An Overview of Revisions to the World Bank Resettlement Policy

Dana Clark


Prepared for the Workshop on Engendering Resettlement And Rehabilitation Policies and Programmes in India

(New Delhi; September 12, 2002); to be published in L. Mehta (ed.), 

Gender and the Politics of Forced Displacement (forthcoming, Sage).


I.  Introduction

The World Bank first adopted a policy on involuntary resettlement in 1980 in order to provide substantive and procedural policy protections for people who are displaced from their homes and livelihoods by World Bank-financed projects.  In doing so, the Bank assumed a leadership role in addressing the cycle of impoverishment that often threatens people who are forced to abandon their homes or livelihoods as a result of development-induced displacement.


In light of the World Bank’s mandate of poverty alleviation, and given the wealth of data showing the extreme risks of impoverishment associated with involuntary resettlement, the resettlement policy is an important recognition of the rights of affected people. It was developed as part of a policy-based effort to shift the Bank’s lending practices towards more sustainable development.    


The World Bank is operating with public money and with a mandate of poverty alleviation.  The World Bank claims to champion the right to development. And the World Bank has consistently documented the extremely high likelihood of impoverishment that accompanies development-induced displacement.  Unfortunately, the hard lesson of twenty-plus years of experience with the resettlement policy is that the majority of people who are uprooted in the name of development do not easily recover, much less improve, their previous standards of living. While the 1980 policy was an important step forward, numerous studies by the World Bank[i] and others[ii] have shown that the policy is not enough. The Bank’s track record in projects involving involuntary resettlement has been a dismal failure.[iii]  This poor track record calls out for a fundamental reassessment of the way in which World Bank projects are conceived, executed, and supervised.


The recent policy revision process that resulted in the December 2001 version of the involuntary resettlement policy was disappointing in many ways. The net effect of the changes that were introduced is to promote a more “checklist” approach to policy compliance and decrease emphasis on accomplishing improved results on the ground. In some cases, the changes will make it easier for project authorities to evict people from their lands while still complying with the policy – because compliance has gotten easier as rights are reduced. This, unfortunately, is the outcome of the Resettlement Thematic Group’s emphasis on “operationalizing” the policy and “enabling” its implementation.


This paper discusses the changes incorporated in the World Bank’s revised policy on Involuntary Resettlement, which, after a five-year revision process, was approved by the Board of Executive Directors in late 2001, and became effective for projects initiated after January 2002.[iv]  The paper highlights significant policy debates at stake in the resettlement conversion. The policy discussion will be prefaced by a brief examination of two controversial resettlement projects.  Where possible, there is an assessment of the gender implications of resettlement in the context of project failures and the policy revisions.


II. Overview of Context: Resettlement and Reality     

The basic objectives of the policy are that displaced persons should benefit from the project, and that they should have their standard of living improved, or at least restored.  All too often, failure to accomplish these objectives relegates displaced communities to impoverishment and stagnation.  In order to provide context, the paper examines two of the Bank’s most significant resettlement project, both in India.  These projects have been extensively studied, and provide vivid illustrations of how resettlement policy violations at the design stage play out during implementation.


Sardar Sarovar and Singrauli share certain characteristics, with the most basic being that they violated the requirements of the resettlement policy and the interests of local affected people.  In both of these cases, senior Bank officials pushed for the projects to be approved by the Board even though there were clear violations of the Resettlement policy. They were both rushed through the loan approval process in order to be funded by the end of a fiscal year.  They had disproportionate impacts on marginalized communities and particularly on ethnic minorities, tribal, and lower-caste people; they failed to adequately consider the social and environmental impacts of the project; and were designed without adequate consultation with affected people. 


A.  The Sardar Sarovar Project

Possibly the most well-known struggle against unjust resettlement has been in the context of the Sardar Sarovar and other dams planned for the Narmada River.  A 1985 World Bank loan jump-started construction of the Sardar Sarovar dam and canal project, at a time when the Indian government had not yet completed its environmental review and the project had not received the necessary clearances. In addition to bypassing the requirements of Indian law, the project also violated the World Bank’s policy requirements relating to indigenous peoples, environmental assessment, and involuntary resettlement. 


Although the resettlement policy had been on the books for five years at that time, the policy requirements were almost completely ignored by the Bank at the time of project preparation and approval by the Board of Executive Directors.


In 1985, when the credit and loan agreements were signed between the Bank and the three states, no one knew the scale of the displacement that would result from the Sardar Sarovar Projects, nor did anyone have anything like a true picture of the peoples who would be displaced, nor had the people themselves been consulted. . . . no basis for designing, implementing, and assessing resettlement and rehabilitation was in place. . . . Nor were there benchmark data with which to assess success or failure.  As a result, there was no adequate resettlement plan, with the result that human costs could not be included as part of the equation. Policies to mitigate those costs could not be designed in accord with people’s actual needs.[v] 


This was one of the many devastating conclusions reached by an independent review of the project.[vi]   The independent review, known as the Morse Commission, was created after a mobilized grassroots movement challenging the project joined forces with an international solidarity campaign to raise concerns with member governments of the World Bank.  In response, the World Bank agreed to the first-ever independent investigation of a Bank-financed project. These events had tremendous impacts on both the project and the institution.


The combined scrutiny of this broad constituency, together with the findings of the Morse Commission of widespread violations of the Bank’s policies to the detriment of local people and the environment, fed into growing civil society pressure for greater transparency and accountability in World Bank operations.  Local and international activists called for the World Bank to withdraw from the project.  In 1993, when the government and the Bank failed to take adequate steps to deal with the findings of the Morse Commission, the government of India indicated that it would not seek any further disbursements under the loan, and World Bank lending for the project was terminated.[vii]


The precedent that was set by having an independent review of the Sardar Sarovar project helped lead to creation of a permanent, quasi-independent accountability mechanism at the World Bank that was designed to be responsive to the concerns of local people relating to violations of World Bank policies and procedures.[viii]  This entity, known as the World Bank Inspection Panel, investigates cases where local people believe that their rights and interests have been or are likely to be harmed as a result of the World Bank’s violations of its own policies and procedures, and issues its findings in a report to the Board of Executive Directors and the President.  One important limitation on the Panel’s jurisdiction is that it can only hear claims that relate to projects where the Bank has not fully disbursed the loan. Once 95% of the loan has been disbursed by the Bank, a claim cannot be filed.[ix]  Thus, Sardar Sarovar and other similar projects, where the loan has been disbursed but not repaid, fall into an accountability gap at the Bank, where there is no formal opportunity for aggrieved parties to seek redress from the Bank for projects that are continuing to violate Bank policies during implementation.


B.      Singrauli

The Bank initially got involved in the Singrauli region in 1976, when it financed the National Thermal Power Corporation’s (NTPC) Singrauli Super Thermal Power Plant, the first coal-fired power plant in the area; it also financed the extension of that plant in 1980.  In 1985, the Bank financed one of the area’s first open-pit coal mines, Dudhichua. The Bank helped connect the Singrauli power plants to the Northern grid with another loan to NTPC in 1985.


Through this series of loans, plus others made to Northern Coalfields Ltd., the World Bank provided the seed money for the transformation of Singrauli, from an area rich in biodiversity and subsistence agriculture to an industrial wasteland.  Local people and the environment have paid a devastating price from the industrialization of Singrauli, which is notorious as one of the world’s worst examples of failed resettlement, because so many people have been displaced so many times. A 1991 World Bank study found that 90% of the people surveyed had been displaced at least once, and 34% had been displaced more than once.[x]  By 1993, when the Bank provided an additional $400 million to NTPC, hundreds of thousands of people had been displaced by the coal activities and the Rihand dam (which was not financed by the World Bank).


This loan took the Bank and NTPC to the $4 billion mark in their lending relationship, and made NTPC at that time the Bank’s largest borrower.  The 1993 loan was to support the expansion of two NTPC power plants in the Singrauli area, Rihand and Vindhyachal, and the related expansion of waste disposal areas known as ash dikes.[xi]  In addition, the 1993 loan included provisions to deal with the ongoing crisis in resettlement and rehabilitation.  However, despite the clear and obvious resettlement problems in Singrauli, and the contemporaneous high-profile critique of the Bank’s handling of similar issues in the Sardar Sarovar project, senior Bank management still failed to comply with the provisions of the Bank’s Involuntary Resettlement policy when preparing the 1993 loan. 


