World Bank Safeguard Policies Are Not Sufficient for Hydropower Projects

Date: 
Tuesday, June 14, 2005

Ms. Janet West
Export Credits Division
Trade Directorate, OECD

Dear Ms. West:

I would like to contribute to the process whereby the Participants to the OECD Arrangement on Officially Supported Export Credits and the environmental practitioners are currently considering strengthening their environmental policies for large hydro projects. May I ask you to kindly forward this memorandum to the Participants and the environmental practitioners? My expertise is that of an academic researcher and a consultant who has been studying the social and environmental impacts of large dams, especially hydro, starting with the Legacy of Dams on the Zambezi: Group Works to Right Wrongs at Kariba Dam on the Zambezi in the mid 1950s. In particular I would like to compare the utility of the World Bank’s safeguard policies with the policy principles of the The World Commission on Dams (WCD). I am familiar with both.

As a consultant to the World Bank on and off since 1964, I was the Bank’s principal resettlement consultant during the 1980s on the Bank–financed Sardar Sarovar Project in India and, at the request of the Bank, I served as resettlement adviser to the Canadian consortium that completed a major feasibility study of China’s Three Gorges Project in the mid 1980s. I have also served on three World Bank–required environmental and social expert panels dealing with Lesotho’s Highland Water Project, China’s Longtan Project and, ongoing, Laos’ Nam Theun 2 Project. As a WCD commissioner, I was able to familiarize myself with the Commission’s entire data base because I had recently retired from the California Institute of Technology.

In this memo I would like to emphasize two major reasons why the approach of the World Commission on Dams to social and environmental issues of large hydro is not only superior to that of the World Bank, but also is the most appropriate for the "do–no–harm" mandate of the Export Credit Agencies. The first reason is that WCD policy principles apply specifically to large hydro while the World Bank’s safeguard policies apply to any Bank–financed project that raises environmental and social issues. The second reason applies to involuntary resettlement which prominent experts representing all sides of the large dams debate agree is the most contentious issue associated with large hydro.

Concerning the first reason, whereas WCD’s policy principles apply specifically to hydro dams, the World Bank has evolved 10 discrete safeguard policies to deal with different types of development interventions regardless of scale. The result is a catch all, least common denominator effort. Because of the large scale of hydro projects, or multipurpose projects with a major hydro component, all ten policies may apply. That requires ten separate studies which may overlap, on the one hand, or, on the other hand, pay insufficient attention to other important environmental and social issues.

Robert Goodland, the recently retired Chief Environmental Adviser to the World Bank Group, was well aware of the problem. As a result he drafted for the Bank an overarching state–of–the–art safeguard policy to deal specifically with large dams. Surely large dams, and, especially contemporary Bank–assisted projects like the Lesotho Highland Water project and Laos’ Nam Theun2 Project, warrant such an overarching policy since they, as with past Bank–financed hydro projects such as Kariba and Nigeria’s Kainji, are the largest single development project in a nation’s development portfolio at the time of their construction. But for reasons never explained the Bank failed to approve and implement such a policy.

The downstream environmental and social impacts of large hydro, and especially large hydro on mainstreams, illustrate the environmental and social risks of the Bank’s failure to require an overarching policy for large hydro. While major environmental impacts to wetlands, deltas and offshore fisheries are apt to be irreversible, insufficient attention continues to be paid to them in the Bank’s policies. The same applies to downstream impacts on riverine communities where natural flows are the basis of economies involving flood recession crop agriculture and livestock management, flood–dependent fisheries, flood–dependent forests essential for provision of building materials, fuel and non timber forest products and community water supply–dependent aquifer recharge. Such economies, supporting tens of millions of people, are especially important in late industrializing countries where most future hydro will be built. While the World Bank in recent years has appointed a committee on environmental flows, these have yet to be adequately incorporated within Bank safeguard policies whereas the importance of environmental flows is covered by the World Commission on Dams.

Concerning the second issue that I wish to address, there is a fundamental and crucial difference between the WCD approach to involuntary resettlement and the World Bank’s. WCD’s policies require that resettlement plans be implemented to improve the living standards of resettlers as does the November 1995 Position Paper on Dams and Environment of the International Commission on Large Dams (ICOLD). While the World Bank states that improvement is the most desirable outcome, Policy OP 4.12 (April 2004 Revision) allows governments and project authorities the fallback position of restoring incomes and living standards. That is the position that governments and project authorities are apt to take as in the case of the Lesotho Highlands Water Project. Yet it is a position that has been shown by research not only to not restore incomes but to leave the majority impoverished which is why I consider Bank policies as partly responsible for failed resettlement. There are several explanations for such a result, five of which are briefly mentioned below.

First, planners assume wrongly that a compensation policy can restore living standards rather than a balance between compensation and development initiatives. The Bank’s OP 4.12 is at fault here since ‘compensation’ is mentioned 19 times while ‘development’ is mentioned only four times. Second, a restoration approach fails to take into consideration the fact that living standards for a majority of resettlers tend to drop during the long planning process (often over ten years) that proceeds construction of a large hydro project and during the initial years immediately after physical removal. One reason is that government and other external investments, including school, clinics and economic development programs, are stopped while those to be resettled are often told to stop improving housing, upgrading farms and making other investments in what planners assume will be a future reservoir.

In addition the process of physical removal makes heavy demands on resettlers that delay for one or more years their re–establishing themselves in a new area, with a new host population with whom resettlers must compete for land, social services and jobs, and with, more often than not, an increased government presence. Third, pre–project surveys carried out to establish a benchmark against which restoration can be measured are known to underestimate income and living standards which have already been lowered due to project–related cessation of investments in the area. Fourth, the Bank’s safeguard policies deal only with direct economic and social impacts. Ignored are a wide range of socio–cultural effects associated with forced removal from a preferred homeland, the psychological stress affecting the elderly and women, and increased rates of illness and death that have been reported in resettlement areas where, more often than not, population densities increase, and water supplies and food – at least during initial years – are inadequate. Fifth, resettlement tends to be associated with increased cash expenditures because rural resettlers, who remain the large majority in connection with large hydro, are moved to less fertile soils which require costly inputs to provide equivalent yields, have less access to common property resources for grazing, fuel, building materials and for foraging, and become more dependent on credit and the risk of indebtedness.

Ted Downing, research professor at the University of Arizona also explains why the World Bank policy "institutionalizes a negotiating system that potentially violates human rights." Moreover, after mentioning that involuntary resettlement is accompanied by impoverishments risks, the Bank’s policy, unlike the policies of the World Commission on Dams, fails "to propose measures to address them. Instead, it falls back on the same flawed economic analysis and methodologies that have been responsible for decades of unacceptable performance." Failed performance has been acknowledged even by such senior World Bank officials as Robert Goodland and Michael Cernea, the Bank’s recently retired Senior Adviser for Social Policy and Sociology, and within the Bank’s Operation’s Evaluation Department (OED). OED’s 2001 Involuntary Resettlement: Comparative Perspectives that dealt only with large dams, concluded that "the record on restoring – let alone improving – incomes has been unsatisfactory." More to the point, the authors, one of whom was OED’s Director, concluded for dam projects that "Above all, displacees must be beneficiaries of the project. Merely aiming to restore standards of living and lifestyles common to isolated river valleys can be a dead–end development strategy." The fact that the Bank’s managers are unwilling to change a resettlement policy which its own most knowledgeable staff conclude is flawed is a telling critique of a policy which should not be used as a model by Export Credit Agencies.

Please feel free to contact me if I can be of further assistance. The issues discussed in this letter are dealt with in greater detail in my The Future of Large Dams: Dealing with social, environmental, institutional and political costs, Earthscan, London, 2005.

Yours sincerely,

Thayer Scudder
Professor of Anthropology, Emeritus

Available in PDF format