The Mekong: Governance Challenges

Philip Hirsch Associate Professor University of Sydney

Paper prepared for presentation at the “Water For Irrigated Agriculture and the Environment: Finding a Flow for All” conference conducted by the Crawford Fund for International Agricultural Research, Parliament House, Canberra, Australia, August 16, 2006

Copyright 2006 by Philip Hirsch. All rights reserved. Readers may make verbatim copies of this document for non-commercial purposes by any means, provided that this copyright notice appears on all such copies.

SESSION: BALANCING THE DEMANDS FOR AGRICULTURE AND THE ENVIRONMENT: WORLD BEST PRACTICES OR DISASTERS?

The Mekong: Governance Challenges

PHILIP HIRSCH Australian Mekong Resource Centre, School of Geosciences University of Sydney, NSW 2006, Australia Email: [email protected]

The Mekong Basin is rich in water and other re- Infrastructure projects that isolate the river from its sources, and the six countries that share it include floodplain are of particular concern. some of the world’s poorest on a per capita income basis. They also include some of the world’s fast- The Mekong River Commission, established in est growing economies. Two of the countries also 1995, provides a mechanism for shared manage- have fast growing rural populations. ment of the water resources of the Mekong River. Water is also used and managed at other levels The Mekong is a shared resource. It is shared within each riparian country. Governance of water across borders, among more than 60 million in the Mekong Basin is a key challenge. people who live in the basin, between those who live in the basin and others who draw on its This paper describes the current and potential resources, between farmers, fishers, power pro- uses of water in the Mekong Basin for irrigated ducers, consumers and many others. agriculture, fish production, the generation of hy- dropower and environmental services. It examines Water appropriation for human use is currently the effects of current and proposed dams on the only about 12 per cent of total flow. Although dams natural environment and on the sustainability of currently regulate only five per cent of the waters in agriculture and fisheries. The paper also considers the basin, many large hydropower dams have the effectiveness of the Mekong River Commission been proposed or are under construction. as an organisation enabling shared management of the basin’s water resources, looking particularly Environmental issues include the need to maintain at governance challenges in a transboundary river the seasonal water flows that underpin the ecology basin that is subject to heavy development pres- of Tonle Sap (Great Lake) in Cambodia, the man- sures. agement of flooding, acid sulphate soils and salt- water intrusion in the delta, and the need to control logging and deforestation throughout the basin. The Mekong River and its basin PHILIP HIRSCH is Director of the Australian Me- kong Resource Centre and Associate Profes- ‘The world’s atlases no longer tell the sor of Geography at the School of truth’5 Geosciences, University of Sydney. He has published extensively on environment, natural Maps of the Mekong River and its tributaries do resource management and rural development not yet lie — unlike the maps of many interna- in SE Asia. He speaks and reads Thai, Lao and tional basins whose rivers only intermittently com- Vietnamese and is learning Khmer. Recently, plete their route to the sea. The Mekong, it seems, he has been engaged in a number of studies has plenty of water — but does it, and for how on water governance in the Mekong region with long? AusAID, Danida and several NGOs. He is joint lead author of the chapter on river basin devel- opment and management for the IWMI Com- 5 Fred Pearce, ‘Growing thirst of an arid earth’, The Austra- prehensive Assessment on Water Management lian, 22–23 July 2006, p. 25. This article was abstracted in Agriculture. from Pearce (2006).

W ATER FOR I RRIGATED A GRICULTURE AND THE E NVIRONMENT 43

The Mekong River drains a basin area of 795 000 The Mekong as a shared resource km2.The Mekong Basin is home to close to 70 mil- lion people, most of whom live in rural areas. The 4800-km long Mekong River is a shared re- Farming, fishing and forest use continue to sustain source. It is shared across the borders of six coun- livelihoods, while urban growth, industrialisation tries — China, Burma, Laos, Thailand, Cambodia and regional integration mark the sharp end of the and Vietnam. It is shared among a population some wider economy. The basin is culturally diverse, 35 times that of the Murray–Darling Basin on a with multiple ethnic groups practising agriculture smaller territory. It is shared between those who and depending on natural resource-based liveli- live in the basin and others who draw on its re- hoods in ways adapted both to exigencies of the sources, both virtually and physically. It is shared physical environment and to the pressures and op- between different types of resource users — farm- portunities presented by social, political and eco- ers, fishers, power producers and consumers, and nomic change. many others. Diversity and dynamism thus characterise the ba- Each dimension of sharing represents, alterna- sin. They also shape its challenges. tively, an avenue for cooperation, for competition or for conflict over the Mekong as a shared re- Bio-physically, the Mekong is an extraordinary source. The relative abundance of water in the Me- river in a number of respects. With 1300 endoge- kong has, to date, meant that competition and nous fish species known and several hundred likely conflict have been limited and localised. It has also yet to be discovered, the Mekong is the second meant that frameworks for cooperation have de- most ichtyofaunally bio-diverse river system in the veloped around notions of abundance rather than world, after the Amazon. This biodiversity is im- scarcity. This has resulted in institutions with a portant in its own right, but it also supports the marked developmentalist orientation in their ethos world’s largest freshwater fishery, one whose catch and charter. Conversely, it has tended to minimise of two to three million tonnes per annum provides the regulatory role and potentials of such institu- 40–80% of the animal protein consumed by the tions. region’s poor. The climate of the basin is largely monsoonal, resulting in monthly mainstream flow Governing the Mekong as a shared resource is be- regimes in the wet season some 15–20 times those coming increasingly challenging, as development of the dry season. Seasonal flow variations of pressures pose choices and tradeoffs. Integrated many tributaries are even greater. Reverse flows in Water Resource Management principles provide an the Tonle Sap River feeding the Great Lake in overall framework for holistic basin management. Cambodia represent a peculiar hydrological phe- However, the IWRM framework remains poorly nomenon and also provide the basis for subsistence understood, broadly interpreted and loosely ap- of a significant part of the Cambodian population. plied. The Mekong River is one of the less developed of the world’s major rivers, nowhere near ‘closure’ in Irrigation development and the the sense of having most of its water committed environment (Molle et al. 2007). Only about 12% of its water is physically regulated or extracted, and less than 5% Most agriculture in the Mekong Basin remains of its hydro-electric energy potential is exploited. rainfed. Irrigation development is largely limited to The Mekong is also home to some of the world’s the delta, areas irrigated by storage dams in north- cash-poorest people, with provincial per capita an- eastern Thailand, and pump irrigation along the nual GDP typically at the US$200 level in central incised tributaries of the Mekong and lower tribu- parts of the basin. Not surprisingly, the river faces tary floodplains in Thailand and Laos. significant development pressures. Overall rice production is more than sufficient to feed the basin population, so endogenous demo- graphic growth is not a reason per se to expand irrigated staple grain production. However, a num- ber of factors result in pressure for expansion of irrigation. First, agricultural production is highly uneven within the basin. In Vietnam, the delta pro- duces about half the country’s rice output, and the