The resettlement plan for the 1993 loan, which was destined to become corporate policy for NPTC for all of its projects in India, was negotiated in two weeks in Washington, just prior to loan approval, in violation of the requirements of the resettlement policy.[xii]  The resettlement plan was not based on consultation with affected people, failed to reflect the reality on the ground, and lacked ownership by NTPC.  Furthermore, the rush to approval seemed to be generated solely by Bank management (including the same Regional Vice President who had been involved in Narmada), who “pressured staff to accelerate processing of the loan in order to meet fiscal year 1993 lending targets.”[xiii]


Members of the World Bank’s Board of Executive Directors were fully aware that the NTPC project was in violation of OD 4.30 when they nevertheless gave it their approval two days before the close of the fiscal year.  The German Executive Director, who abstained rather than support the loan, observed that:


With respect to the Resettlement Action Plans for the two extension projects, we understand that staff has reached last minute agreements with NTPC . . . . These agreements, which are not summarized in the SAR [Staff Appraisal Report], as they should according to OD 4.30, forsee, among other things, a supplementary socio-economic survey of the project affected persons. . . . .our understanding of OD 4.30 is that such surveys should be completed before board consideration, and that respective costs and budgetary provisions should be reflected in the SAR.  In view of the mixed experience with resettlement in the past and great sensitivities in our capitals and parliaments vis-à-vis these issues, we would have preferred to delay board consideration until all provisions of the OD 4.30 on involuntary resettlement are met.[xiv]


Environmental and health studies that were supposed to be a condition of the 1993 loan were simply never done.  The fundamental design and assessment failures ensured further imposition of misery on the local people living in the way of these ever-expanding power plants and coal mines. The decision to rush the project to approval despite the fact that Bank staff, Management and the Board all knew that the resettlement plan was flawed and non-compliant represents a disturbing abdication by all levels at the Bank of its fiduciary obligations to protect the interests of the people living in the project area, affected by their lending decisions. These decisions and policy violations had devastating repercussions during project implementation.


Building on accountability reforms that grew out of the Narmada struggle, people being evicted from their lands in Singrauli without adequate resettlement or rehabilitation chose to file a claim to the Inspection Panel in May 1997. The Inspection Panel recommended a full investigation of the claim, but was blocked from conducting a field investigation by the Indian government’s refusal to allow the Panel access to its territory.[xv]  Instead, the Panel undertook a Washington-based desk review that documented serious violations of the policies.[xvi]


In response to the claim, the Bank and NTPC came up with an action plan that included the creation of a 3-member team of Indian nationals authorized to make recommendations for improving the compensation package in order to bring the project into compliance with the World Bank policy and the 1993 NTPC R&R policy.  Known as the Independent Monitoring Panel, or IMP (not to be confused with the Inspection Panel), the team -- all men -- had extensive consultations with affected communities in Singrauli, and earned some degree of trust and respect from the people. 


As a result of the IMP’s investigations and the negotiations that it undertook with NTPC on behalf of the affected communities, a revised compensation package, which became known as the “IMP package,” was developed and offered by NTPC to approximately 1200 families who were considered by the Bank and NTPC to belong to “Stage 2” displacement in Singrauli. The claimants, and others being displaced by the expansion of waste disposal areas for the power plants being financed by the 1993 loan, were considered Stage 2.


Those people who were termed “Stage 1,” were had been displaced by the original construction projects, including the World Bank-financed Singrauli Super Thermal Power Plant and they constituted a majority of displaced people in Singrauli, living in truly desperate circumstances in grim resettlement colonies.  NTPC ultimately rejected the IMP’s attempts to address remedial measures for Stage 1 people, despite the fact that the 1993 loan agreement specifically provided for improvements in resettlement conditions for Stage 1 people. 


Gender Implications In Singrauli

Another major problem with resettlement in Singrauli, and, unfortunately, with the IMP remedial process itself, has been the marginalization of women. Many people interviewed by the author stated that women had not been separately consulted about their contributions to household income or their economic relationship to their lands by the World Bank, NTPC, the Inspection Panel or the IMP. One woman now living in Dodhar resettlement colony explained how the IMP package failed to compensate for the impacts of displacement on all members of the household:


Two persons’ labor was displaced as a result of the project.  Both husband and wife should have been compensated for their lost livelihood. I was also working in the fields, and I was going to the market to sell surplus vegetables.  I had control over the money I earned from that activity, I had money on hand, and I could decide how to spend it.  I could use it for my family, or if some of my neighbors needed money I could loan it out at interest.[xvii]


The IMP compensation package was supposed to be provided to project affected persons for Stage 2, as opposed to the emphasis on project affected families for Stage 1.  In practice, however, the IMP package was only provided to one member of each household, and in almost every case, that family member was male.  As a result, the compensation package reduced by at least half the household livelihoods of Stage 2 people.  Ironically, the IMP report recognized that:


Unfortunately, the female landless agricultural worker is never acknowledged in any policy or scheme of rehabilitation even though she too loses her source of livelihood, on the dependent lands being acquired.  Article 15(3) [of the Indian Constitution] obliges the State to take special provisions for women and children.  Para. 3.1.22 of the draft National Policy [on Rehabilitation and Resettlement] mandates that special programmes designed for women – their education, health, family welfare and employment opportunities must also be instituted as an integral part of the rehabilitation plan.  This aspect has also been kept in mind.[xviii]


Despite this recognition of the traditional disregard for women’s livelihood, the IMP package continued the trend by failing to recognize or compensate for the role of both women and men in household livelihood.   The freedom and independence of women related to their land-based livelihoods was diminished by their displacement, and their economic relationship to the land was neither recognized nor compensated.


The problems inherent in the IMP package were exacerbated by the actions of the NTPC after the package was finalized.  Many people suffered the additional economic and status blow of being discharged from their jobs after their families chose to accept the IMP package.  Both women and men were fired from NTPC jobs after accepting the IMP package, being told by NTPC that since they had accepted the package, they were not entitled to work any longer.  More recently, one woman reported being told by NTPC that “there is no work for women.” When she replied that some women had jobs, she said that the NTPC person responded by saying “that is a past thing, in the future there is no work for women.”[xix]


Finally, although the IMP made explicit provision that the money paid in compensation through the IMP package should be placed in a joint account in the names of both husband and wife, it did not explicitly require that land purchased with the IMP package funds should be registered in the name of both husband and wife.  This was an oversight that NTPC is now exploiting.  In at least one case, a husband and wife sought to register the land that they purchased with the IMP compensation money jointly in both names, only to be told by NTPC that if they did so, NTPC would refuse to pay them the applicable development charges for that land.  NTPC insisted that the land be registered only in the name of the husband.[xx]


This experience, unfortunately, is all too common. As noted in a report prepared for the World Commission on Dams Thematic Review on Social Impacts:


In the context of resettlement, one of the most glaring instances of gender inequality has been the issue of compensation. As men are treated as heads of households, compensation, either cash or land, is invariably awarded to men. Women are not considered to be farmers or house owners. Single women, widowed women are particularly vulnerable in this situation. Similarly, policy often gives land to major sons, but major daughters are excluded from such provisions.[xxi]


The World Bank’s June 2000 Implementation Completion Report (ICR), evaluating the Singrauli project after the loan was closed, noted that:


Men and women have different needs, opportunities, and access to employment or markets in the Indian context, and the project has affected men and women’s livelihood situation differently. . . . Whereas in the past, women had participated in agricultural production activities, employment in NPTC or through other jobs tended to be dominated by men. Many women were relegated to the domestic sphere, and lost much of their traditional influence and involvement in economic decisions and public life. . . .

. . . .An important lesson learned is that targeting and inclusion of more vulnerable groups can be improved through systematic gender analysis. Gender analysis should also be undertaken as part of developing a consultation strategy for a resettlement project, to ensure that both men’s and women’s views and concerns are addressed.[xxii]


The Singrauli ICR concluded that

As of August 1999, at the time of the ICR missions, implementation of the resettlement action plans and other development programs, intended to benefit the local people in the project sites, remained incomplete.  NTPC was not in compliance with the World Bank’s policy on Involuntary Resettlement as summarized in Operational Directive 4.30.[xxiii]


Thus, the Singrauli project has been in violation of the Bank’s policies from before its inception until after its close – in fact, up until the present moment.  The Bank’s supervision has been ineffective at ensuring policy compliance.[xxiv]


III.  Revisions to Resettlement Policy

The resettlement policy has been revised four times since it was first issued in 1980, most recently at the end of 2001.[xxv]  The latest revision was part of an ongoing effort to “reformat” the World Bank’s policies and switch them from binding Operational Directives (i.e. OD 4.30 on Involuntary Resettlement) into a new format called Operational Policies, Bank Procedures, and Good Practices (i.e. OP/BP 4.12 on Involuntary Resettlement).  The OPs and BPs are binding on Bank staff and borrowers, and are actionable through the Inspection Panel, and the GPs are not.


The most recent revisions to the Involuntary Resettlement policy were approved by the Board of Executive Directors in October 2001, following a protracted and controversial five-year revision process.  Only once during that five-year period, in 1999, did the Bank formally solicit broad public input on the draft policy, although at numerous times civil society groups obtained access to leaked versions of the draft and launched their own comment period in response, forcing the Bank to modify some of the proposed changes.[xxvi] 


As described more fully below, the resettlement policy revision process ultimately amounted to a missed opportunity. The Bank ducked civil society efforts to improve the policy and learn from past implementation failures and emerging principles of international law, arguing that it was simply “reformatting” the policy rather than “revising” it, and that suggestions for strengthening the policy were therefore outside the scope of its mandate.[xxvii] 


In taking this approach, the Bank disavowed past practice with regard to revisions to the resettlement policy.  As noted by the former World Bank General Counsel, earlier revisions to the policy resulted in “repeatedly improved versions, [as] the Bank’s resettlement policy took into account the findings of social science research on resettlement and the lessons from the Bank’s own development projects.”[xxviii]  In this case, the Bank ignored lessons from experience, documented over the 11 years between issuance of OD 4.30 and the new OP/BP 4.12. 


Another reason that the process was frustrating was because at the same time that the Resettlement Thematic Group was using the “we’re simply reformatting” argument to avoid strengthening the policy, it was also seeking to introduce new language that would weaken the policy and thereby place vulnerable people, particularly those without nationally recognized claim to land and other assets, at greater risk of impoverishment. 


IV. Overview of Some of the Key Substantive Issues at Stake in the Policy Debate

The revisions to the resettlement policy, as well as some policy changes that were attempted but withdrawn, tended to lessen the protections available to vulnerable people and communities, while promoting checklists and processes that will make it more efficient for development planners to displace people – an emphasis, in other words, on making resettlement easier for the planners, but not necessarily better for those facing displacement.[xxix]  The following summary of some of the key policy debates covers how the Bank avoided clarifications that would have helped promote the welfare of communities facing displacement, while promoting changes that tend to externalize resettlement costs and risks to those least able to bear them.


A. Policy Outcomes that Undermine the Rights of Affected Communities

1.   Direct and Indirect Impacts: The Bank Limits Its Liability

During the revision/reformatting process, the Resettlement Thematic Group introduced new language that will distort the economic and social considerations of the impacts of resettlement project, leading to flawed decision-making and the further externalization of costs to the local affected community. The new version says that the policy:


covers direct economic and social impacts that both result from Bank-assisted investment projects, and are caused by (a) the involuntary taking of land resulting in (i) relocation or loss of shelter; (ii) lost [sic] of assets or access to assets; or (iii) loss of income sources or means of livelihood, whether or not the affected persons must move to another location; or (b) the involuntary restriction on access to legally designated parks and protected areas resulting in adverse impacts on the livelihoods of the displaced persons.[xxx]


This paragraph contains two significant language changes regarding direct/indirect impacts and restricted access to protected areas that cannot be justified as simply part of a reformatting exercise.


According to this new version, the resettlement policy will only apply to the “direct” and not the “indirect” impacts of resettlement or of the project – even if those indirect impacts are completely predictable and clearly linked to the project itself, such as downstream impacts on fish supplies in the context of a dam.[xxxi]  Regarding the indirect impacts of development projects, it has been noted that:


It is often the down-stream populations, ignored in project planning and resettlement and rehabilitation schemes, who are likely to lose their jobs, access to resources, and become ‘environmental refugees’ – that is the people who can no longer secure a livelihood in their traditional homelands because of environmental factors arising out of planned change.[xxxii]


These indirect impacts can have disproportionately negative impacts on women.  For example,


Fishing communities are likely to be most affected by downstream impacts. . . . Women are usually involved in the processing and marketing of the fish catch in fishing communities. As the means of livelihood gets affected, communities are faced with shrinking economic opportunities. Here again, this has resulted in male migration, leaving women to face an increasingly uncertain economic future. In the case of the Akosombo dam in Ghana, fish population in the main river bodies has declined. As fish is the main source of low cost protein, for women this has meant alimentation and nutritional problems. The inundation of forests also resulted in loss of plants and herbs used for traditional medicines.[xxxiii]


Under the new policy, the World Bank will be able to argue that such downstream impacts, even if they are predictable, are not covered by the policy.[xxxiv]  According to a footnote, it would be “good practice” to consider indirect economic and social impacts, but “good practice” is not a binding policy requirement. In response to criticism of this significant language change, the Coordinator of the Resettlement Thematic Group, Maninder Gill, maintained that the Bank was simply “codifying” unwritten Bank practice over the years and “clarifying” the policy.[xxxv]


2.  Restoration v. Improvement:  Stagnation Carries the Day

The most obvious example of where the Bank failed to incorporate lessons learned from implementation experience was in the debate over “restoration” versus “improvement” in displaced persons standards of living.  The original version of the policy in 1980 said that affected people should become “economically self-sustaining in the shortest possible period, at living standards that at least match those before resettlement. Among major factors bearing on the successful achievement of this objective is the choice of the new area, which should, so far as possible, be one in which the existing skills and aptitudes of the involuntary resettlers can be readily employed.”[xxxvi]


The 1990 policy makes clear that those who are being forced to sacrifice their homes, lands and/or livelihoods in the name of development are entitled to receive development benefits from the project that displaces them, and should have their standard of living improved. The 1990 version of the policy said: “The objective of the Bank’s resettlement policy is to ensure that the population displaced by a project receives benefits from it.”[xxxvii]  This clear statement of objective no longer exists in the 2001 version.  The same paragraph further states, with the emphasis in the original, that: resettlement plans should be conceived and executed as development programs, with resettlers provided sufficient investment resources and opportunities to share in project benefits. Displaced persons should be (i) compensated for their losses at full replacement cost prior to the actual move; (ii) assisted with the move and supported during the transition period in the resettlement site; and (iii) assisted in their efforts to improve their former living standards, income earning capacity, and production levels, or at least to restore them.[xxxviii]


Michael Cernea, principle architect of the original Bank policy, wrote in a 1999 World Bank publication that “The primary goal of any involuntary resettlement process is to prevent impoverishment and to improve the livelihood of the resettlers.”[xxxix] 


During the 1999 public comment period on revisions to the resettlement policy, a broad range of voices from outside the Bank, including anthropologists, economists, NGOs, and representatives of implementing agencies, called for the Bank to clarify and strengthen the “improvement” objective by deleting the final clause, “or at least restore them.” These commenters argued that the clause risked undermining the basic policy objective of improvement. 


Anthropologist Ted Scudder, who has served as a World Bank consultant and as a Commissioner on the World Commission on Dams, submitted detailed comments explaining how an emphasis on restoration leads to the perpetuation of poverty at best, and greater impoverishment at worst, all in contrast with the Bank’s overall mandate of poverty alleviation.[xl]  Others argued that the Bank should “abandon the restoration of income standard in favor of a more development-based objective of improving the lifestyles/livelihoods of project-affected people.”[xli]


However, these arguments calling for deletion of the “restoration” language were deflected by the Bank team redrafting the policy.  Maninder Gill, the Coordinator of the Resettlement Thematic Group replied that “we have not been able to accommodate suggestions that would exceed our policy conversion mandate,”[xlii] notwithstanding the fact that the language change would have fallen within the stated ambit of “clarifying” the policy. 


