Leaked Report Reveals Major Problems with Ilisu Resettlement
An Ilisu Dam Campaign Briefing on the ‘Ilisu Dam’s Resettlement Action Plan (RAP) -- Achieving International Best Practice’

by The Ilisu Dam Campaign and The Corner House

first published 6 September 2000

International groups campaigning against the controversial Ilisu Dam in Turkey obtained a copy of an assessment, commissioned by the export credit agencies considering financial support for the project, of the Turkish Government’s proposed resettlement plan. The assessment highlights serious problems with resettlement and reveals that two to three times more people may be affected than previously estimated - possibly as many as 70,000 people, mainly ethnic Kurds.This article analyses the assessment


International groups campaigning against the controversial Ilisu Dam in Turkey, which the UK government is provisionally supporting through its Export Credit Guarantee Department (ECGD), have obtained a copy of an internal assessment of the Turkish Government's proposed resettlement plan. The assessment, written by ex-World Bank official Dr. Ayse Kudat, highlights serious problems with resettlement and reveals that the numbers who may be affected are 2-3 times higher than previously estimated - possibly as many as 70,000.

The Ilisu Dam

Located on the Tigris river in Turkish Kurdistan, 65 kilometres upstream from the Syrian and Iraqi borders, the $2 billion Ilisu Hydroelectic Project project is part of Turkey's South-Eastern Anatolia Project (Turkish initials "GAP"). The contract for building the dam has been awarded to a consortium led by the Swiss company, Sulzer Hydro. Construction has been subcontracted to a further consortium made up of, among others, Balfour Beatty, Impregilo, Skanska and the Turkish companies, Nurol, Kiska and Tekfen. ABB Power Generation and Sulzer Hydro will supply the generating equipment.

Export credits and investment insurance guarantees are now being sought by the construction consortium from the export credit agencies (ECAs) of Austria, Germany, Italy, Japan, Portugal, Sweden, Switzerland, the UK and the US. Britain's Export Credit Guarantee Department (ECGD) is considering a $200 million investment guarantee for Balfour Beatty, which is leading the construction consortium, and has said that it is "minded" to give conditional approval.

Background to the Kudat Report

When the companies involved in the project first sought export credit support for the dam, there was no resettlement plan for the project. This was in violation of international standards. World Bank policy states: "Involuntary resettlement is an integral part of project design and should be dealt with from the earliest stages of project preparation."1 Under pressure from the public, the ECAs involved made a resettlement plan a condition of their support.

In December 1999, the UK Government released a report prepared for the Export Credit Guarantee Department (ECGD) on the resettlement implications of Ilisu.[REF] The report noted that a large number of villages will be affected necessitating the compulsory resettlement of more than an estimated 16,000 people and affecting a further 20,000. The report concluded that although the Turkish Government was determined to build the dam international guidelines for resettlement were not being observed. Despite the fact that local stakeholders had been waiting for more than 20 years to be informed directly about resettlement (the dam was approved by the Turkish Government in 1982) they had no idea about how they might benefit from the resettlement programme. Land tenure and land title problems had not been addressed in the resettlement planning and no host areas had been identified for resettlement under the Ilisu project. The Report warns that providing appropriate compensation would be complicated by a number of factors: the lack of registration deeds and proof of ownership, forced land confiscation and absentee rural families. It acknowledges that the Turkish Government is in the process of modernising its national policy and legal framework governing involuntary resettlement but adds while "on paper" these measures should be adequate that "evidence from previous resettlement projects indicates that implementation is problematic".

Earlier this year, the Swiss Export Credit Agency commissioned Dr. Ayse Kudat, an ex-World Bank staff member, to help ensure that the resettlement plan and compensation proposals met international standards. A leaked copy of Dr. Kudat's review of the Turkish Government's Draft Resettlement Action Plan has been obtained by the Ilisu Dam Campaign and other NGOs internationally. The Report confirms and substantiates most of the misgivings voiced by the earlier more cursory ECGD report.

