–35, 300, 301, 314, (26 January 2001) 4–5, 31, 120; 329 Latur 119–20; lessons from disaster response preparedness 12 Gujarat and four c-words 31–33; Disaster Risk Reduction (DRR) 385; lethality of 22; Marathwada (30 community-led 396; institutional September 1993) 4, 79, 323–25; development for 399–400 medical management during disasters in India: Chamoli (29 March see medical management, of 1999) 5; comparison of human earthquake disasters; mortality impacts of 27; Great Bengal famine and structural destruction 27; of 1941 19; human impact of 22; non-engineered construction, Kachchh (26 January 2001) 4–5; impact on 8; number of people Marathwada (30 September 1993) aﬀ ected by 20; predictability of 22;

Downloaded by [University of Defence] at 01:15 24 May 2016 4, 79; Uttarkashi (20 October preparedness 389–90; psychosocial 1991) 3–4, 39–46 interventions in diff erent phases disease profi les, of natural disasters of: fi rst few months 320–21; 29–30 initial phases of disaster 319–20; District Disaster Management pre-disaster phase 318–19; Authorities (DDMAs) 244–45 reconstruction in Maharashtra District Industries Corporation (DIC) see reconstruction programme, 155, 156 in Maharastra; reconstruction District Inspector of Land Records programme 47; schemes for (DILR) 167, 180 rehabilitation in Maharastra 418 Ú Recovering from Earthquakes

50–52; shelters 43–44; trauma Geographical Information system after see trauma, after earthquakes; (GIS) 6, 240 types of injuries caused by 29; Global Seismic Hazard Assessment Uttarkashi (20 October 1991) 3–4, Program (GSHAP) 389 39–46, 322–23 Gram Panchayat 50, 104, 129, 146, 149 economic risk mitigation 398 Gram Sabha 104, 122, 229 electricity, provision of 59–60 grant-cum-loan schemes 154 ELT see event loss table (ELT) Grassroots Organisations Operating EM-DAT see Emergency Events Together in Sisterhood (GROOTS) Database (EM-DAT) 121 Emergency Centres 6 grass-roots women’s organisations Emergency Events Database 120, 121, 125–28, 139 (EM-DAT) 14, 22, 25, 27 Great Aceh earthquake 384 Emergency Operations Centres (EOCs) Great Bengal famine (1941) 19 240 Gujarat Development Control emergency phase, of disaster cycle 18 Regulations (GDCR) 228 emergency response systems 390–91 Gujarat earthquake, management engineering model, for catastrophe of: casualty data 306–7; casualty impacts: attenuation module management protocol 304–5; 346–49; hazard module 346; central response 305–6; diﬃ culties vulnerability module 349–50 encountered during 307–9; environmental health services, for disaster impact 303; experience disaster management 297–98; of working and 326–29; local community involvement and response at Bhuj military hospital education 298; environmental 303–4 sanitation 299; food sanitation Gujarat Earthquake Reconstruction 299; personal hygiene 299; and Rehabilitation Policy 228, 235 settlement sites for displaced Gujarat Earthquake Reconstruction persons 299; shelter 299; vector and Rehabilitation Programme control 299; water supply, 298–99 (GEERP) 235, 236 Environment Impact Assessments Gujarat Ecology Commission (GEC) (EIA) 152 152 event loss table (ELT) 362, 363, 364 Gujarat Institute for Disaster exceedance probability (EP) curve 362, 370 Management (GIDM) 6; objectives

Downloaded by [University of Defence] at 01:15 24 May 2016 of 233 falia samitis 193 Gujarat State Disaster Management falia sanskriti 193 Act 235 famines 19, 27; types of diseases Gujarat State Disaster Management caused by 30 Authority (GSDMA) 143, 223, fl oods 14, 19–22; environmental 235; awareness campaign 232–33; destruction, impact of 25; types of certifi cation of masons 235; diseases caused by 29–30 disaster management policy fl oor space index 162 227; engineers’ training 235; fortress houses (garhis) 101 feasibility studies/plans on Index Ú 419