The new policy, regarding objectives, states:

Involuntary resettlement may cause severe long-term hardship, impoverishment, and environmental damage unless appropriate measures are planned and carried out. For these reasons, the overall objectives of the Bank’s policy on involuntary resettlement are the following:

. . . .

(c) Displaced persons should be assisted in their efforts to improve their livelihoods and standards of living or at least to restore them, in real terms, to pre-displacement levels or to levels prevailing prior to the beginning of project implementation, whichever is higher.[xliii]


It is easy to see how this language will perpetuate exactly the stagnation warned against by commenters. The policy objective has been weakened rather than strengthened, and it is therefore even less likely that displaced people will able to achieve the right to development, or that they will benefit from a properly planned project that seeks to improve their standards of living or that reflects their development priorities.  Rather, they are being asked to sacrifice, suffer, and stagnate in the name of development benefits that accrue to others.


3.  Indigenous Peoples: The Bank Introduces Dangerous New Language

During the eleven year interval from when the policy was last revised, there has been a wealth of evidence that involuntary resettlement has imposed disproportionately harsh impacts on indigenous peoples and ethnic minorities. Accordingly, during the policy revision process, many commenters argued that the Bank should bring the resettlement policy into conformity with emerging standards of international law, and the evolving policies of international institutions, by including a provision that required the prior informed consent of indigenous peoples before they can be displaced from their land.  Support for these arguments was drawn from the 1998 version of the Involuntary Resettlement policy of the Inter-American Development Bank (IDB), which provides as follows:

Those indigenous and other low income ethnic minority communities whose identity is based on the territory they have traditionally occupied are particularly vulnerable to the disruptive and impoverishing effects of resettlement.  They often lack formal property rights to the areas on which they depend for their subsistence, and find themselves at a disadvantage in pressing their claims for compensation and rehabilitation.  The Bank will, therefore, only support operations that involve the displacement of indigenous communities or other low income ethnic minority communities if the Bank can ascertain that: (i) the resettlement component will result in direct benefits to the affected community relative to their prior situation; (ii) customary rights will be fully recognized and fairly compensated; (iii) compensation options will include land-based resettlement; and (iv) the people affected have given their informed consent to the resettlement and compensation measures.[xliv]


In addition, support for the informed consent standard is found in the recommendations of the World Commission on Dams, issued in 2000, and in recognitions in both international and national legal systems of the close links between indigenous peoples and their homelands.


Both World Bank board members and outside commenters called for the policy to correct an imbalance in negotiating power by shifting the burden of proof of designing a culturally acceptable project to the project sponsors, by giving the indigenous communities the opportunity to withhold their consent to a project.  Getting their approval would be one of many prerequisites to displacement that affected indigenous communities and ethnic minorities.


However, the Resettlement Thematic Group responded by refusing to adopt the prior informed consent standard (arguing to Board members that it would constitute a “veto right”), and instead inserted dangerous language on indigenous peoples that constitutes a Trojan horse.  The new OP states:


Bank experience has shown that resettlement of indigenous peoples with traditional land-based modes of production is particularly complex and may have significant adverse impacts on their identity and cultural survival.  For this reason, the Bank satisfies itself that the borrower has explored all viable alternative project designs to avoid physical displacement of these groups.  When it is not feasible to avoid such displacement, preference is given to land-based resettlement strategies for these groups that are compatible with their cultural preferences and are prepared in consultation with them.[xlv]


While cynically professing concern about the impacts on indigenous peoples, the new language actually provides no additional protection beyond what is already required in the policy -- i.e. avoidance and striving for land-based alternatives, and is implicitly permissive of the opposite -- i.e. eviction even when cultural survival is threatened and even if a land-based alternative is not offered.


4.  Voluntary v. Involuntary Resettlement

The new policy now “clarifies” that it covers only “involuntary,” and not voluntary, resettlement.  As such, voluntary resettlement financed by the Bank lacks any procedural or substantive policy requirements.  Despite requests from Board members and the public, and despite the recent controversy over “voluntary” and “involuntary” resettlement in the China Western Poverty Reduction Project, the Bank refused to even define what would constitute voluntary resettlement, and failed to provide any policy guidance to cover voluntary resettlement projects.


By failing to provide minimum policy guidance on voluntary resettlement, which is a growing category of projects financed by the World Bank, the draft introduced a perverse incentive for project planners to characterize projects as “voluntary” and thereby avoid the social safeguards that are at the core of the involuntary resettlement policy.  It is easy to conceive of a situation, encouraged by this definition, whereby project-affected people are required by the authorities to give their “consent” to resettlement in order to receive benefits. Their “consent” would, in turn, remove their rights to protection under World Bank policy. The OP does not explain how to determine whether a decision is, in fact, voluntary and free from coercion.[xlvi]


The danger of this approach was demonstrated by the China Western Poverty Reduction Project, rejected by the Board in July 2000, which claimed to promote poverty alleviation through the “voluntary” resettlement of approximately 58,000 settlers who would be shifted onto the Tibetan plateau where they would involuntarily displace approximately 10,000 local people, including Tibetan and Mongol ethnic minorities.  The project’s “voluntary resettlement implementation plan” was never released to the public, because it was considered to be the property of the borrowing government, and China refused to allow its disclosure. Thus, according to the Bank, voluntary resettlement plans are not required to be made publicly available, and voluntary settlers affected by World Bank-financed projects are not entitled to the same participation rights or substantive policy protections as involuntary settlers.


Had the Bank provided a minimum framework of rights for situations involving voluntary resettlement, it would have constituted an important first step towards moving away from the conflict-ridden model of forcible eviction and towards a model of negotiated settlements with local communities. The Bank’s failure to address voluntary resettlement constituted a missed opportunity and allowed the perpetuation of a status quo that has negative impacts on local communities. 


B.  Hazards That Were Avoided by Civil Society Vigilance

1.  Pejorative Language Regarding Illegal Users of Natural Resources Removed

As the policy was being revised, Bank staff sought to introduce new language that would have allowed the borrowers to develop a “participatory process” to exclude from compensation or project planning any persons engaged in the “illegal” use of natural resources.  Thus, footnote 14 of the March 2001 draft stated that  “The participatory process may result in excluding from assistance persons engaged in illegal use of natural resources . .  .”[xlvii] leaving it to the discretion of implementing agencies to determine who or what was illegal.  As discussed more fully in the following section, this proposed language was deleted in response to widespread public criticism.


2.  Recognition of the Vulnerability of Those Who Lack Title to Land

Earlier drafts of the OP also sought to delete the recognition of vulnerability of people who are not covered by national land compensation legislation, while also simultaneously revising one of the only paragraphs in the policy that actually referred to women.  Bear in mind that the 1990 version of the policy stated that:


Vulnerable groups at particular risk are indigenous peoples, the landless and semilandless, and households headed by females who, though displaced, may not be protected through national land compensation legislation.  The resettlement plan must include land allocation or culturally acceptable alternative income-earning strategies to protect the livelihood of these people.


In contrast, the March 6, 2001 version of the Draft OP deleted the reference to those who are excluded from national land compensation laws, and stated: “To achieve the objectives of this policy, particular attention is paid to the needs of vulnerable groups among those displaced: especially those below the poverty line, the landless, the elderly, women and children, indigenous peoples and ethnic minorities.”


After public outrage over the proposed changes resulted in the Bank being inundated with thousands of letters, phone calls, and emails, President Wolfensohn responded by insisting that the policy would not be weakened.  In a speech to the German Parliament he said: “The Bank has the most comprehensive environmental and social safeguard policies of any of the multilateral development banks.  We have a leadership role to play in this area and we intend to play it.  Make no mistake, we will not water down our safeguards as some have alleged.”[xlviii] This edict was reiterated down the chain of command.  Ian Johnson, the Vice President for Environmentally and Socially Sustainable Development (ESSD), stated the following in a letter to CIEL:


We have consistently held that the policy conversion process does include the incorporation of important policy clarifications provided by Bank Management and takes into account important recommendations of OED evaluations and lessons of implementation.  However, since the conversion process is not supposed to alter the key objectives and principles of the current resettlement policy, it is not possible to accommodate suggestions that entail a change in current standards.  At the same time, we have taken great care to ensure that the current standards enshrined in safeguard policies are not, in any way, diluted. [xlix]


In the next internal draft (which was promptly leaked to NGOs), the offensive footnote 14 regarding “illegal users of natural resources” was deleted, and the language about the vulnerability of those lacking title to land was restored to paragraph 8.  The final version of the OP, in paragraph 8, has the following formulation regarding the vulnerability of those lacking title:


To achieve the objectives of this policy, particular attention is paid to the needs of vulnerable groups among those displaced, especially those below the poverty line, the landless, the elderly, women and children, indigenous peoples, ethnic minorities or other displaced persons who may not be protected through national land compensation legislation.


