Home | Databases | WorldLII | Search | Feedback Australian Indigenous Law Reporter

Editors --- "Implementation of Operational Directive 4.20 on Indigenous Peoples: An Independent Desk Review - Digest" [2003] AUIndigLawRpr 14; (2003) 8(1) Australian Indigenous Law Reporter 113

International Developments

Implementation of Operational Directive 4.20 on Indigenous Peoples:

An Independent Desk Review

World Bank

Report Number 25332

January 10 2003

Acronyms & Abbreviations

ADB Asian Development Bank

AFR Africa Region

BP Bank Procedure

CAS Country Assistance Strategy

CDF Comprehensive Development Framework

CODE Committee on Development Effectiveness

DANIDA Danish International Development Agency

DCA Development Credit Agreement

EAP East Asia and Pacific Region

ECA Europe and Central Asia Region

ES Evaluation Summary

ESSD Environmentally and Socially Sustainable Development Network

ESW Economic and Sector Work

GEF Global Environment Facility

ICR Implementation Completion Report

IDB Inter-American Development Bank

IEPS Initial Executive Project Summary

ILO International Labor Organization

IP Indigenous Peoples

IPDP Indigenous Peoples Development Plan

IRIS Integrated Record Information System

LCR Latin America and the Caribbean Region

LIL Learning Innovation Loan

M&E Monitoring and Evaluating

MDBs Multilateral Development Banks

MENA Middle East and North Africa Region

MTR Mid-Term Reports

NGO Nongovernmental Organization

OD Operational Directive

OED Operations Evaluation Department

OEDCR OED Country Evaluation and Regional Relations

OMS Operational Manual Statement

OP Operational Policy

PAD Project Appraisal Document

PCD Project Concept Document

PPAR Project Performance Assessment Report

PSR Project Status Report

QAG Quality Assurance Group

QACU Quality Assurance and Compliance Unit

SAR South Asia Region

SAR Staff Appraisal Report

SIF Social Investment Fund

TTLs Task Team Leaders

UN United Nations

WDR World Development Report

...

Table of Contents

Glossary

1. Introduction and background

Evaluation background and context

Bank’s policy on Indigenous Peoples

Issues related to the evaluation

2. Relevance of Operational Directive 4.20

Long-term and holistic

Client ownership

Partnership

Result orientation

3. Implementation of OD 4.20

Coverage of intended beneficiaries

Protecting interests of IP

Monitoring and evaluating project results

4. Findings and recommendations

Findings

Recommendations

Bibliography

Boxes and Tables

Box 3.1: Who are IP for purposes of bank assistance?
Box 4.1: Good practices at the design stage
Box 4.2: Results of a staff survey conducted by OED

Table 2.1: measuring relevance of OD 4.20 goals and objectives
Table 3.1: Implementation of OD 4.20 at a regional level
Table 3.2: Implementation of OD 4.20
Table 3.3: Quality of OD application
Table 3.4: Improving quantity and quality of beneficiary participation

Attachments

Attachment I: management response
Attachment II: project classification methodology
Attachment III: operational directive 4.20
Attachment IV: sample countries
Attachment V: terms of reference for panel of experts
Attachment VI: report from code
Attachment VII: OED’s presentation at the code meeting

...

1. Introduction and background evaluation background and context

1.1 This evaluation assesses how the Bank has implemented its Operational Directive (OD) 4.20. The directive mandates special measures to protect the interests of Indigenous Peoples (IP). The evaluation examines Bank assistance, namely investment projects and analytical work, in 34 countries^[1] between January 1992 and May 2001. The evaluation questions are:

• To what extent is the OD relevant in delivering effective development to IP?
• To what extent has the OD been applied and how?
• To what extent has the OD been efficacious in achieving its objectives?

1.2 The evaluation is being phased. This report focuses on the first two questions (partially on the relevance aspect and on the implementation of OD 4.20). A second-phase report to be delivered before June 2003 to the Committee on Development Effectiveness (CODE), will assess the results of Bank-financed projects in a small set of selected countries.

1.3 It is worth stressing that this review assesses only the implementation of OD 4.20 in social groups affected by Bank-financed projects and that meet to varying degrees the five characteristics stated in the OD. The resulting findings are without prejudice to the status of these groups as IP under their domestic jurisdiction or under international law.

Bank’s policy on Indigenous Peoples

1.4 The Bank was the first multilateral agency to recognize the need to provide special protection to tribal groups. It did so as early as 1982. The Operational Manual Statement (OMS) 2.34 (1982), which was the Bank’s first guidelines on IP, covered mainly tribal groups—’ethnic groups with stable, low-energy, sustained-yield economic systems as exemplified by hunter-gatherers, shifting or semi-permanent farmers, herders or fishermen.’ In 1987, an internal review found that the OMS was usually applied in projects affecting small, isolated, and unacculturated tribal societies and was not applied to many other groups who needed protection but were acculturated.^[2] The review suggested a need to shift attention from a definition of tribals by their isolation and acculturation to a definition that focused on their socio-cultural systems, modes of production, and forms of ecological adaptation, which are different from that of dominant societies (Davis 1993). A shift was also considered necessary to align the Bank’s policies with international thinking on the rights of IP as well as the general trend to recognize the social and economic rights of poor and marginalized peoples throughout the world (Davis 1993, World Bank 1986).^[3] In 1991, the Bank issued OD 4.20 on IP (see Attachment III for OD 4.20). It aimed to ensure that ‘the development process fosters full respect for the dignity, human rights, and cultural uniqueness’ of IP.

1.5 OD 4.20 states that ‘for an investment project that affects IP, the Borrower should prepare an Indigenous Peoples Development Plan (IPDP)’ (para. 13 of OD 4.20). It is the Bank’s responsibility to identify IP through a preliminary examination of the Borrower’s law, policies, and procedures, and through anthropological and sociological studies where necessary. Task Managers are then responsible, in consultation with regional environment staff, for signaling IP issues and any special action needed, including an IPDP, in the Initial Executive Project Summary (IEPS). When the bulk of the beneficiaries are IP, ‘the Bank’s concerns would be addressed by the project itself and the provisions of the OD apply to the project in its entirety.’

1.6 The IPDP aims to mitigate the potential negative project impact on IP and to ensure that beneficiaries ‘receive culturally compatible social and economic benefits.’ The preparation of the IPDP is the Borrower’s responsibility and it is due for Bank review before project appraisal. The OD states that ‘cases will occur, especially when dealing with the most isolated groups, where adverse impacts are unavoidable and adequate mitigation plans have not been developed. In such situations, the Bank will not appraise projects until suitable plans are developed by the Borrower and reviewed by the Bank.’ The IPDP needs to be based on a comprehensive diagnosis of the socioeconomic context within which the IP operate and on their informed participation. The IPDP addresses the following dimensions as needed: the legal framework and land tenure issues, a strategy for local participation, proposed measures to mitigate any adverse impact and to ensure that IP receive equitable project benefits, measures to strengthen institutional capacity to implement the IPDP, measures for monitoring and evaluation, cost estimates, and a financing plan.

Issues related to the evaluation

1.7 As with all ODs, OD 4.20 is a mixture of policies, recommendations, and good practices. The language of OD 4.20 is broad and subject to various interpretations. It is, therefore, important to clarify the evaluation’s positions in three key areas: First, which groups should be considered to be IP for the purposes of the evaluation? Second, how should the evaluation interpret the term ‘affects’? and Third, under what circumstances should the evaluation consider that the OD has been applied?