Kudat notes the progress that has been achieved in some of the problem areas since the ECGD report was written. But the latest review provides additional social, economic and political background information which shows that the whole resettlement and rehabilitation programme - even if it could be brought up to acceptable international standards - would remain a highly complex and politically sensitive operation.

This briefing analyses Dr Kudat's preliminary assessment of the draft RAP in the light of World Bank and OECD guidelines for "best practice".2

1. Numbers Underestimated

The report says more than two to three time more people may be affected than was supposed by the UK Secretary of State for Trade and Industry, Steven Byers, when he announced he was "minded" to grant UK support for the project.

UK dam builders Balfour Beatty initially put the number of people affected at 12-16,000. The December 1999 report commissioned by the Department of Trade and Industry upped this figure to a maximum of 36,000. The RAP report now says:

"A large number of people are potentially affected and their numbers cannot be estimated. They range between 47,000 and 78,000". (p.1).

Although the RAP Report argues that the actual number affected will be smaller, since some have already been moved and won't seek compensation, it concludes:

"Regardless of the assumptions made, all estimates indicate far higher numbers of affected people than those mentioned in previous reports on Ilisu." (p.4).

2. Failure to Recognise the Significance of the Human Rights/Security Context

The Report promises compensation for those who have already been forcibly evicted from the reservoir area of the dam due to the 16-year conflict in the region. Nonetheless it provides telling evidence that the needs of this category of displaced villagers are being ignored as a result of "security concerns". The full implications for a successful resettlement outcome are not addressed.

Ilisu is to be built in an area that has been devastated by an armed conflict between Turkey's security forces and the Kurdish Workers' Party (PKK). International human rights organisations (such as Human Rights Watch) have reported on the forced depopulation of thousands of villages in southeastern Turkey where the PKK has been active.

Despite government promises to compensate villagers, little effort has been made to facilitate the return of displaced persons to their homes or to compensate them for the destruction and loss of their property.

This is a major complicating factor for the Ilisu resettlement and compensation package. Kudat argues that "the draft Ilisu RAP exceeds international best practice in attempting to benefit those people who were displaced many years ago due to the security situation". (p.2). Yet, her report provides telling information about the neglect of this category of dam-affected people.

For example, Kudat states that:

"DSI field staff calculate that 24 villages (together with their hamlets) lack land/title registration. These villages are still under security, and to most, there is no permission to enter. Their infrastructure and homes are ruined and their populations are spread around the country. Thus, to bring together the land/title registration officials and the villagers is not easy. Despite high levels of landlessness, the displaced communities are still concerned that lack of land/title registration would make it difficult for them to get compensated for whatever they have remaining. The fact that they have lost their trees and gardens is another major concern."7

Kudat argues that it is not appropriate to place the full burden of compensation for those people who have previously been forced to abandon their homes because of the internal security problems on the RAP. But surely, it is appropriate for international donors/guarantors to demand that the Turkish Government assume its obligations to those already displaced (as a result of the security forces depopulation policies) as a pre-condition of support for Ilisu?

3. Sweeping Institutional Reform Required

The report shows that sweeping institutional reforms are needed before "best practice" can be achieved.

The UK government has stated that its support for the project will be conditional on the Turkish authorities drawing up a resettlement plan to international standards. Although this report is upbeat about some aspects of the resettlement plan meeting international standards, it states:

"In a number of areas, additional data will have to be collected if best practice is to be achieved. Also, new institutional arrangements should be formulated and agreed upon between concerned institutions." (p.2).

"The GAP framework alone will not resolve all the institutional complexities of the resettlement projects. There is a need to have a more unified institutional framework, a single earmarked budget for resettlement implementation, and mechanisms for quality assurance, enforcement and monitoring and evaluation..." (p.28).

The Kudat report points out that the ability of the DSI (the State Hydraulic Institute) to implement and enforce the resettlement action plan is hampered by the security situation and by the lack of coordination with other institutions which each act independently and have separate budgets. All five provinces affected by the construction of the Ilisu Dam are still under OHAL (that is governed under "extraordinary circumstances" regime) and as Kudat comments "the Ministry of Interior and the military have very different sets of priorities" (p.27) from the DSI (State Hydraulic Institute) - the project implementation agency. The complexity and magnitude of how best to address the needs of previously displaced population "goes beyond the ability of the project and the solutions require the decisions of security agencies". (p.6).