From a gender perspective, the final OP language is arguably better in terms of the vulnerability of women, because whereas the OD recognized the vulnerability of “households headed by females,” the new language in the OP recognizes more broadly the vulnerability of “women and children” as among others who, though displaced, may not be protected by national land compensation legislation.


C. Other Policy Changes of Note.

This paper cannot cover all the many ways that OP/BP 4.12 differs from OD 4.30; the interested reader is encouraged to do their own analysis, and to review the archived documents and correspondence, many of which are available on the website of the Center for International Environmental Law.[l]  This section will cover a few significant points that have not yet been addressed: the issue of restricted access to protected areas; language regarding counterpart financing; and the Bank’s involvement until resettlement is completed.


(1)     Parks and Protected Areas.

One significant change to the resettlement policy was the adoption of new provisions relating to restrictions on access to parks and protected areas resulting in adverse impacts on the livelihoods of the people who are thereby displaced.  People affected by restrictions on access to protected areas are given differential treatment throughout the policy.[li]  Paragraph 3(b) of the 2001 policy carved out this category of impacts, and Paragraph 7 calls for a “process framework” (defined in Annex A) for these projects.  Through this process framework, project planners develop a plan for implementation, measures to improve or at least restore livelihoods (while maintaining the sustainability of the protected area), and procedures for conflict resolution.


Responding to public concerns about this differential treatment, the Coordinator of the Resettlement Thematic Group insisted that “the objectives of the resettlement program remain the same for those affected by restrictions of access as for those affected by land taking in large infrastructure projects.”  He further stated that “the mitigation plans prepared as a result of this participatory process need to be reviewed and accepted by the Bank and restrictions on access cannot be imposed until the mitigation measures described in the mitigation plan are implemented.”[lii]  This assurance, however, is not reflected in the policy.  Instead, paragraph 31 of the OP states that “during project implementation, and before to [sic] enforcing of the restriction, the borrower prepares a plan of action, acceptable to the Bank, describing the specific measures to be undertaken to assist the displaced persons and the arrangements for their implementation.”  A plan of action is not at all the same thing as implementation of mitigation measures, as history has made all too clear.  These provisions should be monitored carefully and reviewed in the context of the promised two-year review of the policy.


(2)     “Operational” Language Improvements

Certain civil society comments that related to addressing gaps in the borrowers’ legal frameworks and financial responsibilities were incorporated into the policy.  For example, BP paragraph 2(c) now requires Bank staff to “assess the legal framework covering resettlement and the policies of the government and implementing agencies (identifying any inconsistencies between such policies and the Bank’s policy),” and paragraph 2(e) further requires Bank staff to discuss with the borrowing government agencies “measures to address any inconsistencies between government or implementing agency policies and Bank policy.” 


The issue of counterpart financing is an important “operational” issue where some progress has been made in the 2001 revisions.  In some cases, implementation problems have related to the Bank’s insistence that resettlement financing, rather than being built into the loan financing, should come through government-provided “counterpart” funds, combined with inadequate attention to the adequacy of those funds to meet the costs of displacement.  In the case of the Yacyretá dam on the Rio Parana between Argentina and Paraguay, for example, most of the benefits were to accrue to Argentina (electric power for Buenes Aires), which was the borrower for the loan, but most of the most of the displacement (and environmental) impacts would be felt in Paraguay.  Despite obvious problems concerning the political will of Argentina to come up with funding for poor Paraguayans, the deal called for Argentina to provide $300 million in counterpart financing.  Subsequently, Argentina refused to deliver on its commitment, with apparent impunity.[liii] 


In the revisions to the policy, several clauses were added to require Bank staff to assess the feasibility of the resettlement measures during project preparation, including provision of counterpart funding on an annual basis.[liv]  During appraisal, Bank staff must assess the borrower’s “commitment to and capacity for implementing the resettlement instrument,” including the “availability of adequate counterpart funds for resettlement activities.” [lv]  The Project Appraisal Document must summarize basic information about resettlement, “including adequate and timely provision of counterpart funds.”[lvi]


(3)     Clear Statement about Bank Supervision.

Another new provision worth noting is the explicit statement in BP paragraph 16 that “A project is not considered complete - and Bank supervision continues - until the resettlement measures set out in the relevant resettlement instrument have been implemented.”[lvii]  While it is valuable for the policy to explicitly state that the Bank’s role continues until the resettlement measures have been implemented, it must be noted that this language cuts both ways.  First, there is an improper emphasis on the measures in the resettlement instrument (the check-list approach) rather than an evaluation of the results on the ground, and gauging compliance with the objectives of the policy. 


In addition, there is contradiction between how the Bank considers a project to be complete – the above language needs to be contrasted with the (weak and deferential) language of OP paragraph 24, which says that if the borrower determines that the objectives of the resettlement instrument have not been achieved as of the completion of the project, “the borrower should propose follow-up measures that may serve as the basis for continued Bank supervision, as the Bank deems appropriate.” These provisions are described rather optimistically by Maninder Gill as follows: “If standards of living are not likely to be restored, the policy proposes that the borrower should propose a course of action to address this issue.  Thus, the proposed policy attempts to strengthen the likelihood of restoration of standards of living of displaced persons.”[lviii]


V. Overview of the Gender Implications of the Policy Revisions

Given that revisions to the policy were taking place at and after the turn of the 21st century, it is surprising that the revised policy doesn’t have much to say about the gender differentials in resettlement planning or impact.  In fact, the word “gender” does not appear anywhere in the policy, and the OP mentions women only once.  The BP has no reference to women, and the Annex has only one. Nowhere is there even a citation to the Bank’s Operational Policy 4.20, The Gender Dimension of Development (October 1999).


The 1990 version of the policy emphasized the importance of including women and other vulnerable groups in institutional arrangements regarding the development of resettlement plans, the design and choice of resettlement alternatives, and methods of facilitating communication between project officials and communities.  It said: “Particular attention must be given to ensure that such vulnerable groups as indigenous people, ethnic minorities, the landless, and women are represented adequately in such arrangements.”  That language has been included in the Annex to OP 4.12, in paragraph 15(d), though without the emphasis on “particular attention,” and instead only requiring a description of institutional arrangements for communication between affected people and project authorities.


Once again, the Resettlement Thematic Group ignored the lessons of experience, including the findings of the comprehensive 1994 Bankwide Review of resettlement, which noted that:


Women may experience the adverse consequences of resettlement more strongly than men.  First and foremost, this is because compensation payments are usually paid only to the heads of households, converting the collective assets of the family to cash in male hands, and leaving women and children at higher risk of deprivation.  Female-headed households, which in some cases range from 20-40 percent of the affected household, suffer most from such exclusionary policies. . . .In urban areas. . . there is evidence that displaced women are harder hit by resettlement than men. . . .Women may also be affected disproportionately in rural areas since they are often more dependent on common property resources.[lix]


Despite the assurances in the letter from Ian Johnson about taking into account “important recommendations of OED evaluations and lessons of implementation,” the Resettlement Thematic Group failed to deal with the OED’s warning that the Bank seemed to be “largely oblivious of the gender aspects of resettlement,” a conclusion that was drawn from a 1998 OED review of resettlement in six dam projects.[lx]  In its assessment of resettlement in the Upper Krishna dam project in India, the OED, after noting that “the needs of women as a group have been completely ignored” by the project, went on to find that:


Most women (62 percent) thought they now had less personal disposable income than in the old village; only 11 percent felt they had more. This has significance beyond the  income loss alone, as the decisionmaking power of a woman is closely tied to her personal disposable income. Labor opportunities for women were reduced. Their income from farming and livestock decreased and they became more dependent on wage incomes. The availability of fuelwood and fodder decreased. Livestock had to be sold. Women had to migrate for work. . . . .Not only was compensation considered inadequate or squandered but prices increased significantly while self-sufficiency decreased. For example, vegetables were more scarce and had to be bought in towns at high cost.. . . .Women had to bear much of the burden of the decline in income and living standards during the difficult first few years after relocation.[lxi]