1.8 OD 4.20 requires special action where ‘Bank investments affect indigenous peoples, tribes, ethnic minorities, or other groups whose social and economic status restricts their capacity to assert their interests and rights in land and other productive resources.’ It uses the term IP to describe ‘social groups with a social and cultural identity distinct from the dominant society that makes them vulnerable to being disadvantaged in the development process’ and states that these groups may be referred to as ‘indigenous peoples,’ ‘indigenous ethnic minorities,’ ‘tribal groups,’ or ‘scheduled tribes.’^[4]

1.9 Which groups should be considered IP for the purposes of the evaluation? Since no single definition can capture all these groups, the OD states that social groups to be covered ‘can be identified in particular geographical areas by the presence in varying degrees of the following characteristics: (a) close attachment to ancestral territories and to the natural resources in these areas; (b) self-identification and identification by others as members of a distinct cultural group; (c) an indigenous language, often different from the national language; (d) presence of customary social and political institutions; and (e) primarily subsistence-oriented production.’ These characteristics, derived from OMS 2.34 on tribal people, focus on the socio-cultural systems, modes of production, and forms of ecological adaptation, that are different from those of dominant societies.^[5] The OD states that ‘Task managers (TMs) must exercise judgment in determining the populations to which this directive applies and should make use of specialized anthropological and sociological experts throughout the project cycle.’

1.10 The OED team found a range of views among Bank staff and managers on how the OD has been applied. Some felt that only indigenous groups are covered by the OD, others that tribal groups are also covered. Yet others felt that all social groups (indigenous, tribal, and ethnic minorities) who met to varying degrees the five characteristics stated in the OD are covered. Management clarified the official interpretation of the policy, explaining that the OD applied only to ‘social groups who meet the five characteristics’ to varying degrees. The evaluation finds that this flexible interpretation is broadly in line with Bank practice over the past decade^[6] and explicit understandings with some Borrowers such as China, India, and Indonesia.

1.11 The evaluation uses the same approach — the classification of projects was gleaned from actual Bank practice rather than from an independent attempt to ascertain the importance of each of the five characteristics. This was done in the following manner. First, the evaluation applied the domestic approach where it exists and has been accepted by the Bank.^[7] Second, where no approach at the country-level has been agreed, the evaluation examines project documents to identify any underlying practice.^[8] Third, where there is neither an agreed domestic approach nor consistent practice, the evaluation uses a definition consistent with the domestic legal/administrative framework for indigenous or tribal peoples.^[9] Finally, where none of the above approaches can be used, the evaluation exercises judgment to determine whether social groups affected by the project meet to a sufficient degree the five characteristics stated in the OD (see Attachment II).^{[10] [11]}

1.12 When are IP ‘affected’ by Bank assistance? The evaluation interpreted the term ‘affected’ to apply when social groups with characteristics stated in the OD are in the project area and may be directly harmed by the project activity and/or where they are among the intended beneficiaries.^[12] Some external stakeholders view this interpretation as too narrow.

1.13 When did the evaluation consider the OD applied? The evaluation team first separated those projects where the appraisal documents identified the presence of IP. In these projects, the evaluation looked for projects with self-standing IPDPs. It then looked for projects with elements of an IPDP. Such elements were derived from OD 4.20, namely: sound diagnosis of issues related to IP, participation of IP in project design and implementation, measures to protect the interests of IP, and monitoring indicators for IP-related results. If the project had any of these elements to varying degrees, it was considered to have an IPDP. In addition, the evaluation considered the OD applied in those projects that included some measures to protect IP, even if these were inadequate to be considered as IPDPs. Once again, some external stakeholders consider these assumptions too lenient.^[13]

1.14 In the remaining projects, based on its review of the 34 sample countries, the evaluation identified projects where IP were affected but the OD was not ‘applied,’ that is, the presence of IP groups in the project localities were not identified and no measures were taken to protect IP interests. In this category, the evaluation examined other project documents: IEPSs, loan agreements, and Implementation Completion Reports (ICRs). OED sent an initial list of the projects for comment to ESSD staff in October 2001 and again between March and June 2002. A panel of internationally recognized experts was also asked to review the classifications (see Attachment V for the terms of reference and panel composition). In selected cases, OED consulted with external stakeholders. Finally, OED took into account Management’s objections to the classification of 10 projects. Consequently, five projects were shifted to a ‘Not Rated’ category, and five others in SAR and EAP regions were shifted to a ‘Not-Applicable’ category.

1.15 The remainder of this report is organized as follows: Chapter 2 discusses some aspects of the relevance of OD 4.20. Chapter 3 presents findings on the Bank’s experience in implementing OD 4.20. The conclusions and recommendations are found in Chapter 4. Attachment I contains the Management Response. Attachment II provides information on the evaluation’s methodology. Attachment III reproduces OD 4.20. Attachment IV compiles a list of the countries in the evaluation sample and indicates the regulatory structures related to IP in each of them. Attachment V contains the terms of reference of OED’s panel of experts. The CODE discussion held on September 18, 2002 appears in Attachment VI. OED’s presentation at the CODE meeting appears in Attachment VII. In addition, a few background papers are available upon request. Background Paper I presents the detailed results of the review for the 34 sample countries. Background Paper II compiles the list of 234 closed projects reviewed for this evaluation, and Background Paper III is based upon the deliberations of an OED/India Department workshop (December 3-5, 2001).

...

3. Implementation of OD 4.20

3.1 This section assesses the extent and quality of implementation of the OD. The evaluation reviewed the universe of 234 projects that were appraised after January 1, 1992 and closed before May 31, 2001 in the 34 sample countries.^[26] In addition, the evaluation undertook a rapid review of the last five projects approved through FY2001 in each of the 34 countries to understand whether there has been any improvement in OD implementation.

3.2 The evaluation focused on three aspects:

• Coverage of intended beneficiaries. (To what extent did the Bank identify that IP, as covered by the OD, were affected by the projects?)
• Protecting interests of IP. (To what extent did the Bank base its assistance on sound diagnosis? Did it include satisfactory IPDPs? Did it involve informed participation of IP?)
• Monitoring and evaluation of project results. (To what extent did projects include measures to monitor and evaluate results related to IP?)

Coverage of intended beneficiaries

3.3 Projects that ‘affected’ IP. The evaluation found that 89 projects (US$9.5 billion) of the 234 reviewed (US$27 billion) could have affected IP. Attachment II explains the specific criteria and processes for determining whether a project affected IP, including references in project documents and significant presence of IP in the project locality. The project classifications were discussed with regional staff and interested nongovernmental organizations (NGOs). In addition, they were reviewed by a panel of experts established for this purpose (see Box 3.1).

3.4 Projects that applied OD 4.20. Overall, the evaluation found that the OD was applied in 62 percent of the 89 projects that affected IP (55 projects for US$7 billion). Table 3.1 shows the regional distribution of these projects. In these 55 projects, IP groups were identified, and some measures integrated to protect their interests.^[27] In four of these 55 projects, the appraisal documents indicated that IP-related issues had been carefully considered and they confirmed that no further action was necessary.^[28]

Table 3.1:
Implementation of OD 4.20 at a regional level

3.5 Projects where the evaluation found that the OD should have been applied, but was not. In the other 38 percent of the 89 projects (34 projects for US$2.4 billion), the evaluation concluded that there was prima facie evidence that IP were affected, but were not protected by the application of the OD. This list originally included nearly 48 projects, but this number was reduced to 39 through discussion—sometimes spirited—with Management and staff, through consultation with external stakeholders in selected cases, and through review by the OED panel of experts. In the final review, Management disagreed with the OED finding that IP were affected in five projects; these were then shifted to a special ‘not-rated’ group, leaving 34 projects in this category.^[29] Of these 34 projects, 27 included IP as beneficiaries and only seven may have negatively affected them.

3.6 Out of the 89 projects that affected IP, the evaluation concluded that activities were likely to have an adverse impact on IP only in 25 (US$3.9 billion), while in the remainder, the IP were potential beneficiaries. Only 11 of the 25 projects, however, had IPDPs or elements thereof (US$1.6 billion). Self-standing IPDPs were present in only eight, and in another four, there were elements of an IPDP (The nature of these elements is explained in Chapter 1). In the other 14, there were inadequate measures to mitigate the adverse impact of the project activities.^[30] However, this does not mean that the evaluation can conclude that the project adversely affected IP or that IP did not benefit from the project. To arrive at this conclusion, there is need for fieldwork, which will be undertaken in Phase II.

3.7 The review of more recent open projects indicates that there has been improvement in the quality of coverage in projects approved after FY1998 (see Table 3.2). Institutional arrangements for monitoring OD implementation changed after FY1996, when technical departments and subsequently the regional networks, were charged to ensure proper implementation. The Strategic Compact in FY1997 also provided resources marked for social scientists who play an important role in implementing this policy. These institutional changes took some time to have an impact. To check whether implementation has improved more recently, the evaluation reviewed 170 open projects, 140 of which were approved after FY1998. The review of these projects indicates that the application of OD 4.20 has remained at the same level, but that more than 95 percent of the projects likely to have an adverse impact on IP included IPDPs or elements of IPDPs. On a regional basis there is considerable improvement in the LCR and EAP regions, with the OD being applied in more than 85 percent of the projects that affected IP. In the projects where the OD has not been applied, IP are mainly beneficiaries.

Table 3.2:
Implementation of OD 4.20

3.8 The evaluation finds remarkable flexibility in Bank practice in the interpretation of the term ‘social group’ used in OD 4.20 to identify IP. For example, in China the evaluation sample indicates that the Bank works within the definition of 55 national minorities, considering such groups as IP.^[31] Several loan agreements for Indonesian projects define IP as those social groups that: first ‘have a distinct social and cultural identity’ and second, ‘are susceptible to being disadvantaged in the development process induced by the Project or any part thereof.’^[32] Similarly, loan agreements in Vietnam define the term ‘Ethnic Minorities’ as social groups that are disadvantaged in the development process and that have social and cultural identities distinct from those that constitute the predominant ethnic group in Vietnam. In some recent cases in India, not only scheduled tribes, but also backward or scheduled castes who meet the five characteristics to varying degrees have been covered by the OD.^[33] In Pakistan, a more exclusionary approach is adopted. Although tribal populations in many parts of the country meet to varying degrees the five characteristics stated in the OD,^[34] the Pakistani Constitution has conferred a special status on tribal groups to protect their customary and political institutions, and the government has signed ILO Convention 107, the Bank has preferred to take an exclusionary approach and not consider them as IP under the OD. In LCR, some of the rural Afro-Latino descendants are treated similarly to IP, because these groups possess certain characteristics (especially in terms of their attachment to lands and natural resources) similar to those of IP and could have been excluded from project benefits if special measures had not been taken in Bank-financed projects.^[35]

3.9 The OD states that ‘Task managers (TMs) must exercise judgment in determining the populations to which this directive applies and should make use of specialized anthropological and sociological experts throughout the project cycle.’ The evaluation found only limited projects where documents indicate that anthropologists or sociologists were involved during project preparation. There was also little or no discussion in IEPSs on this issue. Perhaps as a result, no consistent practice has emerged on whether a social group needs to meet all five characteristics to varying degrees or to meet just a few of the five characteristics fully. In some projects, once IP are identified under the country’s legal framework, the focus in applying the OD seems to be on remoteness and poverty.

3.10 The evaluation finds that IP are more systematically identified in countries with legal provisions related to IP or when IP-related issues are a part of the countries’ public policy debates.such as in China, India, and several LCR countries. In AFR and MENA regions, the evaluation sample of six countries did not find any sample projects that recognize IP, although the evaluation found the presence of social groups that met to varying degrees the five characteristics mentioned in OD 4.20 in at least three countries. In the AFR region, for example, an initiating memorandum was drafted to prepare a set of regional guidelines for the identification of IP. However, given the impending revision of the OD, which began in 1996, the initiative was suspended. More recently, the OD has been applied in some recent Global Environment Facility (GEF) projects and the Chad-Cameroon Pipeline project.

3.11 The evaluation’s findings on application of the OD are consistent with several self-assessments that were undertaken by the regions despite the fact that OED’s sampling was purposeful and they selected those projects, that either affected IP or had likelihood of doing so. An internal review of 72 projects affecting IP in the LCR region found that the policy was slow to take off in the period between 1992 and 1997, but had improved thereafter. According to this report, the degree of OD implementation was much higher in the 24 projects under preparation in 1997, where several projects intended to incorporate an IPDP. A previous internal review of the application of OD 4.20 in the EAP and the SAR regions between 1987 and 1994 found that although all 48 projects reviewed included IP or ethnic minorities, only 18 projects (37 percent) acknowledged existence of IP and the applicability of the OD.

Protecting interests of IP

3.12 The evaluation examined the 55 projects where the OD was applied to assess how the Bank protected the interests of IP based on three criteria: (i) soundness of underlying diagnosis, (ii) the quality of the IPDP or elements thereof, and (iii) the extent of IP participation. Overall, 58 percent of the projects that applied the OD were assessed to have taken measures that were satisfactory or highly satisfactory. This improved significantly in the later set of open projects, with 77 percent of the projects having a satisfactory or highly satisfactory rating. The findings are summarized in Table 3.3.

Table 3.3:
Quality of OD application

HS: Diagnosis + participation + measures to protect IP interest + monitoring indicator on impact or outcome.

S: Diagnosis + participation + measures to protect IP interest. No indicators.

MS: Participation + measures to protect IP interest. No Diagnosis or indicators.

MU: Lower levels of participation only in implementation + Measures to protect IP interest. No Diagnosis or indicators.

U: References to IP + Some analysis or a measure directed at IP. No participation, Diagnosis or indicators.

HU: References to IP, but nothing else.

3.13 Soundness of diagnosis. Sixty percent of the project documents that identified the presence of IP included some efforts to diagnose IP-related issues, although in 78 percent of the project documents there is poverty analysis that may indirectly deal with IP issues.^[36] Overall, diagnosis of issues related to IP needs to be strengthened and made an integral part of project preparation. Increasingly, as noted in Chapter 2, country-level analysis of issues is being undertaken, and that analysis is then supplemented at a project-level.

3.14 Quality of IPDP or elements thereof. Normally, an IPDP is prepared by the Borrower, and the evaluation finds that this process enhances client ownership of the plan. On the other hand, the extent of Borrower ownership is not as clear in projects where the evaluation only finds some of these elements in project documents. Only 58 percent (32 projects) of the 55 closed projects that applied the OD had IPDPs, elements thereof, or the necessary design features.^[37] Twenty-two percent (12 projects) had self-standing IPDPs and another 31 percent (17 projects) had elements of an IPDP. In five percent (three projects), the bulk of the beneficiaries were considered as IP, and no IPDP was required because the OD applied to the project in its entirety.^[38] The other 23 projects (out of the 55) included one or more specific measures or provisions for IP as required by the OD, but these were not considered sufficient to warrant description as IPDPs. There is significant improvement in the set of open projects, however, in that 77 percent of the open projects that applied the OD had IPDPs or elements thereof. Thirty-five percent have self-standing IPDPs.

3.15 During the evaluation period, there seems to be little difference in the application of the OD to projects that adversely impacted IP and those where IP were beneficiaries. For example, 32 projects had IPDPs or elements thereof; only 11 of these projects were likely to adversely impact IP, and the remaining 20 projects were those where IP were mainly beneficiaries. However, out of the 12 self-standing IPDPs, eight were in projects that adversely impacted IP and four were in projects where IP were beneficiaries.

3.16 Only a handful of so-called ‘process projects’^[39] included strategies to address issues related to IP. Among these, the Peru Second Social Investment Fund has dealt well with these issues. The appraisal documents for two emergency projects (Ecuador-November 1997; Peru-December 1997) are silent on IP issues, but the legal agreements require the Borrower to employ a team of experts under terms of reference satisfactory to the Bank to prepare IPDPs for relevant sub-projects. In these cases, design and diagnosis were postponed to the implementation stage (while the OD requires these steps to be undertaken prior to appraisal). Similarly, most financial intermediary projects that operate in IP locations do not have any references in any documents to the potential impact on IP.

3.17 A review of Bank practice also indicates that the links between the different social policies/directives were not clear during most of the evaluation period. For example, is there a need for a separate IPDP when there is a resettlement or an environmental action plan, required under the other Bank policies? Current guidance on the web page <http://lnweb18.worldbank.org/ESSD/essdext.nsf/47ByDocName/EnvironmentalAssessment> for implementing the policy on environmental assessments suggests that issues related to social policies such as the one on IP, are dealt with separately.^[40] In practice, however, the evaluation did not always find such separate treatment, which sometimes resulted in a failure to implement the provisions of the OD effectively.^[41]

3.18 The evaluation found that the OD was not applied in at least 10 urban development projects in the sample, although in most it was clear that IP would be among the population affected by the projects. Management believes that these projects are not covered by the OD because the affected populations do not meet in varying degrees the five characteristics stated in the OD. In line with Management’s views, this evaluation classified these projects as not-applicable, that is, they were not included in the group of 89 projects considered to affect IP for the purpose of OD 4.20.^[42] The proposed OP takes a slightly different view from OD 4.20 and exempts the policy from applying to groups who: (a) have left their communities of origin, (b) moved to urban areas, and (c) and/or migrated to obtain wage labor.

3.19 Paragraph 13 of the OD states that when the ‘bulk’ of the direct project beneficiaries are IP, the Bank's concerns would be addressed by the project itself and the provisions of the OD would thus apply to the project in its entirety. This provision has been interpreted in different ways. For example, the India Bihar Plateau Development project does not have a separate IPDP perhaps because it states that 50 percent of the project beneficiaries belong to tribal groups. On the other hand, the Implementation Completion Report (ICR) for the Cambodia Social Fund project states that in view of the national scale of the project and the small population of IP, the project does not have an IPDP. In the Pakistan Balochistan Natural Resource Management project, the appraisal document indicates that the bulk of the people affected will be tribals, meeting to varying degrees the five characteristics stated in the OD. The project does not have an IPDP, although it is clear that the customary access to land and natural resources of a number of IP groups will be restricted through environmental protection measures. The review of open projects also indicates some similar approaches. The PAD of the Nepal Rural Infrastructure project states that, ‘virtually all inhabitants of Nepal can be considered to be indigenous’ and adds that therefore the project has been designed to ensure that its benefits reach the poorest group.

3.20 Paragraph 20 of the OD requires that ‘the Borrower's commitments for implementing the IPDP should be reflected in the loan documents.’ The evaluation examined the 55 projects to understand how this requirement was addressed. Eleven out of the 12 projects with self-standing IPDPs satisfactorily reflected IP issues in their legal documents and included covenants or other measures indicating the commitment of the Borrower or the project entity to carry out the IPDP. In projects where there are only elements of IPDPs, about 50 percent of the documents reflected some measures in loan documents. OED understands the OD provision as requiring legal documents to set out Borrower commitments on project implementation arising from the OD, regardless whether there is a self-standing IPDP, elements of an IPDP, or some measures thereof. Management takes a different interpretation of the OD provision.^[43] This suggests the need for greater clarity as to how and when the Borrower’s commitment to implement the project consistent with OD 4.20 should be reflected in loan documents.

3.21 Extent of informed participation. The evaluation examined the extent of ‘informed participation’ as reflected in project design. There is a marked increase in the participation of IP beneficiaries in the design and implementation of projects where the OD was applied, compared with those where the evaluation concludes that IP were affected but the OD was not applied (see Table 3.4). Even among the projects that applied the OD, however, participation of IP in decision-making and in financial management is still low. Their participation may have been constrained by country regulations that control the use of public funds. On the other hand, participation is high in countries without such constraints, for example, in Bolivia, which passed a law in 1994 for the promotion and consolidation of the participatory planning process of IP groups and rural and urban communities. This provided a framework for substantial participation in the case of the Bolivia Rural Communities Development project. In addition, Bank procurement guidelines used to make it difficult for IP (and community groups) to participate in project-financed procurement. These guidelines were eased in 1995 with the introduction of ‘community-related procurement.’

Table 3.4:
Improving quantity and quality of beneficiary participation

3.22 As the results of field assessments indicate, capacity-building of IP groups and their representatives is critical if the participation of IP is to be meaningful. However, only 52 percent of the projects that applied the OD involve some form of capacity-building of IP for the achievement of project objectives. Although lower than desirable, this level of capacity-building is much higher than the 33 percent of projects where IP were affected, but the OD was not applied (see Table 3.4).

3.23 Additionally, the evaluation notes that since 1994 there are several initiatives to build capacity of IP organizations and government agencies in LCR through the use of Institutional Development Funds. Through these country-level programs (14 initiatives in 11 countries), a large number of local organizations were trained in ethno-development, project design and evaluation, organizational development, negotiation skills, and indigenous laws in a gender-aware manner. The links between such training and participation of those trained in Bank-financed projects need to be strengthened, however, if such training is to produce needed outcomes (World Bank 2000).

3.24 During the second phase—field assessments—the evaluation will ascertain whether (a) affected IP were provided with relevant information in a comprehensible manner; (b) their views were solicited on relevant issues; (c) the consultation methods were culturally appropriate; and (d) their views were considered in making final decisions on project design and implementation. It is not possible to base such assessments on a desk review. Initial results from field assessments in Andhra Pradesh, India, indicate that while ‘informed participation’ is an important goal, it is difficult to implement effectively in contexts where traditional systems of disseminating information have either broken down or been replaced by modern state structures; where there are high levels of illiteracy among IP communities; and where the IP are part of a more heterogeneous community. Informed participation is a key tool in the implementation of the policy but it is very difficult to implement effectively. Thus, there may be a need for other tools to achieve the policy objectives.

Monitoring and evaluating project results

3.25 The evaluation assessed the 55 projects that applied the OD in terms of the project’s M&E framework, the extent to which relevant issues were monitored during project implementation, and attention to relevant IP issues in ICRs and other reviews undertaken by different Bank units. The evaluation also assessed the institutional arrangements for monitoring the implementation of OD 4.20.

3.26 Forty-five percent of the PADs included monitoring indicators to measure the impact of the assistance on the poor, but only 20 percent included specific indicators to measure the results or outcomes for IP. The prevalence of monitoring indicators has improved in the open projects, with 55 percent of the projects having one or more indicators for IP.^[44] OD 4.20 requires that ‘legal provisions should provide Bank staff with clear benchmarks that can be monitored during supervision.’ Fifteen loan agreements included reference to monitoring indicators or the need to develop them, or to the IPDP. Others were silent on the issue.

3.27 Project Status Reports (PSRs), prepared after each implementation support mission, do not systematically track effectiveness of actions related to IP in the projects that have applied the OD. Forty-four percent of the 41 PSRs reviewed provided satisfactory or marginally satisfactory information on IP-related issues or indicated that such information was available in other documents.^[45] About half a dozen PSRs paid due attention to the section on Compliance with Safeguard Policies. Task teams regularly left the section on OD 4.20 blank or marked ‘NA,’ even in PSRs prepared in FY2000. The evaluation also examined 37 Mid-Term Reports (MTRs). Twenty-nine percent of the MTRs dealt with these issues satisfactorily, by focusing on achievement of objectives and monitoring indicators specifically related to IP.^[46]

3.28 The evaluation examined 47 ICRs^[47] available for the projects where the OD was applied. Only 25 percent provided information on outcomes for IP, although almost 60 percent made passing references. OED has not done a good job of monitoring these issues in its ICR reviews. There were 24 evaluation summaries (ESs) available for projects with IPDPs or elements thereof. Of these, OED identified IP as part of the objectives in only seven, while six evaluation summaries highlighted IP-related achievement in discussion of outcomes. Four ESs included lessons learned in relation to IP. Only a single ES (Bolivia Second Social Investment Fund project) pointed out that the ICR was deficient in not discussing IP issues. The evaluation assessed four available Project Performance Assessment Reviews (PPARs) for projects that affected IP. They do not focus on IP issues, even in projects where ICRs discuss the issue.

3.29 The evaluation examined the institutional procedures for monitoring the application of OD 4.20. For example, in the case of the environmental safeguard policy, a mandatory screening process conducted in the regions allows operations to determine the likely magnitude of potential adverse environmental impacts of all investment projects. In the case of OD 4.20, there does not seem to have been any such rigorous mandatory screening in place at the Bank level for most of the evaluation period, although some regions required an initial social analysis and consideration of IP issues. More recently, some regions such as SAR, LCR, and EAP, have introduced the use of a social data sheet or upstream project reviews as part of its PCD and PAD review.

3.30 Over the last few years, the Bank’s Quality Assurance and Compliance Unit (QACU) has helped to prepare and integrate systems that requires task teams to identify and address IP issues upstream in the project cycle and explicitly assesses compliance with the policy at the time of project entry.^[48] Similar modifications in the PSR format require project supervision to rate implementation of agreed IPDPs throughout the project. QACU has also helped establish an institutional ‘safeguards compliance monitoring and reporting system’ to monitor compliance with the IP policy. Although no systematic data is available as yet, these new institutional procedures may have been a contributing factor to the improving trend in the application of the OD. Between 1997 and 2000, the Quality Assurance Group (QAG) reviewed 15 out of the 55 projects that applied OD 4.20. (Thirteen reviews looked at the quality of supervision and two at the quality at entry). Only five of the ensuing reports referred to IP. It should be noted that only since FY2000, QAG’s questionnaires for the quality of supervision reviews included questions pertaining to the Bank safeguard policies, specifically in relation to OD 4.20.

4. Findings and recommendations

Findings

4.1 Overall, this evaluation indicates that the OD has positively influenced Bank assistance in many countries—particularly in India, Indonesia, and in several countries in the LCR region—in focusing on the marginalized poor (see Box 4.1). Beneficiary participation is higher in projects that have applied the OD than in similar projects that have not. In addition, the quality of application of the OD seems to be positively correlated with better results for poverty reduction, although this is not conclusive.

4.2 While there is clear evidence of progress over time, the evaluation found that the OD had not been applied in a consistent manner in the sample projects. Only in slightly more than half of the projects that affect IP have IP been identified and measures integrated to protect their interests. Out of these, only 22 percent had separate IPDP and another 36 percent had elements of IPDPs. In general, the evaluation finds that projects where the Borrower prepared a self-standing IPDP have better outcomes, which illustrates the importance of Borrower ownership in protecting the interests of IP. In the other projects, application of the OD was of low quality.

4.3 In general, the Bank has encouraged country ownership by adapting its approach to that of the Borrower. As a result, it has been able to encourage the application of the OD in some countries where it operated within the domestic legal frameworks and used the domestic terminology (sometimes without the term ‘indigenous peoples’). On the other hand, in countries where the domestic legal system does not recognize or address IP-related issues, country ownership is often limited. This is the case in Morocco and Ethiopia, where—given the historical and social contexts—there is little consensus on which, if any, social groups are indigenous.

4.4 Thus, the evaluation finds that the Bank found it technically difficult to identify IP when operating outside the legal framework of a Borrowing country. This difficulty explains to a significant degree the uneven application of the OD. Looking forward to the new OP, the Bank will need to clarify how it will deal with this issue to ensure consistency and transparency in the application of the policy. For example, the new OP will need to address many of the following questions: To be covered as IP, do vulnerable groups need to meet all or only some of the five characteristics to varying degrees? Is it realistic for the policy to require that IP should be primarily subsistence-oriented? The field assessments even in remote parts of Andhra Pradesh in India indicate tribal communities are increasingly integrating into the market economies. Is it important that the group speaks an indigenous language? The Guatemala field assessment indicates that some IP groups were now speaking and using Spanish as a matter of choice. What is an indigenous language? In India the OD has been applied to vulnerable groups who speak a language different from that spoken by the indigenous groups in the area, given that the former migrated from a different part of the country.is their language an indigenous one? Would the policy cover pastoralists or migratory tribal groups, some of whom may not have attachment to any particular ancestral land? There are no hard and fast rules in many of these gray areas, but some levels of clarity would indeed be necessary for staff who are expected to implement the OD.

4.5 The OD requires the provision of culturally compatible benefits, as well as the avoidance or mitigation of potentially adverse effects on IP, that is, both the safeguard and benefits provisions must be applied simultaneously and in direct consultation with IP. In doing this, the policy shifts to an approach that is more difficult to ensure and to assess ex post, posing a particular challenge when part of a social OD.

4.6 The evaluation found no clear understanding in Bank documents or practice of the term ‘affect’ which triggers the application of the OD, that is, whether it refers to direct or indirect effects, as well as if it refers to both positive and/or adverse effects. This has resulted in a lack of consistency in the application of the OD to Bank projects. It has also led to a disconnect between Bank practice and the expectations of external stakeholders, namely IP groups and NGOs, as to which projects are covered by the OD.

4.7 The quality of monitoring implementation of the OD is low, whether it is during project preparation, supervision, or at completion. Bank efforts had been inadequate to meet the ambitious objectives of the OD. Review of the more recent set of open projects, however, indicates that there is significant progress in the implementation of the OD and that the OD has been applied to over 60 percent of projects that affect IP and to 90 percent of those that could have harmed IP.

4.8 A staff survey conducted by OED suggests that there is still confusion in understanding the OD and its requirements (see Box 4.2). The survey also indicates that there is a perception among respondents that task teams do not have adequate resources to implement the OD. At the same time, the evaluation team could not find data on the IP- related financial and administrative costs of doing business, either for the Bank or for the client country.

4.9 The OD lists a set of wide-ranging prerequisites that may need to be considered in preparing IPDPs. Several aspects (such as assessing of constitutional provisions and domestic legal systems and addressing deficiencies in land tenure laws, capacity of government institutions, and so on) may be better addressed at a country-level rather than at a project-level. Individual task teams do not have necessary leverage to address these critical issues and this has led to inefficiencies in the application of the OD.

4.10 The evaluation finds that many Borrowers have poor institutional environments for the informed participation of IP in development planning and implementation. These are issues that are better addressed at the country, rather than at the project-level.

4.11 The approaches of different donors, including that of multilateral agencies, are not harmonized, reducing the potential for genuine partnerships in the area.

Recommendations

4.12 Based on the Phase I review, OED recommends that the Bank:

(i) clarify the intent, scope, and requirements of the revised OP;
(ii) distinguish clearly between the safeguard (do no harm) aspects of the revised OP and its do good aspects. The OP should clearly delineate the extent of the Bank’s safeguard responsibilities. On the other hand, many of the do good aspects would be better specified in the Source Book. The policy relating to projects where IP are only beneficiaries should be moved out of the safeguard section of the policy and placed in the second section in the proposed OP 4.20;
(iii) identify indigenous and tribal groups in a manner consistent with the country’s legal framework. In countries where the legal framework does not meet the standards of the policy relating to coverage of IP, the Bank should ensure that IP are protected within the overall framework of its poverty reduction policies and establish a project-level system to monitor disaggregated impact on IP. In addition, the Bank should address the issue wherever appropriate in its country dialogue;
(iv) ensure that in countries with significant IP populations the Country Director, in consultation with the Environmentally and Socially Sustainable Development (ESSD) Network, engage the Borrower in discussions on how the Bank can best assist the country in providing culturally appropriate assistance to IP within the context of the Country Assistance Strategy (CAS) and agree with the Borrower on IP poverty monitoring indicators. The proposed Bank Procedure (BP) should specify clear accountabilities to this end; and
(v) design regional and sub-regional strategies to implement the OP given the significant differences in circumstances faced by Bank staff in implementing the policy.