W ATER FOR I RRIGATED A GRICULTURE AND THE E NVIRONMENT 44 surplus forms the bulk of Vietnam’s exports that astrous irrigation works of the Khmer Rouge era. rank the country as the second-largest world ex- Irrigation development also includes private in- porter after Thailand. In Thailand, the main part of vestment in tank reservoirs that capture water in the country that lies within the basin, Isan (or the the recession areas of previously-flooded forest North-east), has always been the poorest and natu- around Tonle Sap and release it by gravity to the rally driest part of the country. Geopolitical con- newly-created surrounding rice fields. cerns in the past, and party political concerns more recently, have directed development resources to In Vietnam, the major dry-season water competi- Isan, and water resource development has loomed tion comes from irrigation in the upper delta prov- large in this relatively dry region. Cambodia and inces at the expense of the coastal provinces. Laos remain poor countries with local and national Intensification of rice farming through increased food security concerns. Production varies consid- cropping density has exacerbated this trend. Sec- erably between provinces. Relatively little rice is ondary impacts include high concentrations of fer- grown in the steep gorges of the Chinese section of tiliser and pesticide in runoff that affect fisheries the basin in Yunnan province, except on the low- and water quality throughout the delta. land terraces of Xishuangbanna. The environmental implications of irrigation ex- Thailand, Laos and Cambodia all have irrigation pansion are still poorly understood. At a basin- expansion on their agendas. The main drivers of wide level, increased dry-season abstraction will irrigation development are specific to each coun- exacerbate saline intrusion in the delta. Low water try. levels of tributaries such as the Sedone in southern Laos are reported by farmers and fishers and are Politics have driven various large-scale schemes to perceived, at least, to be a result of irrigation ab- ‘green the North-east’, from the Thai military’s stractions and to contribute to fisheries decline Isan Khiew program of the 1980s, to the Khong- (Bush and Hirsch 2005). In Cambodia, the clearing Chi-Mun grand vision of the 1990s, to the so- of flooded forests and conversion of areas previ- called Water Grid network of the current decade. ously under recession rice farming to more perma- The latter would involve linking river basins, in- nently-bunded fields have far-reaching cluding importing water from the Nam Ngum and implications for fish ecology of the Great Lake. Xe Banghieng tributaries in neighbouring Laos This is even before the larger-scale impoundments though giant siphons under the Mekong River. The are considered — most of these are primarily for water grid vision also includes investment in pipes hydropower, but many have a secondary purpose to replace open channels, in part to achieve effi- in irrigation development. ciencies but in part also to greatly increase the irri- gable area of Isan’s rolling landscape. Energy and the environment In Laos, the major food security drive has come with investment in dry-season pump irrigation de- The Mekong has been estimated to have a hydro- velopment. Subsidised diesel pumps from India electric potential of 30 000 megawatts. Dams have were distributed around the country during the late been on the agenda for the Mekong since the in- 1990s. Dry-season farming that draws water from ception of the Mekong Committee in 1957. A Na- the incised rivers of the left bank floodplain is of- tional Geographic special issue in December 1968 ten uneconomic, particularly with rising costs of featured a front cover entitled ‘River of Terror and diesel. Nevertheless, increasing coverage of the Hope’, the hope being the prospect of a cascade of electricity grid and relatively cheap and partly sub- dams that would have turned the Mekong into a set sidised electric power costs have resulted in sig- of stepped lakes from northern Laos to central nificant expansion of the irrigated area. Cambodia. The Mekong Committee’s indicative plans in 1970 and 1987 respectively featured main- Cambodia is similarly embarking on a program to stream hydropower as their mainstay. The Indo- expand its irrigated area, focusing mainly on sup- china conflict ensured that these plans remained on plementary wet-season gravity-fed systems but paper. During this period, however, several large also involving dry-season projects such as the tributary dams were built in Isan, and the Nam ADB-funded Stung Chinit Project and North-west Ngum dam in Laos represented the main concrete Irrigation Sector Project in the Tonle Sap catch- manifestation of the Committee’s vision. ment. These schemes include rehabilitation of dis-