Such findings are also echoed in the comprehensive report of the World Commission on Dams, which found that "Forests, fisheries and other common property resources, which support subsistence livelihoods, are often not replaced during resettlement with women often bearing a disproportionate share of the resulting costs."[lxii] It also found that “[f]or both indigenous and non-indigenous communities, studies show that displacement has disproportionately impacted on women and children.”[lxiii]  In addition, there are inequities in how impacts are borne, with experience showing that “those with assets and power prior to displacement cornered most of the benefits while most of those belonging to poorer communities were left with unproductive lands and a few temporary jobs.”[lxiv]


In the most recent policy revision process, the World Bank took no proactive steps to address issues of gender or to try to improve the status or rights of women in the context of displacement.  In fact, the net effect of the policy revisions is to reduce rather than increase attention to women in the context of displacement.  The requirement for the provision of livelihood linked specifically to displaced vulnerable people has been eliminated.  The requirement that women and other vulnerable groups be, as a matter of priority, included in consultations has been de-emphasized and shifted to the OP Annex. 


As a result, the Bank’s revised resettlement policy makes zero progress on gender issues.  One could argue that in the 21st century it is unconscionable for the World Bank to continue to be blind to gender implications of its lending decisions, given that being blind to the obviously negative implications for women (and their families) results in the perpetuation of power in the hands of those with privilege.[lxv]


VI. Policy Implementation: The Bank/Borrower Relationship

When considering the revisions to the resettlement policy, it is also important to bear in mind some of the historic tensions between the policy prescriptions on paper and implementation in practice. The World Bank’s studies, independent critiques, and numerous claims to the Inspection Panel have all shown that there is a disturbingly large gap between policy and practices when it comes to involuntary resettlement.  There is tension between the Bank and many borrowing governments over implementation of not just the resettlement policy but most of the environmental and social “safeguard” policies.  As noted in a recent study:

The huge scale of development-induced displacement, and its overwhelmingly negative consequences for affected populations, indicate major policy failures in this area …Explanations advanced for DIDR’s [development induced displacement’s] dismal record include the absence of national level legal and policy frameworks, as well as the lack of the necessary commitment and political will, on the part of borrower countries implementing DIDR projects.[lxvi]


It is clearly the case that in many borrowing countries, the legal framework of rights for people affected by the government taking of land is not as strong as the World Bank’s policy.  In such contexts, the Bank’s policy complements, and to the extent necessary, goes beyond the law and policy requirements of the borrowing country.  In other words, the Bank’s policy sets the minimum requirements.  This can lead to tension with borrowing countries, which frequently resent being asked to provide benefits in excess of national law, in part because doing so is seen as too expensive, and in part because they are concerned about setting a precedent that can lead to increased demands for compensation from people affected by projects which are not financed by the Bank.  In many cases, the Bank seems to lack the capacity, political will, and necessary clout when it comes to dealing with implementation problems in the context of resettlement; this seems to be particularly true when it is a large and powerful borrowing country involved. 


As described by the Bank itself in the Implementation Completion Report for the Singrauli project:

It is difficult for the Bank to try to secure compliance with its policies that require higher standards than those which are provided under state or national regulations. Bank standards are seen by the implementing agency as creating precedents for future projects, which may not be funded by the Bank, as well as future financial liabilities. This should be recognized by all parties and agreement reached on the appropriate action in the earliest stages of the project. Projects with complex R&R and environmental issues should not be undertaken in a social setting as complex as India without the full commitment of all concerned.[lxvii]


Indeed, the lack of borrower buy-in puts more responsibility on the World Bank, which should work to do capacity building, to promote the development of stronger national legislation in borrowing countries to eliminate the policy gap with the World Bank, and help promote informed decision-making and equitable and accessible processes for affected persons. In situations where there is tension between Bank policy and local or national law, the Bank should take greater responsibility to ensure that particular projects do not result in the impoverishment of local people.   Furthermore, if the capacity and commitment to meet the terms of the World Bank’s policy framework are lacking, the Bank should not support the project.


However, this is not a role that the Bank has readily assumed, and it has consistently given support to projects that fail to meet its policy requirements.  Policy and implementation failures are not solely the province of borrowing countries, though of course that is part of the picture.  As noted in the two examples discussed above, resettlement failures can also be linked to substantive and procedural design failures that are arise when Bank staff have insufficient concern for the requirements of the involuntary resettlement policy and other social and environmental policies.  These policy failures can lead to the Bank designing and approving flawed projects that fail to consider or factor into decision-making the consequences for vulnerable people, with devastating effects at the project level.  Problems in implementation can often be traced back to initial project design.


The Bank, when challenged on its record of failed resettlement, for example in the context of claims to the Inspection Panel, invariably seeks to cast at least some of the blame on borrowing country governments.  In addition, there is currently momentum within the Bank to try to shift more responsibility for safeguard policy implementation to the borrowers, and to devolve accountability away from the Bank.[lxviii] This is surely at least in part an attempt to avoid the jurisdiction of the Inspection Panel, which has jurisdiction over the actions or omissions of the Bank but not the borrower, except to the extent that the Bank has failed to adequately supervise or monitor the borrower’s implementation of the loan agreements and compliance with the Bank’s policy framework. If more responsibility is shifted to the borrower, it will be more difficult for local people to seek accountability through the Inspection Panel. There is no evidence that this is a sound development strategy; it seems more likely to be an avoidance of accountability rather than a way of improving implementation and accomplishing sustainable and equitable development.


The shifting of blame between the Bank and the borrower has been going on for years, and is starting to resemble a cynical game that is played, with the lives, livelihoods and futures of affected people at stake.  The money is disbursed, the loan is closed, and yet the problems continue.  When people complain, the Bank and borrower each point their fingers at the other. Even in cases when the Inspection Panel or the Environment Department or the Operations Evaluation Department documents case-specific and/or systemic violations of the resettlement policy, very little emphasis is put on solving the implementation problems and bringing a project into compliance with the policy.  Even though the loan conditions and policies in force at the time of project inception continue to legally apply until the loan is repaid, the Bank fails to supervise the projects or to require that borrowers meet the terms and conditions under which the money was made available.[lxix]  This lack of follow-through renders the policy virtually meaningless during the majority of a project’s lifespan.  The Bank needs to undertake a renewed assessment of its role in supervision and monitoring of projects involving involuntary resettlement and work with borrowers to ensure that project implementation not only respects the loan agreements and policy framework, but also that implementation happens in a way that respects the rights and advances the interests of the people most affected by the project.

The Bank should invest resources in developing capacity in borrowing-country implementing agencies, and provide technical support and real supervision and assistance during project implementation in order to ensure that the objectives of the policies that affect human and environmental rights are respected.


VII. Conclusion

The Bank should not lend its financial and institutional support to projects that result in the involuntarily impoverishment of communities that are deemed to be “in the way” of development projects. This is inconsistent with its mandate of poverty alleviation and engenders social tension. 


The World Bank and borrowers have been shown to lack capacity in terms of policy compliance in project design and implementation. There should be a moratorium on all new projects involving involuntary resettlement until the Bank deals with the problems it has already created, increases its capacity, and focuses on improving the capacity of and its own commitment to borrowing governments and implementing agencies.[lxx] Otherwise, the World Bank will continue to practice abusive development, in violation of its mandate of poverty alleviation, its policies, and international human rights law. 


The Bank must undertake a shift in emphasis, and start prioritizing those who are not traditionally privileged.  The borrowers can generally be expected to be responsive to the concerns of the elites, those with political power. The World Bank should take a different approach and be more responsive to the needs, concerns and priorities of poor and disenfranchised affected communities.  This includes those groups lacking societal privilege, such as women and ethnic minorities.  In addition to taking a more appropriate approach to future projects involving involuntary resettlement, the Bank must take responsibility for the impacts of policy failures to date, including developing a system for bringing projects that are already in implementation into compliance with the terms of the policies in force at the time that the loans were approved.