Kudat deplores the fact that the DSI's resettlement performance has been subjected to such detailed public scrutiny when other bodies, notably the Ministry of Energy, Ministry of Construction and Ministry of Transport have an equally poor or even worse record [footnote 20].

She argues convincingly that Turkey needs to establish a single institution to ensure quality and monitor resettlement performance for all development projects, not just large dams. "Clearly the existing institutional arrangements will not suffice to meet the needs." (p.6).

Kudat says that there is a need to empower the DSI to assume primary responsibility in budgeting and coordinating the RAP as well as to have the flexibility to modify implementation arrangements should institutional arrangements fail to deliver the desired actions in a timely manner.

The admission of the need for such sweeping institutional reform is critical, since implementation of the plan (however well conceived on paper) will depend on such reforms being undertaken. This could take years. There would thus seem to be a prima facie case for the UK government withdrawing support.

4. Failure to Analyse Alternatives

The report acknowledges that the Turkish Government has failed to give adequate consideration to alternatives to the Ilisu project. This violates both World Bank and OECD guidelines for projects involving involuntary resettlement.

World Bank guidelines state: "Involuntary resettlement should be avoided or minimized where feasible, exploring all viable alternative project designs."3

The OECD's Development Assistance Committee guidelines, which the UK government says it will use to assess the project, go further: "Alternatives to displacement and resettlement should be fully considered before decisions on displacement and resettlement are taken... In every case, the alternative to refrain from carrying out the project (the "non-action" alternative) should seriously be considered".4

The RAP Report shows that only a limited assessment of alternatives has been done. There is no assessment of solar alternatives, gas or improving the performance of existing dams. Nor is there any assessment of demand side management or the "non-action" alternative.

In addition, the RAP Report fails to assess the relative resettlement impacts of the 10 alternative designs that the Turkish Government originally considered. The data on these alternatives were not made available to Dr Kudat. The RAP Report states:

"While it is today too late for the Ilisu resettlement plan to have an impact on dam design, the public still wants to know whether or not alternatives were considered and whether adverse impacts on settlements could not be reduced, particularly with respect to Hasankeyf's cultural heritage. Although some 10 alternatives were considered at the time the location of Ilisu was selected, the implications of other technical alternatives for resettlement are unknown." (p.10).

5. Failure to Complete RAP On Time

Both the World Bank and the OECD require a resettlement plan to be drawn up prior to project approval. This condit ion has been flouted.

The World Bank states: "During project preparation, the feasibility of resettlement must be established, a strategy agreed upon, the resettlement plan drafted, and budget estimates prepared."5

The OECD stipulates: "The existence of a time-bound resettlement plan and budget must be a condition of initiating appraisal for projects involving resettlement."

The RAP Report makes it clear that these conditions have been breached, both by the Turkish authorities and (in conditionally approving the project without their being a resettlement plan) by the UK government. The Report says:

"The RAP falls short of best practice by not being done together at the same time the technical design was completed". (p.2).

However the RAP Report justifies this by noting that even the World Bank has failed to uphold this policy:

"... in this the Ilisu RAP is not dissimilar from many others around the world, including those funded by the International Financial Institutions such as the World Bank." (p.2).

The World Bank's breaches of its own best practice policies, which are detailed in the RAP Report, cannot justify or excuse similar faults in the Ilisu planning process.

6. Failure to Carry Out an Adequate Socio-Economic Survey

The Report makes it clear that in violation of World Bank guidelines, the Turkish authorities have failed to carry out a full socio-economic survey. It would appear that the project developers have no intention of fulfilling this World Bank condition.