...

Attachment III:
Operational Directive 4.20 (September 1991)

Introduction

1. This directive describes Bank^[A1] policies and processing procedures for projects that affect indigenous peoples. It sets out basic definitions, policy objectives, guidelines for the design and implementation of project provisions or components for indigenous peoples, and processing and documentation requirements.

2. The directive provides policy guidance to (a) ensure that indigenous people benefit from development projects, and (b) avoid or mitigate potentially adverse effects on indigenous people caused by Bank-assisted activities. Special action is required where Bank investments affect indigenous peoples, tribes, ethnic minorities, or other groups whose social and economic status restricts their capacity to assert their interests and rights in land and other productive resources.

Definitions

3. The terms ‘indigenous peoples,’ ‘indigenous ethnic minorities,’ ‘tribal groups,’ and ‘scheduled tribes’ describe social groups with a social and cultural identity distinct from the dominant society that makes them vulnerable to being disadvantaged in the development process. For the purposes of this directive, ‘indigenous peoples’ is the term that will be used to refer to these groups.

4. Within their national constitutions, statutes, and relevant legislation, many of the Bank's Borrower countries include specific definitional clauses and legal frameworks that provide a preliminary basis for identifying indigenous peoples.

5. Because of the varied and changing contexts in which indigenous peoples are found, no single definition can capture their diversity. Indigenous people are commonly among the poorest segments of a population. They engage in economic activities that range from shifting agriculture in or near forests to wage labor or even small-scale market-oriented activities. Indigenous peoples can be identified in particular geographical areas by the presence in varying degrees of the following characteristics:

(a) a close attachment to ancestral territories and to the natural resources in these areas;
(b) self-identification and identification by others as members of a distinct cultural group;
(c) an indigenous language, often different from the national language;
(d) presence of customary social and political institutions; and
(e) primarily subsistence-oriented production.

Task managers (TMs) must exercise judgment in determining the populations to which this directive applies and should make use of specialized anthropological and sociological experts throughout the project cycle.

Objective and policy

6. The Bank's broad objective towards indigenous people, as for all the people in its member countries, is to ensure that the development process fosters full respect for their dignity, human rights, and cultural uniqueness. More specifically, the objective at the center of this directive is to ensure that indigenous peoples do not suffer adverse effects during the development process, particularly from Bank-financed projects, and that they receive culturally compatible social and economic benefits.

7. How to approach indigenous peoples affected by development projects is a controversial issue. Debate is often phrased as a choice between two opposed positions. One pole is to insulate indigenous populations whose cultural and economic practices make it difficult for them to deal with powerful outside groups. The advantages of this approach are the special protections that are provided and the preservation of cultural distinctiveness; the costs are the benefits foregone from development programs. The other pole argues that indigenous people must be acculturated to dominant society values and economic activities so that they can participate in national development. Here the benefits can include improved social and economic opportunities, but the cost is often the gradual loss of cultural differences.

8. The Bank's policy is that the strategy for addressing the issues pertaining to indigenous peoples must be based on the informed participation of the indigenous people themselves. Thus, identifying local preferences through direct consultation, incorporation of indigenous knowledge into project approaches, and appropriate early use of experienced specialists are core activities for any project that affects indigenous peoples and their rights to natural and economic resources.

9. Cases will occur, especially when dealing with the most isolated groups, where adverse impacts are unavoidable and adequate mitigation plans have not been developed. In such situations, the Bank will not appraise projects until suitable plans are developed by the Borrower and reviewed by the Bank. In other cases, indigenous people may wish to be and can be incorporated into the development process. In sum, a full range of positive actions by the Borrower must ensure that indigenous people benefit from development investments.

Bank role

10. The Bank addresses issues on indigenous peoples through (a) country economic and sector work, (b) technical assistance, and (c) investment project components or provisions. Issues concerning indigenous peoples can arise in a variety of sectors that concern the Bank; those involving, for example, agriculture, road construction, forestry, hydropower, mining, tourism, education, and the environment should be carefully screened.^[A2] Issues related to indigenous peoples are commonly identified through the environmental assessment or social impact assessment processes, and appropriate measures should be taken under environmental mitigation actions (see OD 4.01, Environmental Assessment).

11. Country Economic and Sector Work. Country departments should maintain information on trends in government policies and institutions that deal with indigenous peoples. Issues concerning indigenous peoples should be addressed explicitly in sector and subsector work and brought into the Bank-country dialogue. National development policy frameworks and institutions for indigenous peoples often need to be strengthened in order to create a stronger basis for designing and processing projects with components dealing with indigenous peoples.

12. Technical Assistance. Technical assistance to develop the Borrower's abilities to address issues on indigenous peoples can be provided by the Bank. Technical assistance is normally given within the context of project preparation, but technical assistance may also be needed to strengthen the relevant government institutions or to support development initiatives taken by indigenous people themselves.

13. Investment Projects. For an investment project that affects indigenous peoples, the Borrower should prepare an indigenous peoples development plan that is consistent with the Bank's policy. Any project that affects indigenous peoples is expected to include components or provisions that incorporate such a plan. When the bulk of the direct project beneficiaries are indigenous people, the Bank's concerns would be addressed by the project itself and the provisions of this OD would thus apply to the project in its entirety.

Indigenous peoples development plan^A3

Prerequisites

14. Prerequisites of a successful development plan for indigenous peoples are as follows:

(a) The key step in project design is the preparation of a culturally appropriate development plan based on full consideration of the options preferred by the indigenous people affected by the project.

(b) Studies should make all efforts to anticipate adverse trends likely to be induced by the project and develop the means to avoid or mitigate harm.^[A4]

(c) The institutions responsible for government interaction with indigenous peoples should possess the social, technical, and legal skills needed for carrying out the proposed development activities. Implementation arrangements should be kept simple. They should normally involve appropriate existing institutions, local organizations, and nongovernmental organizations (NGOs) with expertise in matters relating to indigenous peoples.

(d) Local patterns of social organization, religious beliefs, and resource use should be taken into account in the plan's design.

(e) Development activities should support production systems that are well adapted to the needs and environment of indigenous peoples, and should help production systems under stress to attain sustainable levels.

(f) The plan should avoid creating or aggravating the dependency of indigenous people on project entities. Planning should encourage early handover of project management to local people. As needed, the plan should include general education and training in management skills for indigenous people from the onset of the project.

(g) Successful planning for indigenous peoples frequently requires long lead times, as well as arrangements for extended follow-up. Remote or neglected areas where little previous experience is available often require additional research and pilot programs to fine-tune development proposals.

(h) Where effective programs are already functioning, Bank support can take the form of incremental funding to strengthen them rather than the development of entirely new programs.

Contents

15. The development plan should be prepared in tandem with the preparation of the main investment. In many cases, proper protection of the rights of indigenous people will require the implementation of special project components that may lie outside the primary project's objectives. These components can include activities related to health and nutrition, productive infrastructure, linguistic and cultural preservation, entitlement to natural resources, and education. The project component for indigenous peoples development should include the following elements, as needed:

(a) Legal Framework. The plan should contain an assessment of (i) the legal status of the groups covered by this OD, as reflected in the country's constitution, legislation, and subsidiary legislation (regulations, administrative orders, etc.); and (ii) the ability of such groups to obtain access to and effectively use the legal system to defend their rights. Particular attention should be given to the rights of indigenous peoples to use and develop the lands that they occupy, to be protected against illegal intruders, and to have access to natural resources (such as forests, wildlife, and water) vital to their subsistence and reproduction.

(b) Baseline Data. Baseline data should include (i) accurate, up-to-date maps and aerial photographs of the area of project influence and the areas inhabited by indigenous peoples; (ii) analysis of the social structure and income sources of the population; (iii) inventories of the resources that indigenous people use and technical data on their production systems; and (iv) the relationship of indigenous peoples to other local and national groups. It is particularly important that baseline studies capture the full range of production and marketing activities in which indigenous people are engaged. Site visits by qualified social and technical experts should verify and update secondary sources.

(c) Land Tenure. When local legislation needs strengthening, the Bank should offer to advise and assist the Borrower in establishing legal recognition of the customary or traditional land tenure systems of indigenous peoples. Where the traditional lands of indigenous peoples have been brought by law into the domain of the state and where it is inappropriate to convert traditional rights into those of legal ownership, alternative arrangements should be implemented to grant long-term, renewable rights of custodianship and use to indigenous peoples. These steps should be taken before the initiation of other planning steps that may be contingent on recognized land titles.

(d) Strategy for Local Participation. Mechanisms should be devised and maintained for participation by indigenous people in decision making throughout project planning, implementation, and evaluation. Many of the larger groups of indigenous people have their own representative organizations that provide effective channels for communicating local preferences. Traditional leaders occupy pivotal positions for mobilizing people and should be brought into the planning process, with due concern for ensuring genuine representation of the indigenous population.^[A5] No foolproof methods exist, however, to guarantee full local-level participation. Sociological and technical advice provided through the Regional environment divisions (REDs) is often needed to develop mechanisms appropriate for the project area.

(e) Technical Identification of Development or Mitigation Activities. Technical proposals should proceed from on-site research by qualified professionals acceptable to the Bank. Detailed descriptions should be prepared and appraised for such proposed services as education, training, health, credit, and legal assistance. Technical descriptions should be included for the planned investments in productive infrastructure. Plans that draw upon indigenous knowledge are often more successful than those introducing entirely new principles and institutions. For example, the potential contribution of traditional health providers should be considered in planning delivery systems for health care.

(f) Institutional Capacity. The government institutions assigned responsibility for indigenous peoples are often weak. Assessing the track record, capabilities, and needs of those institutions is a fundamental requirement. Organizational issues that need to be addressed through Bank assistance are the (i) availability of funds for investments and field operations; (ii) adequacy of experienced professional staff; (iii) ability of indigenous peoples own organizations, local administration authorities, and local NGOs to interact with specialized government institutions; (iv) ability of the executing agency to mobilize other agencies involved in the plan's implementation; and (v) adequacy of field presence.

(g) Implementation Schedule. Components should include an implementation schedule with benchmarks by which progress can be measured at appropriate intervals. Pilot programs are often needed to provide planning information for phasing the project component for indigenous peoples with the main investment. The plan should pursue the long-term sustainability of project activities subsequent to completion of disbursement.

(h) Monitoring and Evaluation.^[A6] Independent monitoring capacities are usually needed when the institutions responsible for indigenous populations have weak management histories. Monitoring by representatives of indigenous peoples own organizations can be an efficient way for the project management to absorb the perspectives of indigenous beneficiaries and is encouraged by the Bank. Monitoring units should be staffed by experienced social science professionals, and reporting formats and schedules appropriate to the project's needs should be established. Monitoring and evaluation reports should be reviewed jointly by the senior management of the implementing agency and by the Bank. The evaluation reports should be made available to the public.

(i) Cost Estimates and Financing Plan. The plan should include detailed cost estimates for planned activities and investments. The estimates should be broken down into unit costs by project year and linked to a financing plan. Such programs as revolving credit funds that provide indigenous people with investment pools should indicate their accounting procedures and mechanisms for financial transfer and replenishment. It is usually helpful to have as high a share as possible of direct financial participation by the Bank in project components dealing with indigenous peoples.

Project Processing and Documentation

Identification

16. During project identification, the Borrower should be informed of the Bank's policy for indigenous peoples. The approximate number of potentially affected people and their location should be determined and shown on maps of the project area. The legal status of any affected groups should also be discussed. TMs should ascertain the relevant government agencies, and their policies, procedures, programs, and plans for indigenous peoples affected by the proposed project (see paras. 11 and 15(a)). TMs should also initiate anthropological studies necessary to identify local needs and preferences (see para. 15(b)). TMs, in consultation with the REDs, should signal indigenous peoples issues and the overall project strategy in the Initial Executive Project Summary (IEPS).

Preparation

17. If it is agreed in the IEPS meeting that special action is needed, the indigenous peoples development plan or project component should be developed during project preparation. As necessary, the Bank should assist the Borrower in preparing terms of reference and should provide specialized technical assistance (see para. 12). Early involvement of anthropologists and local NGOs with expertise in matters related to indigenous peoples is a useful way to identify mechanisms for effective participation and local development opportunities. In a project that involves the land rights of indigenous peoples, the Bank should work with the Borrower to clarify the steps needed for putting land tenure on a regular footing as early as possible, since land disputes frequently lead to delays in executing measures that are contingent on proper land titles (see para. 15(c)).

Appraisal

18. The plan for the development component for indigenous peoples should be submitted to the Bank along with the project's overall feasibility report, prior to project appraisal. Appraisal should assess the adequacy of the plan, the suitability of policies and legal frameworks, the capabilities of the agencies charged with implementing the plan, and the adequacy of the allocated technical, financial, and social resources. Appraisal teams should be satisfied that indigenous people have participated meaningfully in the development of the plan as described in para. 14(a) (also see para. 15(d)). It is particularly important to appraise proposals for regularizing land access and use.

Implementation and Supervision

19. Supervision planning should make provisions for including the appropriate anthropological, legal, and technical skills in Bank supervision missions during project implementation (see paras. 15(g) and (h), and OP / BP 13.05, Project Supervision). Site visits by TMs and specialists are essential. Midterm and final evaluations should assess progress and recommend corrective actions when necessary.

Documentation

20. The Borrower's commitments for implementing the indigenous peoples development plan should be reflected in the loan documents; legal provisions should provide Bank staff with clear benchmarks that can be monitored during supervision. The Staff Appraisal Report and the Memorandum and Recommendation of the President should summarize the plan or project provisions.

^[1] Size of Bank assistance over the evaluation period was the key criteria for selection of the 34 countries, except in four cases, where sub-regional diversity and evaluation overload was also considered. Initial review and consultation with Bank staff indicated that the Operational Directive (OD) had been applied in Latin America and the Caribbean (LCR), East Asia and Pacific (EAP), and South Asia (SAR) regions, while there was little or no application in Africa (AFR), Middle East and North Africa (MENA), and Europe and Central Asia (ECA) regions. Thus, 50 percent of the borrowing countries in LCR, EAP, and SAR regions, and 10 percent of those in AFR, MENA, and ECA were selected. The four exceptions: Zambia was replaced with Rwanda; Romania with Kazakhstan; Bangladesh with Nepal; and Korea with Cambodia.

^[2] This was a desk review of 33 Bank-financed projects identified, appraised or implemented between 1981 and 1985, known to have demonstrable effects on lands, resources, and cultures of IP.

^[3] See ‘Report on a Workshop on ‘Indigenous Peoples, Forests, and the World Bank: Policies and Practice’’ prepared by Thomas Griffith and Marcus Colchester, Forests People Program and Bank Information Center, August 2000.

^[4] The terms ‘scheduled tribes’ and ‘indigenous ethnic minorities’ are references respectively to tribal groups listed in Schedule V of the Indian Constitution and to the 55 national minorities recognized by the Chinese Constitution.

^[5] Kingsbury states: ‘vulnerability and limited capacity to assert rights and interests continue to underlie these criteria’ (Kingsbury, p.17 and 25).

^[6] See discussion in Chapter 3.

^[7] Eighteen sample countries, including Latin American countries (IP), China (ethnic minorities), India (Scheduled Tribes), and Pakistan (Federally Administered Tribal Areas).

^[8] Two sample countries (Indonesia and Vietnam – as defined in several project loan agreements).

^[9] Six sample countries (Cambodia, Malaysia, Nepal, the Philippines, Russia and Tahiland).

^[10] Eight sample countries (Cote d’Ivoire, Ethiopia, Ghana, Kazakhstan, Morocco, Rwanda, Tunisia, and Turkey).

^[11] Management would have preferred that OED take a different approach to the evaluation methodology, using the process for identification of IP prescribed in OD 4.20, as conveyed to OED by Management in March 2002. Definitional clauses and legal frameworks that are included in national constitutions, statues and relevant legislation of Bank’s borrowing countries provide a preliminary basis for identifying indigenous peoples. The presence of the five characteristics is the starting point of the assessment. However, the OD goes on to say that ‘Task managers (TMs) must exercise judgment in determining the populations to which this directive applies and should make use of specialized anthropological and sociological experts throughout.’ Management would also note that a desk review of this type may not have adequate field-level information at its disposal to determine conclusively the applicability of the OD.

^[12] This is consistent with the current guidance provided by ESSD on the Bank’s web page

^[13] See ‘Report on a Workshop on ‘Indigenous Peoples, Forests, and the World Bank: Policies and Practice’ prepared by Thomas Griffith and Marcus Colchester, Forests People Program and Bank Information Center, August 2000

...

^[26] Initially, OED examined all the 297 projects that were approved after January 1, 1992, three months after the issuance of OD 4.20. This is in line with the view that the OD became effective immediately upon issuance by Management. In fact, OD 4.20 does not contain an effectiveness date nor does it exempt any projects, which may have been at an advanced state of preparation at the time of its issuance. OED, therefore, considered that all projects that went to the Board thereafter could have been retrofitted to comply with the policy. OED considered this an integral part of Management accountability to implement the policy at least on a best effort basis, particularly because OD 4.20 followed an OMS issued in 1982 that required a ‘tribal component’ each time a project had ‘tribal populations’ in its zone of influence. Management explained that it would be more reasonable to expect that the OD would apply to projects for which the EPS/PCD was issued after December 1, 1991, or even after January 1, 1992, since this would be consistent with current approaches in other Board approved OPs. In view of Management's comments, OED amended the evaluation sample to include only the 234 projects appraised after January 1992 and closed before May 31, 2001. Interestingly, the levels of implementation of the OD remain the same for both samples, because several projects appraised between September 1991 and January 1992 included measures as per the previous OMS or had applied the OD.

^[27] In fact, as discussed in the section on the quality of implementation, in seven of these projects there is only a passing reference to IP but no explicit efforts to directly safeguard their interests.

^[28] Pakistan Second Private Sector Energy Development project, Russia Second Highway Rehabilitation project, Indonesia Fifth Kabupaten Roads project, and Indonesia Village Infrastructure project.

^[29] Ethiopia Calub Gas project, Morocco Second Agriculture Investment Loan, Pakistan Balochistan Primary Education project, Pakistan Social Action Program, and the Pakistan Population Welfare Program.

^[30] These included projects that support the drafting of domestic environmental laws and the establishment of institutional frameworks to strengthen environmental assessments in public and private sector investments. See El Salvador Technical Assistance Loan.

^[31] This position was confirmed by the Management in its response to the Inspection Panel in the China Western Poverty Reduction project: ‘For the purpose of OD 4.20, the Bank uses the official designation of 55 minority nationalities in determining the applicability of the term ‘indigenous peoples’.’

^[32] For example, Indonesia Fifth Kabuten Roads project (FY1994).

16

^[33] In the Second Rural Water Supply and Environmental Sanitation project and the Community Water Tanks project, both in Karnataka, other backward castes and some scheduled castes that meet the five characteristics are also treated as IP.

^[34] See Staff Appraisal Report (SAR) on Pakistan Balochistan Natural Resource Management project, which identifies these characteristics (pgs.12-15). They are also recognized as IP by other external stakeholders. (See Burger, The Gaia Atlas of First Peoples, pg. 184).

^[35] These include: (i) rural black communities on the Pacific Coast of Colombia under the Colombia Natural Resources Management project; (ii) Garifuna-speaking Afro-descendant communities along the Atlantic Coast under the Honduras Social Investment Fund IV and V and Coastal Tourism projects; and (iii) Afro-descendant communities along the Pacific Coast of Ecuador and Peru under the PRODEPINE and Indigenous Peoples and Afro-Peruvian Learning Innovation Loan (LIL). Partly, this can be attributed to the fact that some of these countries have recent legislation relating to Afro-Latin populations, although narrower in scope than that for IP or that of the OD.

^[36] Analysis in the appraisal document was considered adequate if it reflected an understanding of the socioeconomic and cultural characteristics of IP and the potential impact of the project on IP. Where this did not exist, the evaluation looked for a reference to a baseline study, or a social analysis/assessment, or consultations with IP during project preparation that would have indicated consideration of these issues.

^[37] IPDPs are not easily available, not maintained at the Bank’s Public Information Center, and in many cases are not traceable although referred to in the documents. The evaluation team contacted all regional IP coordinators to obtain lists of IPDPs for the relevant projects and the counts are based on the responses as well as references in documents. Management recognized that this had been a problem and has taken steps to substantially improve access, including the development of a project documentation system for environmental and other safeguard policy reports. See World Bank Policy on Disclosure of Information, June 2002. Management notes that our previous disclosure policy did not require public disclosure for all IPDPs—only for those classified as ‘A’ and in some cases for ‘B’ projects.

^[38] See para. 13 of OD 4.20 (Attachment III).

^[39] Projects that set up mechanisms to finance a series of sub-projects, the nature of which will be determined during implementation, according to established operational guidelines.

^[40] Management notes that in many cases the IPDPs and Resettlement Action Plans have been prepared as one document, where the primary impact of the project on the IPs was resettlement.

^[41] See Russia Oil Rehabilitation projects.

^[42] Reddy and Kvam (2001) argue along similar lines in ‘Note for Proposed Mumbai Urban Transport Project.’

^[43] Management interprets the OD to require a covenant in the legal documents on the obligation of the Borrower or the project entity to carry out the IPDP, if there is such an IPDP. However, if the bulk of the project beneficiaries are indigenous peoples, then the OD does not require a separate self standing IPDP, and any indigenous peoples issues or elements are addressed through project design, which may not be evident from the project description in the legal documents, and would not in and of itself require any covenants or reference to indigenous peoples in the legal documents. The umbrella covenant in the legal documents relating to the obligations of the Borrower to carry out the project (Section 3.01 of the Loan Agreement or Credit Agreement), and the covenants in the implementation program would be necessary to cover the aspects of the project that relate to IP. Similarly, no covenants are required if the PAD includes simply discussion of IP issues.

^[44] Loan Agreements of open projects were not reviewed.

^[45] PSRs were rated taking into consideration the importance of IP issues in the project.

^[46] For a good practice in Mid Term Reviews, see Indonesia Third Community Health and Nutrition project.

^[47] Eight projects cancelled of the 55.

^[48] Management notes that it has given QACU, in consultation with the Legal Department, the key role in providing Managing Directors with advice on the interpretation of safeguard policies.

...

^[A1] ‘Bank’ includes IDA, and ‘loans’ include credits.

^[A2] Displacement of indigenous people can be particularly damaging, and special efforts should be made to avoid it. See OD 4.30, Involuntary Resettlement, for additional policy guidance on resettlement issues involving indigenous people.

...

^[A4] For guidance on indigenous peoples and environmental assessment procedures, see OD 4.01, Environmental Assessment, and Chapter 7 of World Bank, Environmental Assessment Source Book, Technical Paper No. 139 (Washington, D.C., 1991).

^[A5] See also ‘Community Involvement and the Role of Nongovernmental Organizations in Environmental Assessment’ in World Bank, Environmental Source Book, Technical Paper No. 139 (Washington, D.C., 1991).

^[A6] See OD 10.70, Project Monitoring and Evaluation.