W ATER FOR I RRIGATED A GRICULTURE AND THE E NVIRONMENT 45

By the time the main conflict finished with the have been proposed are eventually built. Indeed, 1991 Paris peace accords on Cambodia, attitudes one dam has already had major transboundary im- toward large dams had changed. Early attempts at pacts on fisheries, health and safety for down- the Mekong Secretariat to dust off old plans, and in stream communities. The Yali Falls dam in particular to implement the Pa Mong project — Vietnam’s Central Highlands generates benefits for which would flood an area of north-eastern Thai- Electricity of Vietnam and the domestic and indus- land and central Laos currently occupied by about trial consumers of Ho Chi Minh City, but at the half a million people — elicited a strong reaction expense of several tens of thousands of mainly from environmentalists and those affected by, or ethnic minority communities living along the familiar with, the dams that had already been built Sesan in Ratanakiri and Stung Treng provinces in in Thailand. A revised vision of mainstream ‘run- downstream Cambodia. of-river’ dams was drawn up, but even the much smaller impoundments were increasingly recog- The environmental and social impacts of dams nised as potentially devastating to the river’s mi- have been much debated, and the Mekong was one gratory fish and to other aspects of the river’s of the focal river basins for the World Commission ecology (Hill and Hill 1994). on Dams study (WCD 2000). Some of the key questions that arise are over the impoundment and While the mainstream dams on the lower Mekong barrier effects on longitudinal migration of fish; have more or less — though not entirely — disap- the impact of a changed flood regime that isolates peared from the agenda, a cascade of large dams is the river from the floodplain, with impacts on pro- underway on the upper Mekong (the Lancang Ji- ductivity and lateral fish migration; the impacts on ang) in China. Two are completed, three more are the flood regime of Tonle Sap; the direct and indi- under construction, and the building of at least an- rect social dislocation impacts; downstream ero- other two will commence soon. Two of the struc- sion; forest inundation; and salinity impacts in Isan tures are gigantic; the Xiaowan Dam is 300 m tall. and Laos that result from raised water tables. An- The combined storage is sufficient to affect both swers to these questions depend in part on the scale dry and wet season flows significantly as far as of impact under discussion. Given the relatively Tonle Sap and the delta, and of course the relative open nature of the Mekong compared to river ba- effects are even greater in the middle reaches. sins with much higher levels of existing abstrac- tion, the incremental impact of tributary dams — While no mainstream dams have been built on the or even the medium-sized Chinese dams on the lower Mekong, there has been an acceleration in mainstream — may not be great on a system-wide construction of tributary hydropower dams. In part basis. However, as the examples of Theun- because of the ecopolitical obstacles to further dam Hinboun, Pak Mun and Yali Falls have shown in construction in Thailand, but in part due to its Laos, Thailand and Vietnam/Cambodia respec- natural advantages, Laos has staked its economic tively, tributary dam impacts are considerable at a future on the export of electricity to Thailand and, more localised level. at a later date, to Vietnam and possibly China. Since the 1971 opening of the Nam Ngum dam, electricity has been the country’s main or second- Governing a transboundary largest source of foreign exchange. Today’s dams basin: the Mekong River are mainly financed through public-private part- nerships, assisted with loans and risk guarantees by Commission the World Bank and Asian Development Bank. The Mekong River has been ‘governed’ by a Despite more thorough EIA studies, significant transboundary cooperation framework for nearly problems have been experienced with most of half a century. The Mekong Committee was estab- Laos’ recent tributary dams. The largest and most lished in 1957, but its main agenda was to plan controversial is the World Bank-supported Nam dams rather than to share water or to regulate al- Theun 2, which is now under construction after a terations to the flow regime. In some respects, the 15-year period of studies and controversy. 1995 establishment of the Mekong River Commis- While tributary dams have somewhat lesser indi- sion (MRC) under the Agreement on the Coopera- vidual impact on the transboundary basin than do tion for the Sustainable Development of the mainstream dams, their cumulative impact would Mekong River Basin was a product of the sustain- be profound if a significant number of those that able development era, seeking to optimise eco-