World Bank guidelines state: "The affected hosts and resettlers need to be systematically informed and consulted during preparation of the resettlement plan about their options and rights."6

And again: "Resettlement plans should be based on recent information about the scale and impact of resettlement on the displaced population. In addition to describing standard household characteristics, socio-economic surveys should describe (a) the magnitude of displacement; (b) information on the full resource base of the affected population, including income derived from informal sector and non-farm activities, and from common property; (c) the extent to which groups will experience total or partial loss of assets; (d) public infrastructure and social services that will be affected; (e) formal and informal institutions...; (f) attitudes on resettlement options. Socio-economic surveys, recording the names of affected families, should be conducted as early as possible..."7

The RAP Report notes that a socio-economic survey has consulted a total of 2,100 households, covering 45 settlements, a third of households in Hasankeyf, and "over 100" displaced households. (p.23) This is out of a total of 183 settlements to be affected by the Ilisu dam. In effect, less than a third of the settlements affected have been surveyed.

Significantly, Dr. Kudat blames the failure to carry out a full survey - which she acknowledges is in breach of World Bank guidelines - on the ongoing conflict in the region:

"Some communities, although not displaced, were not accessible for security reasons at the time of the socio-economic surveys." (p.30).

This may explain why the Turkish authorities specifically excluded a full census from the remit of the Resettlement Action Plan, despite such a survey being required under World Bank standards. According to the RAP Report:

"The terms of references of the consultants preparing the RAP did not require a census and the budget that was allocated for the RAP would have been insufficient to carry out such a census." (p.30)

7. Doubts Over Consultations with Host Communities

International guidelines stipulate that host communities - those which will receive the people evicted by Ilisu - must be consulted. The Report claims that such consultations have taken place. However independent sources suggest otherwise.

The World Bank guidelines state: "The [resettlement] plan should address and mitigate resettlement's impact on host populations. Host communities and local governments should be informed and consulted ... Conditions in host communities should improve, or at least not deteriorate."

The Report claims that

"Consultations were held with governors, majors and representatives of the government agencies responsible for RAP implementation." (p.23).

However, independent interviews conducted in June 2000 established that the Mayor of Batman - one of the towns that will receive considerable numbers of evicted villagers - had not been consulted over the resettlement plan. The Kurdish Human Rights Project has confirmed that this is still the case.

8. Major Information Gaps

The report acknowledges that much of the data required to draw up a resettlement plan to international standards is still not available.

The Report says:

"There are inadequate data at this stage to complete the RAP to a standard of "best practice" in the following areas: conditions of the displaced people and of their communities; land-use patterns; use of common property resources; socio-economic situations of communities not included in the original sample; availability of unused Treasury lands in the impact area; current and projected employment situations in host communities; and a detailed economic analysis of households in the town of Hasankeyf." (p.8).

9. Failure to Produce a Resettlement Budget

No budget has yet been prepared for resettlement, in contravention of World Bank and OECD guidelines. The report also suggests that a paper commitment from the Turkish authorities to make the money available cannot be trusted.

World Bank guidelines state: "Where large-scale population displacement is unavoidable, a detailed resettlement plan, timetable and budget are required."8 Like the OECD, the Bank stipulates that this should be "a condition of appraisal for projects involving resettlement..."9

The OECD similarly lays down: "At least a preliminary estimate of total population to be displaced and overall resettlement costs should be made at the outset, as well as an evaluation of proposed resettlement sites."

The RAP Report says:

"A resettlement budget is yet to be prepared and is likely to exceed previous estimates. The key challenge will be for the national budget to provide a convincing commitment by the State to this budget." (p.31)

10. Relying on Economic Growth to Relieve Post-Project Impacts

According to the World Bank, resettlement projects should ensure that those resettled are not worse off than they were before the project. The Report makes it clear that there are major economic and political obstacles to fulfilling this standard. Contrary to World Bank guidelines, it relies on the hope of future economic growth alone to protect the livelihoods of those who will be resettled.

The World Bank and the OECD both state: "All involuntary resettlement should be conceived and executed as development programmes, with resettlers provided sufficient investment resources and opportunities to share in project benefits. Displaced persons should be ... assisted in their efforts to improve their former living standards, income earning capacity and production levels, or at least restore them."

