Lessons learned from Narmada

96034 C O t ,$C5 CONFIDENTIAL International Bank for Reconstruction and Development FOR OFFICIALUSEONLY S-cM93-516 FROM: The DeputySecretary May 24, 1993 LESSONS LEARNEDFROM NARMADA Attachedfor information is a President's Memorandum entitled"Lessons Learned from Narmzada"and accompanying reportentitled"Bank4ide Lesons Learned from the Experience vith the India Sardar Sarovar (Narmada) Project'. Questions may be directedto Mr. Wyss (Director, Bank Paris Office, X580-3010) or throughthe Officeof the President (Ms.Armitage, X81114). Distribution: ExecutiveDirectors and Alternates Officeof the President ExecutiveVice Presidents, IFC and MIGA Vice Presidents,Bank, IFC and MIGA Directorsand Department Heads,Bank, IFC and HIGA This document has a restricted distribution and may be used by recipients only in the performance * t 1r <.t . r?,¢'>-1ret244';C e's ._1 ? t t?5  BANK THEWORMD D.C. 20433 Washington, U.S.A. T. PRESTON LEWIS May 21, 1993 To: Executive Directors Lessons Learnedfrom Narmada During the Board discussion of the Status Report and Next Steps for the Sardar Sarovar (Narmada)Projects last October, we promised to review the Bank-wide lessons learned from the Narmadaexperience. We asked Mr. Wyss to undertake this review, and his report is attached for your Information. It focuses on lessons learned in three main areas: :i) project quality at entry; (ii) communications and decision-making; and (iii) ownership. We accept the recommendationsmade by Mr. Wyss, which are consistent with those of the Report of the Task Force on Portfolio Management. They will be implemented in the context of the follow-up to that report. Attachment K LESSONSLEARNED BANKWIDE FROK THE EXPERIENCEWITH THE INDIA SARDARSAROVAR PROJECT (NARMADA) HANSWYSS MAY 19, 1993 ( BANKVIDE LESSONS LEARNED FROM THE EXPERIENCEWITE THE INDIA SARDAR SAROVAR (NARMADA) PROJECT Table of Contents Page No. Introduction .1....................................... I Approach. 1 A. Quality of Project at Entry and the "Incremental" 3 ......................... Communicationsand Decision-Making B. ...................................... 7 C. Ownership ...................................... D. Recommendations 7 Final Comment ............................ 10 ATTACHMENTS 1. List of Persons Interviewed 2. Policy Basis for Projects: Project Appraisal (extract from "India: The Sardar Sarovar (Narmada) Projects, ManagementResponse," dated June 23, 1992 (SecM92-849)) 3. Suggestions for Improvementof Quality at Entry for R+R  CONFIDENTI4 (- EzPEa1XNCE VIWM THE INDIA BANIVIDE LESSONS LEANJED P10 T SARDAl SAROVAR(NA4IMADAAPROJECT Introduction 23, 1992 on the above 1. In sumarizing the Board meeting of October review the Bankwide that, "we will.... project, the Chairman said, inter alia, this project." lessons learned from our experienceof may be drawn from what we 2. This note sums up the main lessons which avoid It then outlinesactions designed to know at present about the subject. files in the Asia on a review of the the failures observed. It is based more than 30 persons (listed in InformationCenter and on interviewswith with the project and/or Attachment 1); all were associateddirectly Review. developmentsleading up to the Independent time or another, the attention 3. A myriad of factors required, at one project that is among the largest and of those dealing with the project - a This review zeroes in on those issues most complex ever financedby the Bank. to have an IndependentReview which eventually resulted in the decision main failures occurred in three major carried out. It concludes that the areas: to the 'incremental" (i) quality of project "at entry", leading up approach; (ii) communicationsand decision making; and (iii) ownership of project. were mutually reinforcing. Shortcomingsin each of the three areas "Incremental"Aoproach A. Oualitv of Prolect at Entryand the was prepared in the very best 4. On the engineeringside, the project and reflectingtheir newness in the of the Bank's tradition. In contrast, as resettlementand rehabilitation (R+R) Bank's widening development agenda, not addressed up to the aspects were well as many, if not most, environmental post-appraisal during a period of intensive 1983 appraisal. They were handled were approved in early 1985, there remained, work. When the loan and credit however, major shortcomings: (i) in R+R: for a very small number of * lack of baselineinformation: except of the people affected, the villages affected, no census existed their future options their economic activity, level of income, their preferences. What (most important: land availability)or  2 was available in terms of baseline information I were surveys of differing scope and details; * lack of specific Rolicies covering all the different groups of neonle affected - policies which were to ensure that the objective stated in the legal documentswill be achieved (i.e., after a reasonable transitionperiod, the displaced people would regain at least their