W ATER FOR I RRIGATED A GRICULTURE AND THE E NVIRONMENT 46 nomic, social and environmental values, but within environmental and other water requirements in a very loose — some would say non-existent — their respective areas. regulatory framework. Moreover, the MRC carries some of the baggage of the Committee, and like many other international sustainability agencies its Governance challenges mandate is open to interpretation. In conclusion, water governance in the Mekong The MRC is governed by a Council at the political poses significant challenges, including: level and by a Joint Committee at the senior bu- • Moving beyond national interest. In a shared reaucratic level, with representation from each of river basin with a history of mistrust between the four member states. Neither China nor Burma riparian countries, questions of sovereignty is a member. The MRC Secretariat carries out stud- loom large. The MRC remains a governed ies of fisheries, flood mitigation and other key ar- rather than governing agency. Further, national eas to build up the requisite knowledge base for interests as represented unitarily at the MRC managing the river. It has also housed three key need to reflect the diversity of stakeholder po- cross-cutting programs: the Environment Program, sitions and concerns over water resource man- the Water Utilisation Program and the Basin De- agement in the respective national territories. velopment Plan. The latter has involved a process • Providing pathways for the knowledge base to of sub-area planning, in principle to establish a lead to better development decisions. A great rational, equitable and participatory process of de- deal of work has been done on flows model- veloping water resources in sub-basin areas in a ling, building an understanding of the value of sustainable manner with respect to the basin as a fisheries and so on. Scenarios have been de- whole. The Water Utilisation Program is designed veloped at a basin-wide level. However, the to provide the knowledge required to establish wa- MRC has not been proactive in putting these ter-sharing procedures and rules between riparian scenarios out for public or even policy-level members. The Environment Program is geared to consideration. The Commission has also had understanding the basin as an interconnected sys- little role to date in assessing implications of tem, so that impacts of development can be under- the larger hydropower developments such as stood properly and assessed with regard to their Nam Theun 2. system-wide transboundary implications. An Inte- • Engaging China. China has remained largely grated Basin Flows Management program is cur- outside the framework of water governance, rently examining environmental flow requirements despite the fact that half the river’s length in a multidisciplinary exercise that is mainly tar- flows through Chinese territory and that the geted at the mainstream. major dams are being built in China. There are signs that China is ready to consider more sub- National Mekong Committees are supposed to co- stantial engagement or even membership of the ordinate work relevant to the MRC in each riparian MRC. At the very least, this might open poten- country. The technical and political capacity of tials for negotiating patterns of flow releases NMCs varies considerably from one country to from the dams. another. Further, the NMCs represent quite a nar- • Maintaining a focus on the triple bottom line. row set of riparian interests in each country, in a The MRC’s mission is to promote an economi- river basin whose development agenda no longer cally strong, environmentally sustainable and — if it ever did – limits competition for water to a socially just river basin. The hard decisions national interest level (Hirsch et al. 2006). Inte- and proactive leadership required to achieve grated Water Resources Management requires a such ends require a degree of trust, political governance framework that is more inclusive of commitment and community engagement that diverse stakeholder interests and that engages with to date have eluded the Commission. the basin communities. • Working at multiple levels on the basis of an River basin committees at the sub-basin level have integrated system. The MRC has been princi- been established in Thailand and in some parts of pally focused on the mainstream, while most of Vietnam. While they are stakeholder-oriented, the development pressures and impacts have these river basin organisations are still not func- been on the tributaries, wetlands and flood- tional in managing tradeoffs between agricultural, plains. IWRM requires an approach to basin development at multiple scales and in recogni-

W ATER FOR I RRIGATED A GRICULTURE AND THE E NVIRONMENT 47

tion of the interconnectedness of different parts References of the system. Bush, S. and Hirsch, P. 2005. Framing fishery decline. Aquatic Resources, Culture and Development In conclusion, governing the Mekong as a trans- 1(2), 79–90. boundary and integrated system is about much more than cooperative mobilisation of financial Hill, M.T. and Hill, S.A. 1994. Fisheries Ecology and resources for water resource development. It is also Hydropower in the Mekong River: An Evalua- about more than negotiating between sovereign tion of Run-of-the-River Projects. Mekong Se- riparian member countries. While these two con- cretariat, Bangkok, 106 pp. cerns drove an earlier era of cooperative manage- Hirsch, P., Jensen, K.M., Boer, B. et al. 2006. National ment, the sharing of water between farmers, interests and transboundary water governance in fishers, power producers and consumers and many the Mekong. Australian Mekong Resource Cen- others requires a prior commitment to the well- tre, School of Geosciences, University of Syd- being of the river as an ecosystem that supports the ney, Sydney, in collaboration with Danida. livelihoods and cultures of tens of millions. http://www.mekong.es.usyd.edu.au/projects/mek ong_water_governance.htm Molle, F., Wester, P. and Hirsch, P. 2007. River basin development and management. In: Molden, D. (ed.) Water for Food, Water for Life: the Com- prehensive Assessment of Water Management in Agriculture. Earthscan, London. 664 pp. Pearce, F. 2006. When the Rivers Run Dry: What Hap- pens when our Water Runs Out?, Beacon Press, Boston, 367 pp. WCD 2000. Dams and Development: A New Frame- work for Decision-Making. Report of the World Commission on Dams. Earthscan, London, 404 pp.

W ATER FOR I RRIGATED A GRICULTURE AND THE E NVIRONMENT 48