In the case of landless and jobless affected people, the RAP Report says:

"Whether post project situations will improve this depends on the developments of the next decade that currently appear to be positive but are difficult to predict with any certainty." (p.7).

The RAP Report also says:

"If, on the other hand, the region's economic development kicks off within the next year or two, the landless from the affected communities in Ilisu may start making a gradual shift to an improved livelihood. Thus, whether the poor among the affected populations could be worse off depends on the near term economic and political outlook of the region." (p.16).


"While it is unlikely that rural resettlement could be managed to restore completely and/or improve incomes in all cases, urban opportunities will grow and could be managed to the advantage of affected populations." (p.22).

Relying on future economic growth alone to restore or improve the living standards of the displaced population violates World Bank guidelines. The World Bank states: "Normally, general economic growth cannot be relied upon to protect the welfare of the project-affected population."10

The RAP Report itself acknowledges that "it is generally risky to build a resettlement plan solely on the basis of a region's development." (p.25). In a tacit admission that such development is by no means assured, the Report recommends that, in the case of host communities, "the RAP should ... consider host community needs should the expected positive regional developments not take place in a timely fashion." (p.25).

Significantly, the report clearly spells out the failure of past GAP projects to benefit affected communities, causing significant social problems and economic impoverishment. A footnote is more explicit:

"The urban centres of the Southeast Turkey that received high levels of migrants suffer from widespread unemployment and poverty. Educational levels are high and typhoid, dysentery and other infectious diseases are widespread. Staff shortages in schools and in the health sector are still acute. In some parts of the city of Diyarbkir, people continue to drink water from the irrigation canals." (pp.25-26).

11. Failure to Assess the Gender Implications of Resettlement

Both the OECD and the World Bank lay down that special provision should be taken to protect the livelihoods of women. According to the report, this guideline has been ignored.

The OECD states: "Since women ... contribute significantly to the well-being of their families ... planning for relocation should consider their preferences and should address their specific needs and constraints."11

The RAP Report says:

"Despite the preparation of the survey instruments by female social scientists and despite the conduct of the qualitative fieldwork by them, no systematic effort has been made to capture the gender dimensions of resettlement." (p.23).

12. "Independent Monitoring"

The UK government has publicly stated that it is a fundamental condition of its support that the resettlement of affected people be independently monitored. The RAP Report gives fleeting mention (just one paragraph) to the independent monitoring of the project and planning process. The Turkish government has strongly objected in the past to the idea of 'foreign interference'.

The Report says:

"There is also a need for independent monitoring not to be equated with "external" or "foreign monitoring". The principle here is to: (i) internalize financial oversight, technical construction progress, expropriation payments, etc., monitoring within the institution that implements projects involving land acquisition and resettlement; and (ii) externalize the monitoring of the development results or impacts of projects to enhance objectivity and transparency." (p.28).

Although the acceptance of the need for independent monitoring is to be welcomed, the Report fails to specify how such monitoring would be carried out, how the monitoring team would be composed, or who would decide on its composition.

The report also fails to address the implications of the continuing conflict in the region for independent monitoring. It does, however, acknowledge that the conflict poses major problems (euphemistically referred to as "challenges") for any resettlement process:

"(...) internal and external civil society organizations demand government intentions be put to the test in the context of the Ilisu Dam. This demand is articulated in the request for the creation of a free environment in which views concerning the construction of the Ilisu Dam can be openly expressed. The decisions concerning the resettlement choices for Ilisu and the normalization of the martial law conditions thus become closely interrelated." (p.5).

The author does not explicitly clarify whether she shares the view that participatory rehabilitation projects cannot be planned and carried out under martial law.

13. Failure to Compensate for Loss of Customary Rights

Those who are deprived of access to pasture lands because of the dam's reservoir will not be compensated under current compensation law, in contravention of OECD guidelines.

The OECD guidelines state: "Customary land ownership and usufruct rights must be recognised for compensation purposes to avoid the destitution of former users."