The resettlement policy is supposed to be reviewed again two years after its adoption, which should no later than January 2004.  This time, the Bank should recognize the unrecognized: women, indigenous people, the faceless millions of displaced people.  The Bank should help to facilitate a shift in power and put more of the responsibility for decision-making in the hands of affected communities.  In addition, the policy blinders to gender considerations should be removed.  The WCD has recommended that:

The involvement of women and other vulnerable groups in decision-making should be ensured at all stages of the planning and development process.  There should be clear consideration for the vulnerabilities that expose women to project impacts (displacement, changes in the resource base and resulting disruptions of social and economic resources and networks) and for the specific obstacles that reduce their opportunities to share benefits generated by the project.[lxxi]


The report of the World Commission on Dams was deliberately ignored by the Resettlement Thematic Group, even though the report was issued before the revisions to the policy were finalized.  In future reviews of the policy, WCD findings relating to resettlement, gender, and indigenous peoples should be evaluated and incorporated into revised policy.


In addition, the Bank should take steps to abandon the failed model of forcible eviction and inadequate compensation, and shift instead towards a model of negotiated settlements with affected communities. The new model should be based on respect for local knowledge, local development priorities, and true engagement and participation of affected people, where those who are affected by the project would have to agree to and help shape the terms of any proposed displacement and rehabilitation.

[i] See, e.g., World Bank, Resettlement and Development: The Bankwide Review of Projects Involving Involuntary Resettlement 1986-1993 (1994) (hereinafter the “Bankwide Review”); World Bank Environment Department, Regional Remedial Action Planning for Involuntary Resettlement in World Bank Supported Projects (1995); World Bank Operations Evaluation Department, Recent Experience With Involuntary Resettlement (1998).

[ii] See, e.g. World Bank Inspection Panel, Report and Recommendation, Argentina/Paraguay Yacyretá Hydroelectric Project (1996); World Bank Inspection Panel, Report and Recommendation, Brazil: Itaparica Resettlement and Irrigation Project (1997); World Bank Inspection Panel, Report and Recommendation, India: NTPC Power Generation Project (1997), available at; Anthony Oliver-Smith, Displacement, Resistance and the Critique of Development (Refugees Studies Center, University of Oxford, July 2001). See also The World Commission on Dams, Dams and Development (Earthscan, November 2000), particularly pages 102-116.

[iii] See, e.g., Bankwide Review at x, which found that overall, “projects appear not to have succeeded in reestablishing resettlers at a better or equal living standard and that unsatisfactory performance still persists on a wide scale.”   In addition, a thorough review of all projects in the portfolio undertaken in 1992 revealed that the World Bank was suffering from a general performance crisis, with 37.5% of projects ranked as “unsatisfactory,” and widespread violations of loan agreements. World Bank, Report of the Portfolio Management Task Force (World Bank, 1992).  The document made clear that the policy violations identified by the Morse Commission were not an aberration, but rather reflected a systemic part of Bank culture.

[iv] The new policy (OP/BP 4.12) applies to projects that are initiated after January 1, 2002. OD 4.30 continues to apply to all projects that were approved from October 1990 through the end of 2001. Operational Manual Statement 2.33 was adopted in February 1980 and applies to projects approved after that time.

[v] Sardar Sarovar, Report of the Independent Review (1992) at xv-xvi (emphasis added) (hereinafter “Independent Review”).

[vi] The Morse Commission report noted that “the problems besetting the Sardar Sarovar Projects are more the rule than the exception to resettlement operations supported by the Bank in India. . . .Projects are appraised and negotiated by the Bank despite the absence of resettlement plans, budgets, and timetables that meet the Bank’s resettlement policy.” Independent Review at 53-54.

[vii] Wade, Robert, 1997. ‘Greening the Bank: The Struggle over the Environment, 1970 – 1995,’ in Devesh Kapur, John P. Lewis, Richard Webb (eds.), The World Bank: Its First Half Century, p. 705.  Washington, DC: Brookings Institution.

[viii] For more information about the Inspection Panel, see See also Dana L. Clark, A Citizen’s Guide to the World Bank Inspection Panel, 2d ed. (CIEL, 1999), available for download in English, Spanish, French and Portuguese at  A book evaluating the case history of the Inspection Panel is now available; see Dana Clark, Jonathan Fox, and Kay Treakle (eds.) Demanding Accountability: Civil Society Claims and the World Bank Inspection Panel (Rowman & Littlefield: 2003).

[ix] This requirement effectively meant that the Sardar Sarovar project could not be brought to the Inspection Panel.

[x] National Thermal Power Corporation, ‘Environmental Study of Singrauli Area,’ (performed by Electricité de France International), p. 52 (July 1991).

[xi] As a result of financial irregularities, the Bank chose not to support the Rihand expansion and the money was diverted to the Kayamkulam project in Kerala.  The World Bank, Implementation Completion Report for the NTPC Power Generation Project.

[xii] Inspection Panel, Report on Desk Investigation, Dec. 23, 1997, para. 62.


[xiii] Inspection Panel, Report on Desk Investigation, Dec. 23, 1997, para. 84; see also para. 65.  The government of India was in no apparent rush for the money, and did not sign the loan agreement until December 1993, and there were no disbursements until March 1995. Id., para. 65. 

[xiv] World Bank, Executive Directors’ Meeting June 29, 1993, INDIA: NTPC Power Generation Project (R93-116), Statement by Mr. Fischer, p. 3. 

[xv] See, World Bank, Implementation Completion Report, India: NTPC Power Generation Project (Loan 3632-IN) (June 14, 2000), footnote 21: “There were strong complaints from the Indian Government against what they saw as an investigation into internal affairs, and they argued that the Inspection Panel’s mandate was only to investigate the Bank’s own conduct, not the actions of the Indian Government or its agencies. Since the complaint against the Bank was based on the argument that NTPC did not adequately implement the provisions of the project resettlement plans, and that the Bank failed to take adequate action to improve matters, it would be difficult to conduct a full investigation without also pointing out shortcomings of the Borrower.”

[xvi] Inspection Panel, Report on Desk Investigation, Dec. 22, 1997.

[xvii] Dashmatia, interviewed by Dana Clark, Dodhar resettlement colony, September 8, 2002.  The loss of access to independent income over which women had control, from the sale of excess vegetables and milk, was confirmed by Pulmati, interviewed by Dana Clark, Dodhar resettlement colony by Dana Clark, September 8, 2002.

[xviii] Independent Monitoring Panel, Final Report, p. 43.

[xix] Basanti, interviewed by Dana Clark, Bijpur resettlement colony, September 7, 2002.

[xx] Ram Narain Kumari, interviewed by Dana Clark, Dodhar resettlement colony, September 8, 2002.

[xxi] Lyla Mehta and Bina Srinivasan, Balancing Pains and Gains: A Perspective Paper on Gender and Large Dams, Report Prepared for the World Commission on Dams Thematic Review: Social Impacts, p. 31 (November 2000).

[xxii]World Bank, Implementation Completion Report, India: NTPC Power Generation Project (Loan 3632-IN) (June 14, 2000), p. 83.

[xxiii] Id., p. 74.

[xxiv] Inspection Panel, Report of Desk Investigation, December 22, 1997, para. 109.  “The Panel found no evidence of any Bank actions to ensure that required RAP measures were implemented prior to shifting people.  Management recognized problems in the effectiveness of its project supervision, as reflected in difficulties in having appropriate remedial action taken to overcome project implementation problems.  Had the supervision been effective, some of the harm suffered in the project area may have been avoided.”

[xxv] The former policy was known as Operational Directive (OD) 4.30, and the new version is known as Operational Policy (OP) and Bank Procedure (BP) 4.12.  The Bank first adopted a policy on involuntary resettlement in 1980, and the policy was subsequently modified in 1986, 1988, 1990, and 2001.

[xxvi] There was also a limited public comment process in 1997, when approximately 50 NGOs were invited to comment.   See, e.g., letter from Dana Clark, CIEL to Dan Aronson, World Bank, April 7, 1997.  In 1999, the draft policy was posted on the Bank’s external website for a six-month comment period.  It was translated into 11 languages, and the Bank held consultations in 14 countries.  The public comment process largely failed to include communities that had suffered from development-induced displacement, but it did result in a significant broadening in the flow of information.  However, after the 1999 public comment period, the draft policy disappeared from view.  Considerable effort was expended by NGOs to force the Bank to abide by a commitment that it had made to provide a summary of responses to the feedback that it had received; this matrix was finally released in 2001.  Also in 2001, at least three more versions of the policy were leaked and subjected to public comment, despite the Bank’s desire to keep the process closed to the public.

[xxvii] See, e.g, letter from Ian Johnson, World Bank, to Dana Clark, CIEL, April 24, 2000:  “I should point out that it was not our intent to either strengthen or weaken the resettlement and rehabilitation objectives of the current policy, OD 4.30.  Rather, this exercise is rightly referred to as a ‘conversion’ of existing policy, because its aim has been to put the original OD, together with relevant OPC decisions and OED recommendations from the past few years, into a format that separates basic policy objectives from Bank-specific procedures and non-mandatory good practice.”  See also, Dana Clark, “World Bank Resettlement Policy Compromised,” 17 World Rivers Review, No. 1, p. 10-11 (Feb. 2002).

[xxviii] Ibrahim F.I. Shihata, ‘Legal Aspects of Involuntary Population Displacement,’ in Antropological Approaches to Involuntary Resettlement: Policy, Practice, and Theory, Michael M. Cernea and Scott E. Guggenheim (eds.); Boulder, Colorado: Westview Press (1991).

[xxix] Clark, World Rivers Review, at 10.


[xxx] World Bank, Operational Policy 4.12, para. 3 (December 2001).

[xxxi]  World Bank, Conversion of the World Bank’s Policy on Involuntary Resettlement, Substantive Comments from Executive Directors on Draft OP/BP 4.12, dated July 25, 2000 (March 6, 2001), comment no. 29, pp. 6-7, available at

[xxxii] Christopher McDowell, Introduction, Understanding Impoverishment: The Consequences of Development-Induced Displacement, p. 6 (Christopher McDowell, ed., 1996).

[xxxiii] Lyla Mehta and Bina Srinivasan, Balancing Pains and Gains: A Perspective Paper on Gender and Large Dams, Report Prepared for the World Commission on Dams Thematic Review: Social Impacts, p. 19. (November 2000).

[xxxiv] See World Bank matrix of responses to comments from Board members, specifically the response to a question posed by the US Executive Director regarding the downstream impacts of a dam project; the matrix of responses is available at

[xxxv] Personal communications with the author; see also World Bank, “The World Bank and Social Safeguard Policies,” question 11 (March 2001): “The draft OP/BP merely clarifies a long-standing interpretation of Management under the existing OD 4.30 – that the Bank’s policy on involuntary resettlement covers only direct impacts resulting from Bank-assisted projects and caused by land takings or restrictions of access to protected areas.”

[xxxvi] The World Bank, Operational Manual Statement 2.33, Social Issues Associated With Involuntary Resettlement in Bank-Financed Projects, para. 18 (February 1980).

[xxxvii] The World Bank, Operational Directive 4.30, Involuntary Resettlement, para. 3 (June 1990).

[xxxviii] Id.

[xxxix] World Bank, The Economics of Involuntary Resettlement (Michael Cernea, ed., 1999), p. 6.

[xl] See, e.g., Thayer Scudder, “Critique of Resettlement Draft OP/BP 4.12,” (July 28, 1999). Mr. Scudder noted that “acceptance of restoration undercuts the Bank’s emphasis on the resettlement component as a development project.”

[xli] See, e.g., letter from Center for International Environmental Law and International Rivers Network, on behalf of 125 organizations and 33 individuals, to President James Wolfensohn, World Bank, February 22, 2001.  Indeed, the Bank claims to support the “right to development,” a right that has been defined as "an inalienable human right by virtue of which every human person and all peoples are entitled to participate in, contribute to, and enjoy economic, social, cultural and political development, in which all human rights and fundamental freedoms can be fully realized."  Declaration on the Right to Development, G.A. Res. 41/128, U.N. GAOR, 41st Sess., 97th plen. mtg., Annex, art. I, U.N. Doc. A/RES/41/128 (1986).

[xlii] Memo from Maninder Gill, World Bank, to Dana Clark, sent via email August 19, 2001 (available at

[xliii] The World Bank, Operational Policy 4.12, Involuntary Resettlement, (December 2001).

[xliv] Inter-American Development Bank, Resettlement Policy OP 710 (1998).

[xlv] The World Bank, Operational Policy (OP) 4.12, para. 9 (2001) (emphasis added).

[xlvi] The policy defines “involuntary” as follows: “For purposes of this policy ‘involuntary’ means actions that may be taken without the displaced person’s informed consent or power of choice.” Id., footnote 7.

[xlvii] See footnote 14 of the March 6, 2001 draft of the policy, which is on the website of the Center for International Environmental Law,

[xlviii] James D. Wolfensohn, “The Challenges of Globalization: The Role of the World Bank,” Speech made to German Parliament, Berlin, April 2, 2001.

[xlix] Ian Johnson, World Bank letter to Dana Clark, CIEL (April 4, 2001), available at

[l] See (last visited April 17, 2003).

[li] See, e.g., OP 4.12 paragraphs 3(b), 7, 10, 13, 17(c), 25, 30, 31, and Annex A, paragraphs 26, 27.

[lii] Letter from Maninder Gill to Juliette Majot, International Rivers Network, 12 July 2001.

[liii] See Kay Treakle and Elías Díaz Peña, 2003.  ‘Accountability at the World Bank: What Does it Take? Lessons from the Yacyretá Hydroelectric Project, Argentina/Paraguay,’ in Dana Clark, Jonathan Fox and Kay Treakle (eds.), Demanding Accountability: Civil Society Claims and the World Bank Inspection Panel, p. 69-91. Lanham, Md: Rowman & Littlefield Publishers.

[liv] See BP para. 5(d).

[lv] BP para. 10.

[lvi] BP para. 11.

[lvii] BP para. 16.

[lviii] Letter from Maninder Gill, World Bank to Juliette Majot, International Rivers Network, 13 July 2001.

[lix] The World Bank, Resettlement and Development: The Bankwide Review of Projects Involving Involuntary Resettlement 1986-1993 (1994), at P. 2/9.

[lx] World Bank Operations Evaluation Department, Recent Experience With Involuntary Resettlement: Overview, p. 60 (1998).

[lxi] World Bank Operations Evaluation Department, Recent Experience With Involuntary Resettlement: India – Upper Krishna (Karnataka and Maharashtra) (1998), p. 20-21.

[lxii] World Commission on Dams, Dams and Development, at 114 (2000).

[lxiii] Leopoldo Jose Bartolome, et al., Displacement, resettlement, rehabilitation, reparation and Development 6 (WCD Thematic Review 1.3, Working Paper, 2000).

[lxiv] Smitu Kothari, “Whose Nation? The Displaced as Victims of Development,” Economic and Political Weekly, p. 1476-1485 (June 1996).

[lxv]It is interesting, perhaps, to note that the Board of Governors of the World Bank is 94.5% men and 5.5% women, and that the Board of Directors of the World Bank is 91.7% men and 8.3% women. Gender breakdown of top posts at the world’s financial institutions, prepared by Women’s Environment and Development Organization, 2002, available at

[lxvi] Alan Rew, Eleanor Fisher and Balaji Bandey, “Addressing Policy Constraints and Improving Outcomes in DIDR Projects,” (2000).

[lxvii] World Bank, Implementation Completion Report, India: NTPC Power Generation Project (Loan 3632-IN) (June 14, 2000), p. vi, 20.

[lxviii]  World Bank, “Safeguard Policies: Framework for Improving Development Effectiveness,” (October 7, 2002), available at

[lxix] Michael Carter, the World Bank’s Country Director for India, acknowledged that a “vast minority” of projects are supervised by the Bank after the funds are disbursed.  Personal communication, meeting with the author, August 25, 2003.

[lxx] Some of these ideas regarding increased accountability for problems and the development of greater capacity for handling resettlement and other social and environmental issues are addressed in a recent article by the author.  For more information, see Dana L. Clark, “The World Bank and Human Rights: The Need for Greater Accountability” in 15 Harvard Human Rights Journal, p. 205-226 (Spring 2002).

[lxxi] World Commission on Dams, Dams and Development, at 216.