disaster management themes removal 151–53; temporary 234; institutional set-up of houses and permanent slums 225; masons’ training 234–35; 156–57; post-Bhuj disaster 142; workshops/seminars 234 ‘public–private’ partnership 143 Gujarat State Disaster Management housing units, constructed by MHADA Policy 235–36 95 Gujarat Town Planning and Urban HPC see High Power Committee (HPC) Development Act 163, 193 on Disaster Management Gujarat Urban Development Company human impact, of disasters in (GUDC) 163, 164, 186 India 22 Hyogo Framework for Action 246 hands-on masons training, for post- earthquake rehabilitation 135 impact phase, of disaster cycle 18 Hanshin earthquake 166 India Disaster Report 330 hazard risk zonation 389 India Disaster Resource Network 252 healthcare centres for disaster Indian Journal of Social Work 330 situation, organisation of 294–95 Indian Ocean tsunami 330, 384, 390, healthcare preparedness, plan for see also tidal waves disaster management 293–94; Indian Red Cross 322 disposal of dead bodies 297; ethics ‘Initiative for Planned and and teamwork 297; hospitals Participatory Reconstruction’ 161 295–96; operation theatre (OT) injury proﬁ les, of natural disasters 29 296; organisation 294–95; Institute of Seismological Research rehearsal and revision of plans (ISR), Gandhinagar 234 297; search-and-rescue units 295; insurance cover, for natural disaster volunteer services 297 losses 9 heritage conservation, proposals for insurance portfolio: composition of: 178 commercial portfolio 352–53; Heritage Society 179 reinsurance structures 353–54; High Power Committee (HPC) on residential portfolio 351–52; Disaster Management 239–41 depiction of 351; insurance risk hospitals, disaster preparedness of metrics and 362–66 312–13 insurance risk metrics 362–66 Housing and Urban Development Inter-Agency Standing Committee

Downloaded by [University of Defence] at 01:15 24 May 2016 Corporation (HUDCO) 210, 213, (IASC) Guidelines 316 216 International Decade of Natural housing project, for gas victims at Disaster Reduction (IDNDR) 239, Bhopal 214 252, 291, 292 housing reconstruction 228–30 international NGOs (INGOs) 150 housing rehabilitation: failures: in International Strategy for Disaster damage assessment 143–46; Reduction (ISDR) 246, 292 economic recovery 153–55; performance of NGOs 146–51; Jabalpur earthquake 258, 385, 397 planning 155–56; in rubble joint-family households 4 420 Ú Recovering from Earthquakes

Kachchh earthquake 4–5, 31, 120, Maharashtra Housing and Area 128, 207, 210, 385, 389, 393, 394, Development Authority (MHADA) 396, 399 95 Kashmir earthquake 302, 390, 391 Mahila Mahiti Kendra (MMK) 135 Killari earthquake 56, 207, 209, 210, Mahila Mandals 119, 136; 213 construction supervision Kobe earthquake 9, 22, 185, 339, 386, workshops for 130; for co- 395 ordination between oﬃ cials 129; Kutch Mitra Daily 153, 156 earthquake rebuilding programme 131; for encouraging village land readjustment schemes 186 participation in reconstruction landslides, impact on environmental programme 129; social and cultural destruction 25 activities 128 land-use planning and development man-made disasters vs. natural 174 disasters 22–25 Latur earthquake 34, 47, 119–20, 127, Marathwada earthquake 4, 79; 138, 139, 142, 385 epicentre of 82; experience lifeline infrastructure, strengthening of working 323–25; in situ of 393–95 reconstruction and participatory Lions Club, township developed by 149 development: assessment of livelihood restoration and policy impacts 93–105; politics of sustainability, initiatives for: participation 89–91; World Bank agriculture 232; cottage industries policy in early 1990s 87–89 230–31; industries 232; mason-training programmes 125, 132; self-employment activities 231–32; for seismic-safe construction 154 women’s livelihood 231 mass casualty event (MCE) 311, 312 load-bearing building systems 273 mass casualty management 304, 311, London Interbank Oﬀ ered Rate 312 (LIBOR) 380 Mata Amritanand Mayi (religious sect) loss estimation technologies, for 149 insurance purposes 345–46 medical interventions, during post- disaster situation 7–8 Maharashtra Emergency Earthquake medical management, of earthquake Rehabilitation Programme disasters 291; emotional stress

Downloaded by [University of Defence] at 01:15 24 May 2016 (MEERP) 48, 58, 324; activities disorder 300; medical store supply for organising women 75; beneﬁ ts 300–303 of implementation 66, 72, 73; medical supply management 300, 301–2 mitigation of future disasters 106; MEERP see Maharashtra Emergency negative impacts of 81; political Earthquake Rehabilitation inﬂ uences on housing policies 81; Programme (MEERP) scheme of housing entitlements melavas 129, 137 under 53; training component of mental health care: Indian experiences 107; training of unskilled labour in disaster-related 321–22; sponsored by 134 psychiatric disorders 324 Index Ú 421

MHADA see Maharashtra Housing per 100,000 inhabitants in India and Area Development Authority 17; mental health consequences (MHADA) of 7; mortality generated by ‘micro-zonation’ 173, 399 25; occurrence in India 23–24; Millennium Development Goals 246 proportion of deaths in India 16 model buildings: diff erent types of non-engineered construction of plans and material options 133; houses, impact of earthquake on 8 for post-earthquake housing non-engineered structures: concept of rehabilitation 132–33 retrofi tting and 266–76; hurdles Modifi ed Mercalli Intensity (MMI) 349 in applying retrofi tting measures Monte Carlo simulation, for to 271–72; impact of earthquakes catastrophe bonds 378 on 255–58; performance in recent multi-hazard disaster management disasters 262–65; relationship to plan 293 local context 259; signifi cance of construction 259–61; vulnerability National Act for Calamity of 261–62 Management 239 non-governmental organisations National Calamity Management Act (NGOs) 79, 104, 121, 122, 205 246 Northridge earthquake 9, 339, 345 National Centre for Disaster Management see National Institute Occurrence Loss Exceeding Probability of Disaster Management (NIDM) (OEP) 363, 365, 366, 367 National Council for Cement and Offi ce of US Foreign Disaster Building Materials (NCCBM) 229 Assistance (OFDA) 14 National Disaster Management Act on-site mass casualty management 242, 244, 246 311 National Disaster Mitigation Resource ‘owner-driven’ policy, for housing Centres (NDMRCs) 252 rehabilitation 147 National Disaster Response Force Oxfam (voluntary organisation) 322 (NDRF) 243, 252 National Institute of Disaster Panchayati Raj 122, 245, 390 Management (NIDM) 7, 243, 250 participative rehabilitation 204–6; natural disasters: average number attitudes and time required to of people aff ected per 100,000 spread participation in 209–10;

Downloaded by [University of Defence] at 01:15 24 May 2016 in India 17; characteristics and categories of participants in human impact of 18–22; disease 208; macro models of 206–8; profi les of 29–30; distribution by project managers as managers of category 15; distribution in India 215; slum upgradation project, 15; distribution of people aff ected implementation model of 219–21 by 16; injury profi les of 29; Participatory Rural Appraisal (PRA) 88 insurance cover for losses suff ered People’s Science Institute (PSI), during 9; long-term impact of Dehradun 45 30–31; vs. manmade disasters pilot demonstration project, for post- 22–25; mean number of deaths earthquake rehabilitation 133 422 Ú Recovering from Earthquakes

PML see probable maximum loss Rastriya Swayamsevak Sangh (RSS) (PML) 149, 150 portfolio risk management 366–67; Rebuilding Housing in Emergency approaches for: reinsurance and Recovery Projects 88 solvency 369–71; understanding reconstruction phase, of disaster cycle risk 367–69; underwriting and risk 18 selection 371–73; levels of 367 reconstruction programme: after post-disaster intervention: Kachchh earthquake 4–5; after empowerment of grassroots Marathwada earthquake 4; asset- women’s groups 125–28; women building and income generation and rehabilitation 122–25; women 63–67; and empowerment of and relief 121–22 women 4; equity and gender post-disaster reconstruction 47, 48, 67–71; involvement of non- 74, 76, 159, 202, 237 governmental organisations post-disaster rehabilitation in 5; in Maharashtra 47; data programmes 275 description 49–50; emerging post-disaster relocation projects, policy issues 74–76; household list of 80 characteristics 55–56; individual post-earthquake rehabilitation: entitlements 52–54; main disaster preparedness and, 138–39; ﬁ ndings and observations model buildings for 132–33; 56–71; programme assistance pilot demonstration project 133; and household characteristics ‘public–private’ partnership in 143; 50; schemes for earthquake role of SSP in 137–38; training rehabilitation 50–52; and post- components: for artisans 134–35; rehabilitation status of women 70 in practice 136–37; in theory 136; Reconstruction, Repair and Seismic for women masons 135 Retroﬁ tting Programme (RRSP) post-traumatic stress disorder (PTSD) 51, 101; for demonstration houses 324, 326–28 133; faulty policies and technical probable maximum loss (PML) 351, guidance 103; limited success 362, 363 of participation 103–4; missed Project Management Unit (PMU) 93, opportunities for 105; ongoing 104, 125, 133 hazardous living conditions and Property Claims Services (PCS), for 104

Downloaded by [University of Defence] at 01:15 24 May 2016 earthquakes and hurricanes 343 rehabilitation phase, of disaster cycle psychiatric disorders, risk factors for 18 developing 324 Reinforced Cement Concrete (RCC) ‘Psychosocial Consequences of 219, 220 Disasters’ conference 329 relief and rehabilitation work 40, Public–Private Partnership 42, 138; post-disaster 276; at Programme (PPPP) 143, 228, 236 Uttarkashi 3 relocated type ‘A’ villages 94–96, Random Rubble Masonry (RRM) 216; 106; abandonment problems delamination of 263 98–99; construction problems and Index Ú 423

increased hazardous conditions Sakhi Federations 137 98; environmental problems and Samvad Sahayaks 129, 130; ‘best reduced labour output 96; house practice’ villages and 131; design problems and reduced earthquake rebuilding programme livelihood opportunities 97; 131 infrastructure problems and Sarva Shiksha Abhiyan (SSA) wasted public resources 97–98; programme 390 socio-spatial problems and caste Savings and Credit Groups (SCGs) 126 dependency 96 seismic belt, installation of 282 relocated type ‘B’ villages 99–101, 106 seismic risk mitigation, strategy for relocation villages: diff erence between 389 new and old houses 56–57; seismic risk reduction, priorities for area of old and new houses 58; 388–89 building materials used in old and seismic-safe construction, mason- new houses 57; condition of old training programmes for 154 houses 58; improvements upon seismic strengthening of structures: old house types 57–58; provision indicative specifi cations for: cast of bathrooms and toilets 58–59; in-situ bond element 280–81; provision of civic amenities 60–61; ferrocement (FC) or seismic belts provision of electricity and water 281; fl oor level diaphragm, for supply 59–60; safety and quality stone on joist fl oor 287; gable of construction 61; over in situ strengthening, in pitched roof reconstruction 61–62 houses 286–87; pitched roof in- Repair and Reconstruction (R&R) plane diaphragm 287–88; shear project 223, 236, 237 connector 281–83; tie rod for residential policies 351–52 ferrocement belts 283–84; vertical resource mobilisation 74 reinforcement 284–86; suggested retrofi tting of structures: concept of specifi cations 280 266; cost of 270–71; economics Self-employed Women’s Association and advantages of 271; hurdles in (SEWA) 322 applying measures for 271–72; for self-employment activities, for long-term safety 272; measures livelihood restoration and 267; principal steps involved in sustainability 231–32 267–70 self-help group (SHG) 231, 398

Downloaded by [University of Defence] at 01:15 24 May 2016 Rio Declaration 246 shelter reconstruction programme 143 risk management, stakeholders in shelter rehabilitation projects 209 342 Shree Manfara Nav Rachna Samiti 148 risk sharing instruments, Sichuan earthquake 384 development of market-based 398 slum upgradation project 216–19; RRS see reconstruction, repair implementation model of and seismic retroﬁ tting (RRS) participative 219–21 programme Societies Registration Act 225 RRS type ‘C’ villages 101–3, 106, 124 Society for Promotion of Area Resource rural reconstruction package 162 Centres (SPARC) 104, 121 424 Ú Recovering from Earthquakes

solid waste management 175, 179 national-level mental health Special Act for Disaster Affl icted initiatives 329–31; psychosocial Urban Areas, Japan 166 interventions in diff erent phases: SSP see Swayam Shikshan Prayog (SSP) fi rst few months 320–21; initial standard operating procedure (SOP), phases of disaster 319–20; pre- for disaster management 293 disaster phase 318–19; relief and State Disaster Management rehabilitation phase 329 Authorities (SDMA), 145, 244 tsunami see Indian Ocean tsunami structural vulnerability, mitigation of 391–93 United Nations Development Swayam Shikshan Prayog (SSP) 4, Programme (UNDP) 50, 144, 322 104, 118–19; activities during: United States Agency for International Kachchh earthquake 128; Latur Development (USAID) 253 earthquake 127; role in earthquake urban development projects 95, 163 rehabilitation work 137–38 urban earthquake risk mitigation 397 Urban Earthquake Vulnerability Task Force on Mental Health and Reduction Programme (UEVRP) Psychosocial Support in Emergency 249 Settings Guidelines (IASC 2007) 316 urban reconstruction package, for techno-legal and techno-fi nancial Bhuj 162 mitigation regime 395–96 Uttarkashi earthquake 3–4; damage technological disasters, occurrence caused in 39; experience of of 26 working 322–23; implementation tidal waves 22, see also Indian Ocean failure: inadequate feedback 45; tsunami management of NGO eff ort 45–46; Town and Country Planning Acts 247 media management 44–45; policy town planning 4 failures for: clarity regarding Town Planning Act 166 relief entitlement 42; design of town planning offi cer (TPO) 187, 192, relief package 42–43; earthquake 193 shelters 43–44; political town planning schemes 186, 193–98 failures 41–42; relief policy training programmes, post-earthquake and programme 40–41; rescue rehabilitation: for artisans 134–35; operation in aftermath of 39–40 in practice 136–37; in theory 136;

Downloaded by [University of Defence] at 01:15 24 May 2016 for women masons 135 vernacular housing 102 trauma, after earthquakes: in Vulnerability Atlas of India 261 disaster-related mental health care 321–22; documentation 329; wada 58, 68, 83, 96 experience of working: Gujarat Walled City of Bhuj: destruction earthquake 326–29; Marathwada caused by earthquake 196; earthquake 323–25; Uttarkashi objectives of town planning earthquake 322–23; lessons learnt scheme in 187; preparation for psychosocial care 331; mental and sanction of town planning health/psychosocial care 316–17; schemes for 186–87; Draft Town Index Ú 425

Planning Schemes 187–90, 192; women in post-disaster infrastructure planning 198–200; reconstruction: grassroots preliminary Town Planning women’s groups 125–28; Kachchh Schemes 193–98; problems faced earthquake 120; Latur earthquake and lessons learned 190–93; 119–20; rehabilitation activities timeline for 189; problems 122–25; restoration of livelihoods associated with 183–86 121–22; revised approach for warning phase, of disaster cycle 18 128–32 water harvesting system 152 Women’s Livelihood Restoration water management system 152, 174, Project (WLRP) 231 177 World Bank: assessment of policy water supply, provision of 59–60 impacts of 93; post-disaster windstorms, number of people projects 93; in-situ reconstruction aﬀ ected by 20 and participatory development women-initiated activities: during 87–89 Kachchh earthquake 128; World Disaster Reduction Day 252 during Latur earthquake 127; for World Health Organisation (WHO) 291 rehabilitation 122–25; supervision World Vision 149 of masons in building construction 132 Yokohama Conference (1994) 291 Downloaded by [University of Defence] at 01:15 24 May 2016