previous standard of living). Thus, the commitmentto this broad policy objectivewas not translated into specific, monitorable policy measures agreed with the Government of India (GOI) and the three state governmentsconcerned; and * the ganal (Rart of the) Droiect did not grovide tor R+R since it was expected that people would be affected only to a limited extent and that their compensation would not raise special issues; (ii) on environment: * the effects of the large canal works were not the subject of a review of environmentalaspects, though water logging and salinity were addressed extensivelyin the engineering studies in the light of experiencefrom other irrigationprojects in the region; and * the upstream environmentalreview was limited to the area to be inundated rather than covering the full basin. In sum, shortcomingsat 'project entry" were especially glaring on R+R. In the light of subsequentdevelopments,this review concentrateson R+R. True, there are also important environmentalissues, including with respect to forest cover (especiallyon differences in the assessment of the degraded nature of such forests), biodiversityand health aspects. 5. The absence of adequate baseline information,in combinationwith the poorly specified policies, was certain to make future implementationof R+R difficult - even more so as R+R touched on politically sensitive points in GOI and particularly in the three state governments. Therefore, the resettlementplans prepared for, and agreed to during negotiations,did not have much operationalmeaning, especially since the institutionalarrangements for implementingthese plans were not well developed. Finally, the monitoring arrangements for R+R were woefully weak: there was initially reliance on poorly selected independent research institutionsretained by each of the three states and on GOI through the Narmada Control Authority which was unproven in this area. The term "baseline information"is used here in the narrow sense of the detailed informationbase essential for R+R planning purposes. (In much of the Region's discussion on the project in recent years, "baseline"became a measure against which readiness of R+R was to be judged.)  3 6. Project implementation was postponed by GOI for about two years because of delays in environmental clearances. But evidence already had built had large differences in their up by 1987/88 that the three state governments on their different interests vis- views on the project's R+B dimension,based Vigorous involvementby the a-vis R+R and distinct political environments. overcome these serious shortcomings. Bank appeared to offer the only chance to supervision:detailed action plans Thus, R+R became the subject of intensive every six months. This were being introduced,reviewed and redefined only for obtaining the not approach, termed incremental,was designed but also in the broader GOI the Narmada project, necessary progress on R+R in context (variousother Bank-assisted projects with R+R elementswere being pursued with special vigor when an implementedin India). The approachwas and activist team of first-in-line managers was brought in during late 1988 project supervisionresponsibility shifted to the New Delhi office. was at Headquarters [HQ] with (Previously,overall supervisionresponsibility office - inter alia, the latter some aspects being handled out of the field and after the 1988 shift of overall had a junior person on R+R matters before the AgriculturalUnit at New Delhi project responsibility. The new head of R+R aspects, and requested took a strong personal lead on the project's and/or consultants.) The total HQ staff periodicallysupport from specialized years exceeded 80 staff weeks annual level of supervision inputs in many being devoted to (includingconsultants),the largest part issues. Nevertheless,first-in-line engineering/procurement and institutional provided with more resources for supervisorsfelt that they should have been project supervision. but also considerable 7. The result was a very close Bank involvement, to improve R+R policies, state governments Bank pressure on EOI and the three At no stage was there an agreed and to prepare and implementR+R programs. of R+R implementation on a project basis against which to measure progress were plans of varying scope in wide basis, although from 1990 onward, there paragraph, there were preceding the three states. As pointed out in the the light of the findings of in periodically (re) defined specific targets of moving targets (often supervisionmissions. This led to an appearance "new conditions"imposed by the as regarded, and resented, by Indian officials over time, mainly after 1988, in Bank). Overall, major progress was achieved obtaining better knowledge of the terms of defining outstandingpolicies, resettlementplans, leading a people affected and enacting specific, 'rolling" involved to believe - even now - good number of the Bank staff and managers to be an excellent development prove that, overall, Narmada will eventually of this progress was needed simply to overcome project. However, much Only partial adherence to, and non- original shortfalls, and back-sliding. commonly giving way, in the light of fulfillment of, specific targets occurred to newly defined targets for later subsequent supervisionmission findings, of a widely agreed basis for periods. Most important,in the absence the finger on specific measuring progress, it became easy to point shortcomingsof this approach. B. Communicationsand Decision-Making in Operations Policy 8. On the advice of the Agriculture Department concerned decided after the 1983 (OPS/AGR), the Regional Assistant Director  4 apraisal that the project was not ready for negotiationsuntil R+R and environmentalaspects were addressed in line with Bank policy. Subsequent staff work on R+R was done by a (world renowned) consultantwhile a Bank felt member dealt with the environmentalaspects. The consultant eventually plan brought by the Indians under strong pressure to accept the resettlement to neg tiations - no Bank staff in his profession was at his side during in negotiations though his earlier reports had been reviewed (favorably) still experience with its recent 11980] R+R policy was OPS/AGR. (The Bank's out in very limited at that time; analytical and planning tools were spelled in the Bank's 1988 Working Paper No. 80 on 'InvoluntaryResettlement Development Projects"). When the loan documents were sent to the Loan has been Committee Chairman, the cover memorandum stated: "a good R+R program for implementationand for agreed, togetherwith institutionalarrangements monitoring and evaluation." 9. During supervisio, differencesemerged periodicallybetween staff as to what importanceshould be attached to issues faced in project the vis-a-vis implementationand as to how these issues should be tackled 1987 onward as to borrower. Differencescame up at various instances from with suspension of disbursements whether the Bank should threaten the borrower conditions. institutional and/or R+R in order to enforce agreed technical, more than date for the canal credit was used From 1989 onward, the closing three areas once for putting specific conditionsonto its extension in the threatening listed. (Previously,in 1988, a letter was sent to the borrower eventual suspensionof disbursements, but a partial response was accepted as sufficientlyresponsivenot to proceed with suspension.) 10. Decisionswere made first by the responsibleTask Manager (TM) in Department (CD) New Delhi and by the Agriculture Division Chief. The Country Reports Director was fully brought into this process through Supervision missions. (especiallydetailed since 1988) prepared by semi-annual supervision are Differences of views regarding supervision findings and recommendations, and, Chief not unusual. Clearly, supervisingmanagers, i.e., the TM, Division if needed, the Director have to decide on conflicting conclusions/recommendations. Some special factors characterizedthis case: (i) the immediate supervisors and the Country DepartmentDirector believed stronglv that Bank involvement was critical for imDroving the R+R outcome for the DeoDle affected by the proiect. to Suspension, in their view, would reduce the Bank's ability improve the situation for these people - as one interviewee put it: "short leash rather than suspension* was the motto. Given in the judgement of the TM on a sustained "overall net progress" from 3 to 2 implementation,the supervisionrating was improved from late 1988 onward; faced (ii) the Bank increasinglyrecognized the political difficulties states. GO1 in achieving the R+R objectives across the three to being well aware of this issue, at times leaned on the Bank not "give up" (and some NCOs expressed similar views); (iii) there was no strong tradition in the India country context for using threats of susDension - at least this is what staff  5 concerned said they felt, especially those who had worked extensively in other Regions, and finally, greater continuityin HO rofassional Ito*taff ass (iv) RaE would have been helpful (continuity at field offi level was referred to in paragraph 6); in this contextl ived different views have been expressed among those directl staff assignments. Kos! on the reasons for changes in HQ . made important, no senior Bank grofessionalstaff. recognize dL= Mae and outside the Bank for his/her capacity on R+R, lImited charge of this Rroiect coMM=nent (even for a take direct Recruitment of senior, highly experienced staff and period). between their career development,and much closer collaboration central staff are clearly required - the Country Department and new ' this is receivingmore management attentionunder the arrangements which were introduced in January 1993. organizational At a different level, there were also too many discontinuities at Division Chief level - the present is the sixth since 1984. at least one manager now concludes that suspension in 1989 With hindsight, up R+R actions. would have been a healthy "shock" and might have speeded the project 11. Communications with high2r levels of Bank Management on Operations, 1986, the Senior Vice President (SVP), were unusually close. In reviewed all ongoing Bank-financed resettlementprojects and asked for action reorganization,the new programs, including on Narmada. After the 1987 by the Country Department RegionalVice President (RVP)was kept informed memoranda from Executive early on (the flow of letters from the outside and to his attention). From his visit to Directors [EDs] had brought Narmada about the project. India in late 1988 onward, the SVP also was well informed COnsistentwith the 1987 reorganization, project specific action was left (CD). Given the very largely to the initiativeof the Country Department SVP on any important close interactionbetween the CD Director, RVP and communications were often oral. The only written matter, especially on India, in a draft letter recommendationfor threateningsuspensionwas contained in direct. handled prepared by the CD in 1991 - this was subsequently to the letter sent by the CD discussionswith GOI (this event was in addition Director in 1988, referred to in para. 9 above). preparation, 12. From 1983 onward, i.e., from the period of project began with both Indian and international Non-Governmental communications criticism about R+R, some Organisations(NGOs). These generally expressed degraded] forest land to also about environmentalaspects (mainly on [highly took place with NOOs, and be lost in the inundated area). Detailed exchanges coming into the form letters were prepared in response to 'chain letters' through copies. informed Bank. On major letters sent to NGOs, GOI was kept Bank communications with NGOs evolved as follows: (i) Drior to loan/credit aDproval: NGOs were assured that Bank had a Manual progressive R+R policy (which was outlined in Operationa4 1980) and that Bank finaicing Statement (OMS] 2.33 of February C01 "&isLstent would be extended only if there was agreement with : with this policy;  . 6 (ii) from loan/creditanDroval till about 1987/88: NGOs were assured that the Bank had a progressiveR+R policy, that this policy was fully incorporatedin the loan/creditagreements, that there was an agreed resettlementprogram (with details for phas- I and a rolling plan thereafter),and that carefully designed monitoring arrangementsyere in place; (iii) after 1987/88: letters referred to the intensive dialoguebetween the Bank and GOI/state governmentsand to the progressive implementation plans which had been designed, and were being implementedwith a view to ensuring that the agreed R+R objectiveswill be achieved. 13. Communications with EDs started almost at the same time as those with NGOs - indeed the same international NGOs tended to inform their respectiveED about the concerns communicatedto the Bank. Many written briefings were sent to individual EDs in response to specific requests - in addition to many oral briefings. Beginning in July 1988, "informal'meetings took place on a semi-annualbasis between Mr. Arlman and other EDs (mainly attended by Assistantsto EDs), and the Region. It should be noted that the Indian ED, although invited, did not attend these meetings; this changed in 1990. The RVP and/or the CD Directorwere present at these meetings with a group of project managers and staff. The meetings generally coincidedwith the completion of the semi-annual supervisionmissions. The exchanges (records of these meetings in the files are not detailed) tended to focus almost exclusively on R+R, although staff tried also to address other major project issues. From the limited records, especially in the light of the detailed written briefs for the first meeting, it appears that the message from the staff was that important progress on pending R+R matters was underway and/or agreed to be addressed though major difficultiesremained (especially in Madhya Pradesh). Managers and staff participatingin these meetings believe they presented complete factual briefings at these meetings. 14. Over time the EDs became increasinglyimpatient about what they considered as slow progress and even backsliding. They felt Bank staff gave them ever hopeful, rosy reports with emphasis on selected areas of progress while they receivedvery different reports from NGOs. Eventually, in early 1991, meeting among themselves, they concluded that they needed an Independent Review to satisfy their constituents. When they brought the subject to the President,he already had significantknowledge of the project: he had received a briefing on Narmada in preparation of his visit to India in late 1987 (during that visit, a meeting with NGOs had been quite friendly and he had not obtained a sense of a major problem brewing). Subsequently, especially in 1990, he had answered many letters sent to him on the subject. 15. In the light of the foregoing,it is clear that channels of communicationwere not clogged, either within the Bank or with the outside. If anything, communicationswere 'too good': a number of messages went through various channels where, later on, recipients felt the message had been misleading. This was the case, first, internallyat the time of negotiations and subsequentlyat Board approval, when R+R arrangements appeared well in hand given the agreed resettlementplan and monitoring systems. This line was reflected fully in exchanges with NGOs and EDs. As the Bank moved into its activist "incremental"approach, communications became more difficult. The  7 initial answers did not suffice anymore. True, the Bank's efforts were recognizedfavorablyby NGOs interestedin the betterment of R+R conditions for project affected people. But those who opposed the project as such found it convenientto undermine the Bank's credibilitywith 'its' moving targets (often not entirely fulfilled);the same NGOs pushed the former group from center stage and vigorously attacked the Bank thereby putting pressure on "their" EDs for an IndependentReview. Differences of views between staff members, referred to earlier,became known both to NGOs and to EDs. This added to the conflictingnature of signals perceived by EDs. C. Ownership 16. Looking over the developmentsrelated to this project, the most striking feature is the Bank's gradual "loss of control over the agenda." A group of very determinedpersons, operating through NGOs and other constituentsof Bank shareholderseventuallytook over the agenda - up to having a group of EDs induce the President to contract an IndependentReview. This happened at a time when Bank staff were making strong efforts to ensure that the R+R objectiveswould be achieved according to the agreementsreached in only very broad terms in 1985. 17. Underlying this extraordinarydevelopment was the following:the Bank had (been) moved into a situationwhere it was seen more and more as the owner of, or at least the major force guiding, the R+R aspects of the project (the dam/canal investments as such had the unflagging support from GOI and the three state governments through numerous government changes). NGOs succeeded in working directly with, and on, the Bank to take action on R+R - as if the Bank were the owner of the project. The Bank itself fuelled this tendency in two ways: first, when responding to NGOs, it addressed on its own the substance of the issues raised by the NGOs rather than bringing these to the attention of GOI for its handling (and, as appropriate, advising GOI on technical aspects). Second, the Bank tried to engage NGOs to exert a positive influence on GOI and the State Governmentsto improve R+R. 18. Beyond the NGOs' looking to the Bank as their chosen instrumentfor ensuring desired R+R results, GOI itself - in subtle ways - also turned to the Bank for resolving the difficult R+R problem which it, or at least a part of the Central Government, considered to be mainly in the hands of the three state governments. 19. As the pressures from some NGOs mounted on major Bank shareholders, a group of EDs turned to Bank management for resolving the R+R problems of the project rather than to the borrower, thereby heightening further the Bank's exposure as the "main actor". Finally, Bank staff and management, in turn, added to this tendencyby taking a very active role in pushing for improvementsto overcome the original project deficiencies. Thus the Bank, rather than the borrower, found itself tagged as the non-performing party. D. Recommendations 20. This section contains a set of recommendationsfor actions designed to help avoiding the main failures observed in this project. The proposals  8 are presented along the three priority areas identified in this note. Given the special importanceof the last factor reviewed ("ownership'), this is where we begin: (i) Ensuring greater ownershipby borrower/executingagencies (a) Bank/borrogerrelations: there is nuch room for engaging the borrower more deeply in R+R, as has been done by the Bank in other parts of the project (e.g., through the large team of supervisoryengineering consultants as well as two specializedpanels of high level experts for the dam and for the canal - all engaged by GOI/state governments in accordancewith their agreementswith the Bank). The borrower/executing agency(ies),as a matter of course, should be asked to engage consultants,financed under the project, for supervisingexecution of R+R programs. There should also be a preference for Bank financing of R+R components, in which case supervising consultantswould certify reimbursementrequests. As appropriate, i.e., depending on the complexityof the project's R+R issues, specializedpanel(s) may also be requested (this is already done, for instance, in the large Lesotho Highlands Water Project). The objective of these requirements is to deepen the borrower's ownership,but also to strengthen the Bank's ability to play its "arms length" supervision role, i.e., to hold the borrower/executing agencies to the commitments under loan/credit agreements. (Thus, such 'arms length' role still requires an activist approach by Bank staff.) At the same time. the proposed approach would reduce the undue load now put on Bank staff for R+R supervision.2 (Note: this is much broader than an R+R issue. The above approach ought to be consideredsystematicallyin other soft areas as well). It is recognized that the role of consultants in supervisingproject implementation has been a controversial subject in India; nevertheless, in the light of the Narmada experience,a review of this matter may be merited. The Narmada project suffered from an unusual complexity of contractual arrangementsunder which the Bank had to deal with GOI as the Borrower and three states as executing agents. These arrangementslaid the ground for the Bank to play an active, if not 'directive",role of coordinator. 2 Since OD 4.30: InvoluntaryResettlement,needs updating in any case (in the light of decisions listed in Attachment 2), it should be reviewed against the apparent perception of some staff that the intent of the OD is to get the Bank directly involved in project implementation.  9 to States in India (unlike The Bank does not lend directly as has been the case in in some other countries)although, agreementshave been entered the Narmada project, project limits the development of into with States. This situationrelation with the decisive of a healthy direct (financial) case. It is the result authorities,the States in this Federal government and the fundamental arrangements between and these are therefore not susceptibleto change. States, project, there would appear Thus, perhaps in this particular setting up a cormon authority to have been good reasons for critical directly taking charge of all (and which would have received investment/operational issues Bank financing). In any case, agreementsshould be set n to make the Bank a forth in a way that will tend and thus an active party to de facto project coordinator implementation. of Bank/borrowerrelations, Finally, on the same subject mission has probably supervision through a resident of - the Bank's increased - at least the perception While such an involvement in project implementation. be a primary factor of Bank arrangement is unlikely to more carefully where "ownership",it should be considered the Bank toward direct other factors tend to draw coordination. involvement in implementation relations with NGOs have evolved (b) NGO Relations: The Bank's 1980s. The Bank has recognized significantlysince the early of obtaining inputs from NGOs in particular the usefulnessNGO involvement in specific into policy development,and become much more important. Bank-financedprojectshas vis-a-vis complaints from NGOs However, the Bank's practice are not a direct contractual on projects in which these normally such complaintsought party, merits to be reviewed: by the Bank but be given to not to be addressed in substance with Bank advice if the borrower for handling (preferably the Bank should continue course, government so wishes). Of as part of its by NGOs to look into matters raised and take such matters up, as supervision responsibilities appropriate,with the Borrower. major (c) EDs/Shareholders Relations: discussion of anyoperationby a Bank-financed complaints regarding a specific should only take place group of EDs with the Bank staff the borrowing country in the presence of the ED representing Secretary of the Bank (or even concerned - if needed, the to get such ED involved. the President) should intervene in to the Bankwide actions covered (ii) Oualitx at Entry: in addition Review, to the Independent the initial management response 23. 1992 (reproducedas Attachment addressed to the Board on June emerging from various 2), a number of specific suggestions, 3. interviews, are given in Attachment  C 10 At a more general level, our main recomeendationrelates to the need for setting specific guideposts(for proper M&E) at the time of project entry. There is a need for the borrower and the Bank to ensure that a cormon basis has been establishedand agreed for AUl relevant R+R factors so that progress can be monitored properly, and also that a meaningfulproject evaluation can eventually take place. The interest of OED in such guideposts is self-evidentand its role in setting appropriate ground rules for defining such guidepostsneeds special consideration. (iii) Communicationsand Decision Making: (a) first, the importanceof the highest professional standards, includingcandor, for internaland external communicationscannot be overstressed. This has been a,hallmarkin the Bank's traditionalareas of operationalactivitiesand the Bank cannot afford to see these standards weakened in the newer areas of the Bank's developmentagenda. Second, the bureaucratic tendency of passing the good news - mainly in the sense of (over) optimism on timely borrower actions - toward EDs needs greater attention. Realism is required, togetherwith timely action on difficult issues. Third, on decision-making, a more explicit/open recognition of different staff views needs to be encouraged; this would also help maintain broader staff support of final decisions; (b) the weakest element underlying decisionmaking in the present case was the inability of the responsible Division and Department management to attract and nurture senior professional(s) in sociology/anthropology (with on-the-ground R+R management experience) into responsiblepositions for the handling of R+R aspects - this has not been achieved even over a long period. (This should be of special concern given the many and large Bank- assisted projects with R+R componentsin India). Such a professionalvoice is needed to carry weight, both within and outside of the Bank. Final Comment 21. At least one of the senior intervieweessuggested that the institutionalcost incurredby the Bank in being associatedwith large projects involving R+R may be prohibitive. This is the only major suggestion which the writer of this note does not share. To limit the Bank's assistance in the area of irrigation (and "clean" hydro power), would be a strong indictment that the Bank has lost its ability to remain an effective agent in assisting its members in critical areas of development. An effective response to the points raised in this note is well within the capacity of the Bank. Hans Wyss 5/19/93 . Atttagb=et1 LIST OF PERSONS INTEPVIEVKD (in alphabeticalorder) Bilsel Alisbah Shawki Barghouti Thierry Baudon Michael Baxter Tom Blinkhorn David Butcher Michael Cernea John Clark Pamela Cox Gloria Davis ChristophDiewald Carlos Escudero Gerald Fauss Robert Goodland Scott Guggenheim Marianne Haug Chaim Helman Eveline Herfkens Attila Karaosmanoglu Anupam Khanna Jochen Kraske Olivier Lafourcade Per Ljung Ngozi Okonjo-Iweala William Partridge Moeen Qureshi Daniel Ritchie Sven Sandstrom Thayer Scudder Ernest Stern Heinz Vergin Jan Wijnand Joe Wood  Attachment2 Page 1 II. POLICYBASIS FOR PROJECTS; PROJECTAPPRAISAL 6. The IRR examinesin detailBank operational guidelines and procedures as they appliedin the 1980s to the Bank'swork on the Narmada projectsl. It:' reviews the evolution of the guidelines up until the present. characterizing them as setting the highest standards of any aid or lending organization in the world for attempting to influence Borrowers and implementation agencies to mitigate adverse consequences to human vell-being causedby involuntary resettlement.The IRR also examinesthe aspectsof the Indian Narmada Water Disputes Tribunal award.dealingwith resettlement as vell as the Bank Loan and Credit Agreements for the Narmada projects. The IRR concludes that despitethe existence of explicit operational guidelines, the Bank failed to insiston a properpreparation of R&R plans by the Government and accordingly did not appraise adequately the resettlement components of the SardarSarovar projects. 7. Response. Bank management acceptsthe findings of the IRR with respectto the guidelines in forceand the failureof the preparation and appraisalto take these fully into account. Thesedeficiencies were particularlynoteworthyin areas that todayhave become important parts of the preparationand appraisal process -- i.e. insistenceon Government consultationwith affectedparties;development of socio-economic data on projectaffectedpeoples;assessment of the Government'simplementation capacity for resettlement. 8. As the IRK acknowledges, the application of relatively new operational guidelines in the early 1980s demanded high quality resettlement projectcomponents. This was not a simpletask for eitherBorrowers or Bank staff. It required major improvements in the practices of many Bank Borrowers as well as a comprehensive change in approaches and resource allocation for all Bank units dealing with such projects. During this adjustment process. particularly in the first few years after the 1980 adoption by the Bank of the new guidelines, there were a significant number of departures from the guidelines, including in the Narmada projects. 9. These departures led management early in 1986 to undertake a thorough reviewof the operational experience with the guidelines.As a resultof the review,seniormanagementissuedinstructions that potentialor known resettlementissuesbe flaggedprior to appraisal missiondeparture. In addition.Bank lead advisersin centraloperating departments were directedto raiseany unresolved problemswith regional managementand, if necessary, with the Bank-wideLoan Comittee. 10. RemedialAction. Taking into accountrecommendations of the IR!. and based on the Bank's own broader experience management with resettlement, has decidedto initiateseveralorganizational measures-- both to and operational ensure adequateattentionto projectsin the currentportfolioand to strengthen furtherreviewprocedures for new projects. 1 Theseinclude(a) OperationalManualStatement No. 2.33, 'SocialIssues Associatedwith Involuntary Resettlement in Bank-Financed Projects' approvedin 1980; (b) OperationalManual StatementNo. 2.34, 'Tribal People in Bank-FinancedProjects'approvedin 1982; (c) Operational Manual Statement No. 2.36, 'Environmental Aspects of Bank Work'. approvedin 1984.  2 Attachment . Page 2 C (a) Bank-wide Resettlement Review. The Environment been requestedto carryout, in cooperation with Department all has regions.a Bank-wide analysisof all ongoingprojectsentailing resettlement, regarding qualityend consistency of project implementation by the Borrower of the Loan and Credit agreements and Bank guidelines. A report.updatingthe 1986 review,to determine the extentto which post-1986 appraisal work and implementation assistance on RLR conformsto the directives, will be submitted to the Bank's operational management, and subsequently to the Board within about 12 months. In parallel, a set of remedial measureswill be taken immediately in the case of the Narmadaprojects.These measures are outlinedin sectionsIII and IV. (b) Staffingand SkillHix. CountryDepartments (CDs),which have a largeportfolioof projects with resettlementand environment impacts,will review their staffingpositionswith a viev to strengtheningsocialand environmental specialistsin their core staff. (c) Asia Environmental and SocialAffairsDivisions. The Environment and SocialAffairsDivision(ASTEN)in the Asia Technical department will be restructured into two units, one dealing with environmental issues and the other with socialand resettlement issues. Staff will be augmented to support adequately the requirements of the South and East Asia Regions. td) Training for Bank Staff. New trainLng courses will be developed for concerned operational staff covering Bank policy. procedures and best practices in resettlement. EDI TrainingAvailableto Borrowers. EDI will. introduce (e) special, free-standing training courses on resettlement, the current training courses available on complementing be environmental matters. These training courses will organized at the Borrower's request, on a regional or country options, basis, and will disseminate knowledge on R&Rpolicy processes, legal issues, compensation approaches, participatory and lessons of experience from best practices in carrying out resettlement. Improved Review Procedures. Projects that entail involuntary (f) ensure resettlement require special care and review to compliance with the Bank operational directives. Henceforth, stage of processing of such projects, country at the initial project proposed R&ER departments will be required to highlight or in relation to past R&Rexperience in the country features Regional concerned in consultation with the local jurisdictions the Environment Department staff. Special Technical and attention will be given during pre-appraisal to the adopted by Government for consultation with arrangements affected people and to assessment of the Government's R&R implementation capacity. Any divergence of view at that stage and, if will be flagged to the relevant Regional Loan Committee unresolved at that level, to the Bank-wide Loan Committee.  Attachment2 Page 3 Bank will (g) Policy Dialosue. Through its policy dialogue, the encourage Borrowers with ongoing or proposed Bank-supported issue domestic operations involving resettlementto develop and by policy and legal frameworks for resettlementcaused and development projects. In formulatingsector strategies review to sector investment priorities, Borravers vill be asked energy. ezperience with resettlement in key sectors such as Ci) agriculture, infrastructureand urban, in order to to avoid or ascertain whether priority investmentsare designed minimiz, resettlement, and (ii) develop acceptable approaches of to implementing R&1, which may be an unavoidable consequence certain projects. (_  AttachmenLI SUGGESTIONSFOR IMPROVEMENTOF OUALITY AT ENTRY FOR R+R 1. Imorove SubstantiveKnowledge through Guidelines/BestPractice Dissemination: * on consultationprocesses with people affected by R+R; * on land acquisitionand related compensation; * on restoring income to at least earlier, "pre-project"levels; * on transitioncosts (which may be a multiple of eventual [annual] income levels); * on relevant training for people affected; * on resettlementon marginal land (especiallyrelevant for tribal people living on marginal lands). 2. Need for Resource Book and Training of Staff in above (and other Relevant) R+R Related Subiects: * such efforts should be *dovetailed'with knowledge from NGOs; e relevance of eventual trainingmaterial (and training technologies)to people affected (e.g., videos?). 3. Review Process: review at Regional Loan Committee stage, i.e., just prior to appraisal, is too late for effective handling of R+R. Need for consistent review of R+R projects/components at IEPS stage, includingwith ENV resettlementadvisors; in projects with especially sensitiveR+R issues (as identifiedat the IEPS stage), a small team of highly experienced Bank officers may be assigned to review these issues with a view to putting such project to the Bankwide Loan Committee for a decision as to whether to go ahead with such project. -- 7 -7W