The Report acknowledges that "access to pasture lands commonly held by the communities ... will be jeopardised in communities that will be fully affected." Those forced to sell their cattle, however, will not be entitled to compensation.

According to the Report:

"The national legal framework does not address the issue of common property resources for pasture in a comprehensive manner. The policy vacuum with respect to common property resources has so far meant the disregard of the potential income restoration for the livestock owners." (p.26).

14. Previous Record of the Turkish Authorities on Resettlement. The Report Acknowledges the Turkish Authorities' Appalling Record on Resettlement but Fails to Consider the Full Implications for Future Resettlement at Ilisu

The key message of the World Bank's own detailed research into involuntary displacement is that "resettlement is nearly always more difficult, more expensive, more time consuming than is generally realized".

Given this fact, British NGOs and academics would argue that any decision by the DTI to support a project that entails involuntary displacement should be informed by the country's previous record in dealing with resettlement. Turkey's poor record on resettlement was clearly not taken into consideration by the DTI when giving conditional approval for the project.

Kudat comments: "In the Turkish context, past failures have been particularly severe with respect to inadequate and inappropriate delivery of resettlement housing, lack of concern with the well-being of self-settlers, transparent participation of affected populations in resettlement decisions, and monitoring of social impacts during and after dam construction". (p.6).


The RAP Report does not analyse the economic, political, environmental and cultural impacts of the Ilisu dam. Concerning resettlement and rehabilitation, it puts forward at least nine project-specific and institutional recommendations. Most of these have never worked in practice -- in Turkey or in World Bank projects in other countries. The report counts on the economic situation of the region to improve, so that displaced people can find an income away from their land. It implies that martial law must be abandoned so that any participatory approaches to rehabilitation can be considered. And it heads the list of risks by pointing out that "the real test of this commitment (to the implementation of Ilisu's RAP) will be manifest in the RAP's implementation stage and especially in the annual budget allocations that are effectively made available to the RAP." (p.36).

Who will carry all these project-related, economic and political risks? It will be the affected people, and most of all, the poor, landless majority. This is even more likely since most export credit agencies refuse to make their loans or guarantees conditional on the actual implementation of the plans and promises of the project authorities.

As the Report points out, "there are still a large number of people affected by previously constructed dams [in Turkey] who are still waiting to be resettled". If the project authorities are indeed serious about their commitment to equitable rehabilitation programmes which conform to international standards, they should start by resolving the problems which have already been created. ECA representatives have suggested that the Ilisu dam is a test case for Turkey's compliance with international standards, and that the country would have problems raising further export credits if it failed this test. The Ilisu Dam Campaign and other NGOs continue to consider past projects as the test cases. They will support the people affected by Ilisu and their representatives if they continue to oppose this new dam project which again shifts all the risks to them.

Notes and references

1 World Bank Operational Directive (OD) 4.300 'Involuntary Resettlement', June 1990: OD 4.3(3)

2 Other international standards also cover forced resettlement. Article 11 of the International Covenant on Economic, Social and Cultural Rights covers the right to adequate standard of living, including the right to adequate housing. There is a very clear General Comment on Forced Evictions which sets out stringent measures that governments (including donors) need to take to minimise harm to affected groups and individuals.

Article 22 of the ICESCR places an obligation on the international community including multilateral and bilateral donors to ensure that development cooperation is "advisable" and does not erode people's rights (in the context of displacement of Kurds at a time of such instability in the region, as DTI must be aware, the potential for misinterpretations of the resettlement of the population even in optimal circumstances is obvious).

3 WB OD 4.3(3a))

4 OECD Development Assistance Committee (DAC), 'Guidelines for Aid Agencies on Involuntary Displacement and Resettlement in Development Projects, Paris 1992: DAC p.6.

5 WB OD 4.3(29)

6 WB OD 4.3(8)

7 WB OD 4.3(11)

8 WB OD 4.3(4)

9 WB OD 4.3(30)

10 WB OD 4.3(18)

11 DAC p.7.


Related articles: