Hostname: page-component-6bf8c574d5-7jkgd Total loading time: 0 Render date: 2025-02-21T00:00:42.097Z Has data issue: false hasContentIssue false

The World Bank, human rights, and organizational legitimacy strategies: The case of the 2016 Environmental and Social Framework

Published online by Cambridge University Press:  24 June 2019

María Victoria Cabrera Ormaza
Affiliation:
International Labour Office, 4 route des Morillons, CH-1211, Genève 22, Switzerland; Universidad Espiritu Santo-Ecuador, Avenida Samborondón, Km. 2,5 vía a Samborondón, Samborondón 092301, Ecuador, Email: mvcabrer@uees.edu.ec
Franz Christian Ebert
Affiliation:
Max Planck Institute for Comparative Public Law and International Law, Neuenheimer Feld 535, D-69120, Heidelberg, Germany, E-mail: ebert@mpil.de
Rights & Permissions [Opens in a new window]

Abstract

While civil society groups have been urging the World Bank to integrate human rights concepts into its policies, borrower countries have increasingly made the case for flexibility and deference to domestic standards in the implementation of bank-funded projects. This article analyses how the World Bank has navigated these conflicting legitimacy demands in the context of its 2016 Environmental and Social Framework (ESF). Drawing on insights from organizational sociology, we focus on practices of decoupling, which allow organizations to correspond to legitimacy demands by different audiences while not having to substantially adjust their core activities. The labour and indigenous peoples’ safeguards serve as cases in point. Specifically, the article argues that the Bank has decoupled its discourse concerning the ESF from the framework’s actual content by making statements about the ESF’s coherence with key human rights concepts which, upon closer scrutiny, do not fully correspond to its actual requirements. The article also shows how the design of the ESF furthers a decoupling of relevant requirements from its actual implementation. In particular, the confined scope ratione personae of the relevant safeguards and the discretion granted to the Bank’s staff and the borrower to add meaning to undefined key concepts may render their human rights-related requirements in a number of cases inconsequential. By and large, the decoupling practices identified regarding the Bank’s ESF entail problematic effects for the normativity of relevant human rights concepts and may, in the long run, undermine the Bank’s legitimacy as a whole.

Type
ORIGINAL ARTICLE
Copyright
© Foundation of the Leiden Journal of International Law 2019 

1. Introduction

Addressing the human rights impacts of the World Bank’s operations has been on the agenda of civil society groups and other actors for some time.Footnote 1 While the Bank has, so far, refused to acknowledge that it is bound by international human rights law,Footnote 2 it has sought to legitimate its activities vis-à-vis human rights-oriented audiences, including by adopting operational policies aimed at preventing adverse effects of the projects it finances.Footnote 3 This is also reflected in the Bank’s ESF of 2016.Footnote 4 Replacing earlier World Bank safeguard policies, the ESF contains ten ‘Environmental and Social Standards’ (ESSs) to be observed by borrower countries when implementing projects.Footnote 5 While not including a dedicated human rights safeguard,Footnote 6 the ESF touches upon several human rights-related subject mattersFootnote 7 and references the Universal Declaration of Human Rights.Footnote 8

However, human rights-oriented groups constitute only one out of several audiences that confront the Bank with their legitimacy demands,Footnote 9 the most important of which are its donors and, increasingly, its borrower countries.Footnote 10 In a time of multipolar power constellations within the Bank and new alternatives for financing, in particular large borrower countries, such as China, India, or Brazil, have become more and more vocal about their legitimacy demands for respect of borrowers’ sovereignty and project ownership.Footnote 11 These legitimacy demands often stand in stark tension to those emphasized by human rights-oriented civil society organizations as well as by certain donor countries, which presents the Bank with complex trade-offs when designing its policies.Footnote 12

Against this backdrop, the present article examines how the World Bank has navigated conflicting legitimacy demands of different audiences with regard to human rights-related issues in the context of the ESF.Footnote 13 We develop this analysis in three steps. Drawing on insights from organizational sociology, Section 2 sets out a conceptual framework to better understand how international organizations manage their legitimacy in the face of conflicting demands. Our focus is on practices of ‘decoupling’, which allow organizations to respond to different legitimacy demands while leaving their core activities largely unaffected. On this basis, we explore how different forms of decoupling are reflected in the ESF’s human rights-related requirements, focussing on labour and indigenous peoples’ issues. In this vein, Section 3 analyses how the Bank has decoupled its discourse regarding the ESF from the framework’s actual content. The Bank has notably done so by making statements about the ESF’s coherence with human rights-related conceptsFootnote 14 which, upon closer scrutiny, are not fully matched by the ESF’s requirements. We show that this has in some cases involved a re-articulation of the contours of these concepts so as to harness their legitimating potential vis-à-vis human rights-oriented audiences while avoiding clashes with the demands of borrower countries. Section 4 then describes how the design of the ESF furthers a decoupling of its human rights-related policy requirements from the Bank’s operational practice. In particular, the covertly restrictive design of the scope ratione personae of the relevant ESSs may render their human rights-related requirements inconsequential in a number of cases. Furthermore, due to the significant discretion granted to the Bank’s staff and the borrower countries in this regard, affected groups can, in certain cases, be excluded from the application of the ESF’s substantive requirements on a case-by-case basis. Section 5 concludes by pointing to some implications of this analysis for the debates on the legitimacy of the World Bank as well as for international human rights law.

2. Organizational legitimacy, institutional constraints, and human rights concerns at the World Bank

2.1 Competing legitimacy demands and new power constellations

The legitimacy of an international organization is widely considered a significant determinant of its power and, in the end, of the successful implementation of its activities.Footnote 15 Such legitimacy is dynamic and hinges on the perceptions of the relevant audiences of an international organization.Footnote 16 As these perceptions tend to change more rapidly than the organization’s institutional design, legitimacy drifts arise which can lead to legitimacy deficits.Footnote 17 Aside from a failure to live up to extant standards of legitimacy, changing legitimacy audiences, the change of legitimacy standards by the relevant audiences as well as competing legitimacy demands may account for such drifts.Footnote 18

As far as the World Bank is concerned, highly consequential legitimacy demands have been voiced by human rights organizations and related social movements.Footnote 19 Often harnessing donor governments’ domestic institutions with leverage on the World Bank,Footnote 20 their pressure has contributed to important organizational changes within the Bank,Footnote 21 including the creation of mechanisms to tackle adverse human rights impacts.Footnote 22 This has involved procedural aspects, such as the establishment of the World Bank Inspection Panel,Footnote 23 as well as substantive ones, in particular the adoption of safeguard policies defining social and environmental requirements for the Bank’s operations.Footnote 24

The legitimacy demands of human rights-oriented civil society actors stand, however, in tension with those of other important legitimacy audiences, notably the Bank’s borrower countries. Increasingly defying the Bank’s normative authority,Footnote 25 large borrowers have demanded that the Bank undertake reforms, both of its decision-making processesFootnote 26 and of its policies. While not being uniform, the policy-related legitimacy demands of these countries have tended to emphasize sovereignty concerns, namely non-intervention in domestic affairs and project ownership.Footnote 27

The legitimacy demands of borrower countries have gained increasing weight in recent decades. A key factor in this regard has been the decreased dependence of borrower countries on Bank funding due to the increasing availability of alternative sources of finance,Footnote 28 including by new international lending institutions.Footnote 29 At the same time, large borrowers and other middle income countries remain central for the Bank’s business model.Footnote 30 While accruing major legitimacy deficits with human rights-related audiences can significantly complicate the implementation of the Bank’s agenda, legitimacy deficits with (large) borrower countries are likely to undermine the Bank’s policy objectives more severely and ultimately put at risk the organization’s survival altogether.Footnote 31 In light of the above, the hypothesis can be formulated that the World Bank is likely to make substantial efforts to abide by the legitimacy demands of borrower countriesFootnote 32 while at the same time trying to accommodate some of the human rights organizations’ legitimacy demands.

2.2 Managing clashing legitimacy demands: The role of decoupling

Whenever external legitimacy demands collide, addressing them through compliance alone, i.e., through institutional reforms or a change in operational practice, will be difficult for the organization concerned.Footnote 33 In such instances, organizations can be expected to employ a mix of compromise strategies – i.e., complying at least with the minimum demands of either audience to ensure their basic support – and avoidance strategies, which can involve ‘disguising nonconformity behind a façade of acquiescence’.Footnote 34 For this purpose, an organization is likely to resort to ‘coping mechanisms’.Footnote 35 This involves visible but mainly symbolic organizational responses that allow the organization to refrain from substantially changing its core processes while suggesting to the relevant audiences that their legitimacy demands have been met.Footnote 36

A key strategy to manage legitimacy in scenarios involving conflicting demands is ‘decoupling’. This can been understood as aligning formal structures with relevant legitimacy demands while not significantly affecting the organization’s core activities and social impact.Footnote 37 Decoupling suggests to external stakeholders that relevant ‘formal structures are really working’ while minimizing conflicts between legitimacy demands, thereby allowing the organization to ‘mobilize support from a broader range of external constituents’.Footnote 38 A number of studies have shed light on instances from different societal contexts where decoupling has been successfully used to obtain or maintain legitimacy.Footnote 39

Two main forms of decoupling have been identified in organizational sociology literature. The first category – known as policy-practice-decoupling – refers to ‘gaps between norms and behavior’.Footnote 40 These can be achieved, among others, by minimizing or ‘ceremonializing’ the possibilities for inspecting or otherwise controlling relevant activities, by handling relevant practices informally, and by providing for a substantial amount of discretion regarding the implementation of the relevant policies.Footnote 41

A second category has come to be named ‘means-ends-decoupling’. This describes the phenomenon that a given policy is adopted – and implemented – whose features, however, by design do not significantly contribute to the objectives pursued by that policy.Footnote 42 Especially in contexts where the importance of accountability is emphasized, resorting to means-ends-decoupling has been described as a more viable option for organizations.Footnote 43

Building on this literature, an additional variant of this phenomenon can be discerned, which we term ‘discourse-content-decoupling’. This type of decoupling involves cases where organizations make statements about a policy that do not correspond to its actual substance.Footnote 44 As we argue below, such decoupling has been performed by the Bank in the context of the ESF when affirming the coherence of its policies with concepts from the human rights discourse, which turns out to be nonexistent upon closer scrutiny.

In this vein, the two following sections subject relevant practices of decoupling to an in-depth analysis. The ESF is a particularly interesting object of study in this regard given that its adoption by the Bank’s Board of Executive Directors followed a deliberation process that reportedly involved ‘the most extensive consultation ever conducted by the World Bank’.Footnote 45 On the whole, no less than 2,000 stakeholders from more than 60 countries took part in the consultations on the first draft of the ESF, while the consultations on the second draft involved almost 3,000 participants from around 90 countries.Footnote 46 The World Bank was, thus, keenly aware of the stakeholders’ positions, which included trade unions, indigenous organizations, academics, and numerous NGOs. This allows us to shed light on the lines of conflict around the different human rights issues concerning the ESF.

3. Alike but not the same: Adapting human rights-related concepts to the World Bank’s operational context

Human rights-related concepts appear in the ESF in several contexts. This section looks at how the World Bank accommodates human rights language in a way that formally meets, at least in part, the legitimacy demands of civil society actors while minimizing undesired interference with borrower countries’ concerns for project ownership and deference to domestic standards.

This is done, for one, by referring to human rights-related concepts in a rather symbolic manner. More specifically, the Bank declares in the ESF’s Vision Statement that ‘[its] activities support the realization of human rights expressed in the Universal Declaration of Human Rights’ and vows to ‘continue to support its member countries as they strive to progressively achieve their human rights commitments’.Footnote 47 While expressing a political commitment towards international minimum human rights standards, the Bank is careful to avoid any statement that would imply commitments beyond what is expected from its borrowers under international law.

For another, in its operational part the ESF involves, as noted above, a form of discourse-content-decoupling concerning the human rights-related requirements. Essential in this regard is that the Bank refers to concepts from the human rights discourse in a way that suggests coherence between the ESF and these concepts, thereby harnessing their legitimating potential. As will become apparent in the following, the Bank’s discourse regarding these concepts does, however, not fully correspond to the actual content of the ESF. In the following, we investigate this form of decoupling in more detail with reference to the concept of ‘core labour standards’ as well as the indigenous peoples-related concepts of ‘consultation’ and ‘free prior and informed consent’.

3.1 Core labour standards à la carte Footnote 48

Largely drawing on the labour requirements of the International Finance Corporation (IFC)’s Performance Standards, Environmental and Social Standard 2 (ESS2) covers a variety of work-place-related subjects.Footnote 49 Much of the controversies regarding ESS2 revolved around the ILO’s ‘core labour standards’ (CLS), a short-cut for the term ‘Fundamental Principles and Rights at Work’.Footnote 50 Set out by a Declaration adopted under the auspices of the ILO in 1998, the CLS cover workers’ and employers’ freedom of association, non-discrimination at work, the elimination of child labour, and the eradication of forced labour.Footnote 51 During the consultation phase prior to the ESF’s adoption, human rights oriented civil society actors,Footnote 52 trade unions, as well as the ILOFootnote 54 and certain governmentsFootnote 55 insisted that the ESF should embrace and require protection of the CLS.Footnote 56 Meanwhile borrower countries underscored the need for the ESS not to impose unfeasible requirements and insisted that domestic law ought to serve as the main point of reference for the relevant standards.Footnote 57 Notably, the Chinese Government argued for providing ‘some flexibility’ and against one-size-fits-all labour standards requirements so as to ‘tak[e] into account the Borrower’s realities, development stages and relevant laws’.Footnote 58

The clashing legitimacy demands by human rights-oriented organizations and borrower countries, respectively, are reflected in how the Bank has dealt with the concept of CLS rhetorically. The Board Paper on the ESF, as submitted by the Bank’s Management, affirms that ‘[a]ll core labor standards … are reflected in the proposed labor provisions’.Footnote 59 This formulation implies at least a significant amount of coherence between ESS2 and this ILO-coined concept and suggests that the ESF intends to contribute to their protection. It thereby caters to the legitimacy demands of human and labour rights groups and indirectly harnesses the ILO’s CLS as a source of legitimacy. The Bank’s Management also noted though that specific wording had been inserted into the ESF ‘[t]o respond to borrowers’ concerns about implementability’ on this matter.Footnote 60

While ESS2 as such refrains from referring to the concept of CLS or any related ILO instrument,Footnote 61 it addresses issues regarding all four CLS. However, in order to resolve the conflict of embracing the concept of CLS and ensuring a high degree of flexibility for borrower countries the Bank has partly detached the notion of CLS from the meaning assigned to it by the ILO. In this regard a closer look at the individual CLS is warranted.

Coherence between ESS2 and the ILO’s notion of CLS is at its maximum with regard to forced labour. ESS2 even largely reproduces the wording of the definition of this term used in the ILO Forced Labour Convention.Footnote 62 The situation is somewhat more ambiguous in relation to child labour. ESS2 sets the minimum age of employment at 14 years, while the ILO’s Minimum Age Convention stipulates a minimum age of 15 years.Footnote 63 Only where the country at hand has invoked an exception clause, a minimum age of 14 years is in compliance with the Convention, so consistency is less than perfect in this regard.Footnote 64

More substantial inconsistencies are apparent concerning ESS2’s requirements on non-discrimination. In countries with laws that are incompatible with ESS2’s non-discrimination requirements the borrower is only obligated to ‘seek to carry out’ the project at hand in conformity with ESS2’s requirements ‘to the extent possible’.Footnote 65 The borrower, hence, does not need to fully observe the standards embodied in the relevant ILO instruments.Footnote 66 By making the observance of CLS-related requirements dependent on the borrower country’s legislation, the ESF turns the logic of the CLS on its head. Indeed, the very idea of deference to domestic law is alien to the ILO’s understanding of this concept, which this organization considers universal standards that can and must be upheld regardless of the country’s domestic context.Footnote 67

Particularly insightful for understanding how the Bank has reacted to human rights-related legitimacy demands are ESS2’s requirements pertaining to workers’ freedom of association. Unlike the IFC’s Performance Standards,Footnote 68 the First Draft of ESS2 did not establish any requirements for scenarios where domestic law does not protect workers’ freedom of association.Footnote 69 After this had raised substantial criticism, notably from trade unionsFootnote 70 and, importantly, the United States,Footnote 71 a clause was inserted into the second draft of ESS2 which precludes the borrower from impeding the establishment of ‘alternative’ mechanisms for grievance and workers’ interest protection.Footnote 72 Following additional advocacy efforts,Footnote 73 this provision was eventually complemented by an obligation not to ‘discriminate or retaliate’ against workers exercising their freedom of association or bargaining rights or using the said alternative mechanisms.Footnote 74

As a result, ESS2’s level of protection is now largely equivalent to that of IFC’s Performance Standard 2.Footnote 75 Nonetheless, significant inconsistencies with the ILO’s understanding of CLS persist. Similarly to the situation regarding non-discrimination at work, borrowers are not obligated to observe workers’ freedom of association where domestic law prohibits this.Footnote 76 The fact that ESS2’s requirements are, in part, applicable only to the extent that domestic laws do not stand in the wayFootnote 77 renders them incongruent with the respective CLS, as conceived of by the ILO.Footnote 78

The World Bank’s compromising approach to the CLS can be seen as paradigmatic of the trade-offs between human rights-related legitimacy demands and borrower countries’ demands for flexibility in project implementation. Arguably, a full incorporation of the ILO’s CLS into its policies could have entailed considerable challenges for the Bank’s operations. For example, the laws of some large borrower countries, notably China, restrict or prohibit the exercise of workers’ freedom of association.Footnote 79 Under these circumstances, requiring borrowers to allow for independent trade unions might have made it virtually impossible for the Bank to finance projects in those countries, putting at risk a substantial part of its operations.Footnote 80 That being said, by affirming coherence of ESS2 with the ILO’s CLS, the World Bank raises expectations that the substantial requirements of this policy do not live up to and, indirectly, dilutes the meaning of the concept of CLS as a whole.

3.2 Upholding or re-writing international standards on indigenous peoples? Consultation and free prior and informed consent

Regarded as one of most complex components of the ESF,Footnote 81 Environmental and Social Standard 7 (ESS7) aims at ensuring that the development process supported by the World Bank ‘fosters full respect for the human rights’ of ‘Indigenous Peoples and Sub-Saharan African Historically Underserved Traditional Local Communities’ (IP/SSAHUTLC).Footnote 82 Notably, ESS7 does not refer to the ILO Indigenous and Tribal Peoples Convention (ILO Convention No. 169)Footnote 83 or the UN Declaration on the Rights of Indigenous Peoples (UNDRIP), which nowadays constitute the most important sets of human rights standards on indigenous peoples.Footnote 84 Though the Guidance Note to ESS7 acknowledges the support given by a number of countries to international or regional instruments for the protection of IP/SSAHUTLC, mentioning explicitly the UNDRIP, it does not reflect any specific commitment of the Bank to them.Footnote 85

In order to build local project support, ESS7 requires borrowers to undertake ‘meaningful consultations’ with IP/SSAHUTLC in a way that is ‘culturally appropriate’.Footnote 86 When defining this consultation process, ESS7 appears to distance itself from relevant human rights standards in different ways. According to ILO Convention No. 169 and the UNDRIP, consultations should be undertaken through indigenous peoples’ own representative institutions guided by the principle of good faith.Footnote 87 Also, under Convention No. 169, indigenous representation ‘should be the result of a process carried out by indigenous peoples themselves’.Footnote 88 By contrast, ESS7 ambiguously provides for the involvement of ‘where appropriate, other community members’ in the consultation process.Footnote 89 Moreover, the Guidance Note defines appropriate representation of IP/SSAHUTLC as ‘the individuals who are considered by the majority of the affected IP/SSAHUTLC to be legitimate authorities to make decisions on collective support on their behalf’.Footnote 90 ESS7 also limits indigenous peoples’ right to make use of their own decision-making institutions, mentioning explicitly the UNDRIP,Footnote 91 by inserting a majority rule as it makes clear that the consultation process should ‘allow sufficient time for internal decision-making processes to reach conclusions that are considered legitimate by the majority of the concerned participants’.Footnote 92 Such an understanding is shaped by a liberal conception of democracy, which, historically, has not sufficiently allowed for the expression of indigenous perspectives.Footnote 93 In this regard, when addressing consultation with indigenous peoples, the United Nations Development Group has recalled the importance of ‘bearing in mind possible differences among community decision-making processes, which may be very different from mainstream decision-making based on representation, delegation of power and/or majority rule’.Footnote 94

However, the most innovative aspect of ESS7 is the incorporation of the requirement of free prior and informed consent (FPIC) with respect to three situations: Firstly, when the project will ‘have adverse impacts on land and natural resources subject to traditional ownership or under customary use or occupation’ of IP/SSAHUTLC; secondly, when it will cause their ‘relocation’; and, thirdly, when it will have ‘significant impacts’ on their ‘cultural heritage that is material to their identity and/or cultural, ceremonial, or spiritual aspects’ of their lives.Footnote 95

The incorporation of FPIC into the ESF can be seen as a response to the claims made by indigenous organizations,Footnote 96 as well as a way for the Bank to gain legitimacy by aligning its safeguards with the language of relevant human rights instruments. Free and informed consent was articulated for the first time in Article 16 of ILO Convention No. 169, in connection with removal and relocation of indigenous peoples. But it was only after the adoption of the UNDRIP in 2007 that a growing number of international legal instruments, decisions of judicial and quasi-judicial bodies, as well as documents of international expert bodies in the UN system started to refer to and endorse FPIC as a ‘principle’Footnote 97 or as a ‘human right norm’,Footnote 98 despite of its elusive languageFootnote 99 and contested meaning.Footnote 100 Hence, FPIC appears to now be the legitimating standard to be used by both governments and the private sector to validate projects affecting indigenous communities.Footnote 101

Reacting to the pressure exercised by some stakeholders requesting the Bank to consider the issue of FPIC in the safeguard policy review,Footnote 102 the Bank set up an Expert Focus Group to address this issue. The experts concluded that ‘FPIC should be implemented in line with the provisions specified in the UNDRIP’.Footnote 103 On the other hand, focusing on a possible veto right understanding of the FPIC, borrowers pointed to the problems that this notion could entail for the implementation of development projects. Brazil, for example, insisted that FPIC should be understood as a ‘consultation process’, arguing that a ‘veto right’ is neither guaranteed under ILO Convention No. 169 nor under the UNDRIP.Footnote 104 Kenya, for its part, noted that the inclusion of FPIC ‘may create situations where the rights of the few may prevent satisfying the needs and wants of the majority’.Footnote 105

In the ESF-related Board Paper, the Bank firmly states that the ESF ‘requires borrowers to obtain the FPIC of affected Indigenous Peoples [and SSAHUTLC]’.Footnote 106 However, the way the notion of FPIC has been articulated and defined in the ESF strongly reflects the Bank’s intention to accommodate borrower countries’ concerns. The scope of application of FPIC under the ESF appears to be narrower than under the UNDRIP. The Guidance Note clarifies that ‘in some projects, FPIC may be required only in relation to specific portions of land or aspects of a project’ and refers to the example of ‘linear projects that pass through multiple habitats which may require FPIC for the component that traverses IP/SSAHUTLC lands’.Footnote 107 Following this rationale, large-scale infrastructure or extractive projects which could have negative consequences on indigenous lives, such as environmental pollution, deforestation in surrounding areas, or limitations to their access to water resources,Footnote 108 may not be covered by FPIC. This contrasts with the UNDRIP, which refers to FPIC not only in connection with projects affecting the lands and resources of indigenous peoples but, in a broader sense, in connection with the adoption and implementation of ‘legislative or administrative measures that may affect them’.Footnote 109 Also, both ESS7 and the related Guidance Note remain silent on the question of how to determine the adversity or significance of impacts of proposed projects on the community, which, according to ESS7, is the criterion for when to apply FPIC.Footnote 110 Under ESS7, this assessment is an exclusive task of the borrower.Footnote 111 In contrast, under ILO Convention No. 169, indigenous peoples should take part in this initial assessment.Footnote 112

At the same time, ESS7 appears to treat FPIC as a self-standing concept, notwithstanding that under both ILO Convention No. 169 and the UNDRIP consent constitutes the objective of every consultation process.Footnote 113 For Anaya, FPIC was designed to intensify the demand on the state ‘to make every effort to build consensus on the part of all concerned’ within a consultation process.Footnote 114 By the same token, in its comments on the draft ESF, the Office of the High Commissioner for Human Rights recommended that ‘securing consent or agreement should be the objective of all consultations processes’.Footnote 115 However, this approach does not seem to be fully reflected in ESS7. Rather, aware of the absence of a universal definition of FPIC and in order to appease borrowers, who were against an interpretation of FPIC as a veto right of indigenous peoples,Footnote 116 the Bank equates it to ‘collective support’.Footnote 117 In this regard, the Bank makes it clear that ‘FPIC does not require unanimity and may be achieved, even when individuals or groups within or among affected [IP/SSAHUTLC] explicitly disagree’.Footnote 118 Also, the Guidance Note ambiguously defines ‘good faith’ in connection with FPIC, as – among other things – the ‘willingness to engage in a process and availability to meet at reasonable times and frequency’.Footnote 119 In this regard, however, the ILO has emphasized that:

[t]he adoption of rapid decisions should not be to the detriment of effective consultation for which sufficient time must be given to allow the country’s indigenous peoples to engage their own decision-making processes and participate effectively in decisions taken in a manner consistent with their cultural and social traditions.Footnote 120

Hence, by re-articulating certain elements of the requirement of consultation with indigenous peoples and by defining FPIC according to its own logic, the Bank may dilute the burden imposed on borrowers with respect to indigenous peoples, compared to what is required under ILO Convention No. 169 and the UNDRIP. This could have as a result a manipulated form of consultation involving actors that are strategically selected, thereby undermining the element of good faith.Footnote 121 Under ESS7, consultation with IP/SSAHUTLC and FPIC could, thus, turn out to be legitimation tools to validate Bank-sponsored projects and not primarily consensus-building devices and safeguards for indigenous peoples.

4. Diluting the World Bank’s human rights impact through the backdoor? The role of the ESF’s scope ratione personae

In addition to the discourse-content-decoupling related to human rights-related concepts described in Section 3, the ESF also facilitates a decoupling of relevant standards from the Bank’s operational practice. Apart from introducing vague concepts into key provisions of the ESF,Footnote 122 this is notably done by reducing the scope of the relevant requirements. Indeed, a general tendency is apparent within the ESF to compensate for the increased substantive requirements in the ESSs as compared to earlier safeguards by reducing their scope of application.Footnote 123 Notably the ESF allows, in certain cases, for the replacement of the ESSs’ requirements by the borrower country’s own legal and institutional framework,Footnote 124 with an unclear role of the Bank’s Inspection Panel in this regard.Footnote 125 This is the case if the Bank finds the latter to be ‘likely to address the risks and impacts of the project, and enable the project to achieve objectives materially consistent with the ESSs’.Footnote 126 Reduced ESS requirements apply furthermore regarding sub-projects other than those qualified by Bank staff as ‘high risk’.Footnote 127 In both cases Bank staff retains a considerable amount of discretion to preclude the ESF’s application,Footnote 128 thereby allowing for a decoupling of the relevant policies from the Bank’s project management practice.

A similar decoupling between the Bank’s policy and practice can be discerned concerning the specific human rights-related ESS requirements. As will be argued in the following, this encompasses both a restriction of the scope ratione personae by narrowly defining relevant legal concepts and the granting of palpable discretion to the Bank and the borrower countries regarding the definition of the requirements’ scope in concrete cases through the insertion of vague terminology.

4.1 Segmenting the coverage of workers at the discretion of the Bank and the borrower

An important feature to calibrate the ESF’s labour standards-related requirements to the Bank’s operational context consists of fragmenting the workforce depending on their link to the project at hand. This occurs, first, by establishing categories of workers to which a reduced set of requirements applies and second, by introducing ambiguous terminology that allows the Bank to exclude certain groups of workers from the protection of ESS2 altogether.

ESS2 applies in full only to so-called ‘direct workers’. This notion is defined as comprising any workers ‘employed or engaged directly by the Borrower … to work specifically in relation to the project’.Footnote 129 Even though the definition of ‘borrower’ is a rather broad one,Footnote 130 the coverage of the term ‘direct workers’ tends to be limited to a fraction of the workforce involved in Bank-funded projects. Indeed, in many projects only a minority of the workers are ‘direct workers’ and consequently subject to the full set of ESS2’s requirements.Footnote 131

For the remaining workforce, ESS2 creates three supplementary categories. Each of these categories of workers comes with its own level of protection, all of which are considerably lower than that applicable to ‘direct workers’. The first deals with ‘contracted workers’, a category that was added to the second draft of the ESF after pressure by trade union organizations, among other actors.Footnote 132 This category comprises workers employed by ‘contractors, sub-contractors, brokers … intermediaries’ and certain other third parties.Footnote 133 For these workers, borrowers must take certain measures with a view to ascertaining the third party’s compliance with ESS2. In particular, borrowers must insert the relevant ESS2 provisions into their contracts with third parties and carry out some initial screening of that third party.Footnote 134 Borrowers are also required to ‘establish procedures for managing and monitoring [their] performance’ regarding ESS2’s requirements and to ensure the existence of a grievance mechanism for contracted workers’ complaints.Footnote 135

Secondly, ESS2 covers ‘primary supply workers’, i.e., workers employed by ‘suppliers who, on an ongoing basis, provide directly to the project goods or materials essential for the core functions of the project’.Footnote 136 Their level of protection is lower than that available for contracted workers. Here, the requirements incumbent on borrowers are largely confined to taking measures to prevent or address instances of child and forced labourFootnote 137 and to address ‘serious safety issues’Footnote 138 regarding the workforce of their ‘primary suppliers’.

Thirdly, ESS2 addresses ‘community workers’. This refers to persons ‘providing community labor’ in a variety of scenarios, e.g., under ‘food for works programs and public workers as safety net programs’, ‘as a contribution [of the community] to the project, or where projects are designed and conducted for the purpose of fostering community-driven development’.Footnote 139 These workers, too, are subject to a significantly lower level of protection than ‘direct workers’. Indeed, the bulk of ESS2’s requirements is applied in a case-by-case manner, depending on issues such as ‘(a) the nature and scope of the project; (b) the specific project activities … and (c) the nature of the potential risks and impacts to the community workers’.Footnote 140 An exception applies only to the requirements concerning forced labour and child labour.Footnote 141 Moreover, many of the terms in which the relevant requirements for all three aforesaid categories are phrased are indeterminate, referring, e.g., to ‘significant risks’ or ‘serious safety issues’.Footnote 142

The protection of workers under ESS2 is, thus, highly segmented. Additionally, a substantial part of the workforce runs the risk of falling out of the scope of ESS2 altogether. The main reason for this is that certain key categories, such as ‘contracted workers’ or ‘primary supply workers’, have been defined in restrictive and rather vague terms. Notably, the definition of the term ‘contracted worker’ covers only those workers of third parties which ‘perform work related to core functions of the project’.Footnote 143 This term, in turn, is defined in a narrow and less-than-clear manner, comprising ‘those production and/or service processes essential for a specific project activity without which the project cannot continue’.Footnote 144 In the absence of a proper definition of the term ‘essential’,Footnote 145 borrowers will find it easy to make the case that numerous activities are not vital enough for the project to constitute a ‘core function’. This risk is exacerbated in the case of ‘primary supplier workers’, a notion which is confined to ‘those suppliers who, on an ongoing basis, provide directly to the project goods or materials essential for the core functions’.Footnote 146 Consequently, borrowers and the Bank itself are accorded significant discretion to preclude suppliers – and thereby also the suppliers’ workforce – from the coverage of ESS2 regarding a given project.

Finally, as ESS2 does not regulate the outsourcing of work in any way, borrowers are at liberty to delegate any project activity to others. As a result, it will in many cases be possible to avoid the requirements embodied in ESS2 through a rearrangement of the project structure. On the whole, the substantive requirements under ESS2 may turn out to not be applicable to the majority of the workforce of many future Bank-funded projects.

4.2 Who qualifies as ‘indigenous’? Adding an alternative category while reaffirming old criteria

One of the most controversial aspects in the design of the ESF’s section on indigenous peoples concerned the determination of the groups covered by the safeguards. During the first phase of the consultations on the safeguards review, indigenous organizations and different civil society actors urged the Bank to revise its old definitional criteria of indigenous peoples. In particular, the Bank was called upon to adopt a new definition that could apply in different national contexts, especially in Africa,Footnote 147 where the notion of indigenousness is controversial.Footnote 148 Indeed, on more than one occasion, Bank-funded projects undertaken on this continent were scrutinized by the Inspection Panel because of the Bank’s failure to recognize the presence of indigenous peoples in project areas.Footnote 149 Some borrower countries opposed a definition of indigenous peoples that could contravene their national laws.Footnote 150 Peru, for instance, argued that determining indigenousness on a project-by-project basis, despite the existence of an official list of indigenous communities, would be ‘costly and tedious for the Borrower’.Footnote 151

To address the concerns of indigenous groups, without ignoring the complexities that the term ‘indigenous’ has in certain contexts, as acknowledged by the Bank,Footnote 152 the term ‘Sub-Saharan African Historically Underserved Traditional Local Communities’ (SSAHUTLC) was coined and introduced into the text of ESS7 to cover ‘distinct social and cultural groups’ which do not qualify as indigenous peoples (IP) in their countries.Footnote 153 In this regard, ESS7 tries to reflect a flexible approach to its scope ratione personae by clarifying that the terms IP/SSAHUTLC are meant to be ‘used in a generic sense to refer exclusively to a distinct social and cultural group’ that meets, ‘in varying degrees’, the identification criteria set out by ESS7, as verified in the screening undertaken by the Bank.Footnote 154

Still, the definitional criteria set out in ESS7 to identify IP/SSAHUTLC largely reproduce those of previous Bank policies, which are inconsistent with relevant ILO standards.Footnote 155 One such criterion is self-identification as indigenous peoples, which ILO Convention No. 169 regards as ‘fundamental’ for the group to receive special protection as such.Footnote 156 In contrast, under ESS7 the status as an IP/SSAHUTLC is conditioned on others’ recognition of this identity, which involves a higher threshold to prove a group’s indigeneity than under ILO Convention No. 169.Footnote 157

Another definitional criterion of relevance is the collective attachment of the community to their ancestral territories as well as to the resources located thereon. The Bank’s understanding of collective attachment appears to be inspired by an economic logic. According to ESS7, ‘[c]ollective attachment means that for generations there has been a physical presence in and economic ties to land and territories traditionally owned, or customarily used or occupied, by the group concerned’.Footnote 158 This differs from the understanding of ‘collective attachment’ under the relevant international human rights instruments, though, where the relationship between indigenous peoples and their traditional lands is fundamentally of a cultural and spiritual nature.Footnote 159 But most importantly, as it has been in the past, the determination of the presence of IP/SSAHUTLC in project areas, under ESS7, rests exclusively with the Bank.Footnote 160 This is problematic given the ambiguity of the relevant criteria, which provides the Bank with significant leeway when deciding whether or not to grant a group special status under ESS7.

The preceding analysis reveals that the inclusion of a new group category in ESS7 did not come with changes in the Bank’s old definitional criteria and procedures for identification, which are narrower than those under ILO Convention No. 169. Overall, ESS7 is basically a reaffirmation of the Bank’s exclusive power and discretion to decide which groups will qualify as IP/SSAHUTLC under its safeguards.

5. Conclusion

This article adds to the debate on the legitimacy of international organizations by dissecting how the World Bank has dealt with conflicting legitimacy demands by different audiences in the context of the ESF. To this end, we have analysed how the Bank has accommodated human rights-related elements into its policies so as to signal to interested stakeholders that their concerns have been taken into account but at the same time abide by borrower countries’ demands for project ownership and flexibility. Key in this regard are different forms of decoupling, which allow the Bank to symbolically cater to human rights-oriented actors’ legitimacy demands while not necessitating substantial changes to its core activities.

Two main forms of decoupling have been identified with regard to the human rights-related content of the ESF. The first one can be described as discourse-content-decoupling. In the case at hand, this has consisted of a discrepancy between the Bank’s statements on how certain key human rights concepts are accommodated in its policies, on the one hand, and the actual features of these policies, on the other. In regard to ESS2, the Bank’s claim that the ESF ‘reflects’ the ILO’s core labour standards contrasts with the ESS2’s actual requirements, which diverge – in part drastically – from the standards contained in the pertinent ILO instruments. As concerns ESS7, the human rights-related concepts the Bank affirms to align its policies with – in particular FPIC – have indeed been directly incorporated into the relevant policy requirements. This has, however, involved a re-articulation of the concepts at stake in such a way that part of their normative content is diluted.

Second, the article has discerned a practice of decoupling the ESF’s policy requirements from the Bank’s operational practice. In contrast to what has tended to be the focus in the organizational sociology literature, here the policy-practice decoupling does not occur through a simple disregard of the relevant requirements. Rather, it is the policy itself that enables such decoupling by confining the ESSs’ scope ratione personae – or by providing the Bank or the borrower with discretion to do so on a case-by-case basis.

In light of how human rights concerns have at times been instrumentalized by the Global North against the Global South,Footnote 161 the reluctance of borrower countries via-à-vis the insertion of such language into lending arrangements of international financial institutions should not come as a surprise. This notwithstanding, the decoupling practices we have identified in relation to the Bank’s ESF are problematic in several ways. Notably, the manner in which the Bank has accommodated the aforesaid human rights-related concepts into its institutional context affects the determinacy of these conceptsFootnote 162 by giving rise to ‘identically phrased norms’, which, however, come with different legal consequences.Footnote 163 In this context ‘norm inconsistency can provide states with a valid argument to comply with the least onerous of applicable norms or the least onerous of competing interpretation of the applicable law’,Footnote 164 thereby weakening the normativity of the relevant concepts as a whole.

In addition, by purporting to embrace certain human rights-related concerns whilst actually leaving borrower countries numerous possibilities to evade relevant requirements, the ESF runs the risk of creating deceptive expectations among affected local communities and individuals. This also entails a latent risk to the ESF’s legitimacy vis-à-vis human rights-oriented audiences. Indeed, if these audiences obtain the impression that the ESF is ill-suited to prevent human rights violations in the relevant areas and mainly amounts to a ‘legitimacy façade’, its positive effects on the Bank’s legitimacy may soon be goneFootnote 165 and difficult to restore.Footnote 166

Some of the decoupling effects discussed above could arguably be addressed at the implementation stage provided that the Bank’s staff in charge uses its discretion accordingly. This may also require an adjustment of the incentives the Bank provides to its staff, which typically do not encourage complicating the finalization of projects due to social or environmental problems.Footnote 167 Whether the Bank will have the institutional capacity and the political will to proceed in this direction in the face of increasing competition by new development finance providersFootnote 168 remains to be seen.

Acknowledgments

The views and opinions expressed in this article are the authors’ own and do not reflect the position of the International Labour Organization or its members. Sections 2.2 and 3.2 were drafted by the first author while sections 1, 2.1, and 3.1 were drafted by the second author. The remaining sections were drafted by both authors jointly. The authors would like to thank Margherita Brunori, B. S. Chimni, Giedre Jokubauskaite, Martin Oelz, the Journal’s two peer reviewers, and the participants of the workshop ‘The World Bank Environmental and Social Framework (ESF) in a wider realm of public international law’, held in Durham on 18 September 2017, for their valuable comments on earlier drafts. Funding from Durham University, School of Law and Global Policy Institute, is gratefully acknowledged. A special acknowledgment is due to Violetta Ritz for invaluable editorial and conceptual suggestions at various stages of the drafting process, as well as to Jan Höffler for his comments and support. All errors remain, of course, our own.

References

1 See out of many, Clark, D. L., ‘The World Bank and Human Rights: The Need for Greater Accountability’, (2002) 15 Harvard Human Rights Journal 205 Google Scholar, 206.

2 See W. van Genugten, The World Bank Group, the IMF and Human Rights. A Contextualised Way Forward (2015), 3–5. For a critical perspective see, e.g., S. Skogly, Human Rights Obligations of the World Bank and the IMF (2001), 93 et seq.

3 Heupel, M., ‘Human Rights Protection in World Bank Lending: Following the Lead of the US Congress’, in Heupel, M. and Zürn, M. (eds.), Protecting the Individual from International Authority. Human Rights in International Organizations (2017), 241 CrossRefGoogle Scholar, 267.

4 See World Bank, The World Bank Environmental and Social Framework. Setting Environmental and Social Standards for Investment Project Financing (adopted 4 August 2016).

5 World Bank, Borrower Requirements—Environmental and Social Standards 1–10 (adopted 4 August 2016). The ESF also contains a policy setting out the Bank staff’s own due diligence requirements; World Bank, World Bank Environmental and Social Policy for Investment Project Financing (adopted 4 August 2016) (Environmental and Social Policy).

6 This has been called for by a number of actors; see, e.g., P. Alston, Report of the Special Rapporteur on Extreme Poverty and Human Rights, UN Doc. A/70/274 (4 August 2015), paras. 70, 80–5; see also Human Rights Watch, ‘Human Rights Watch Submission: World Bank’s Draft Environmental and Social Framework’, April 2015, 4–5, available at consultations.worldbank.org/consultation/review-and-update-world-bank-safeguard-policies.

7 This includes the areas of labour, health, property and housing, and indigenous peoples’ rights. See World Bank, ESS2. Labor and Working Conditions; ESS4. Community Health and Safety; ESS5. Land Acquisition, Restrictions on Land Use and Involuntary Resettlement; ESS7. Indigenous Peoples/Sub-Saharan African Historically Underserved Traditional Local Communities (adopted 4 August 2016, respectively).

8 World Bank, A Vision for Sustainable Development (adopted 4 August 2016), para. 3.

9 For the purpose of this article we rely on the definition of ‘legitimacy’ provided by Suchman who understands this concept as ‘a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions’; Suchman, M. C., ‘Managing Legitimacy: Strategy and Institutional Approaches’, (1995) 20 Academy of Management Review 571 CrossRefGoogle Scholar, 574.

10 For a review of the dynamics regarding development countries’ influence on the Bank and related constraints see M. Naim, ‘From Supplicants to Shareholders: Developing Countries and the World Bank’, in G. K. Helleiner (ed.), The International Monetary and Financial System. Developing-Country Perspectives (1996), 293.

11 See also Dann and Riegner in this issue (doi: 10.1017/S0922156519000293) and on the new role of BRICS countries regarding international organizations in general see Zürn, M. and Stephen, M., ‘The View of Old and New Powers on the Legitimacy of International Institutions’, (2010) 30 Politics 91 CrossRefGoogle Scholar, 97 et seq.

12 For an example regarding requirements related the insertion of non-discrimination standards into the ESF see D. Van Den Meersche, ‘Accountability in International Organisations: Reviewing the World Bank’s Environmental and Social Framework’, in E. Sciso (ed.), Accountability, Transparency and Democracy in the Functioning of Bretton Woods Institutions (2017), 157, 172.

13 The emphasis of external legitimacy demands incumbent on the World Bank in our analysis does not mean to downplay the role of the World Bank’s own agenda as well as the internal dynamics and conflicts within its staff, management, and governance bodies. See in this regard Sarfaty, G. A., ‘Why Culture Matters in International Institutions: The Marginality of Human Rights at the World Bank’, (2009) 103 American Journal of International Law 647CrossRefGoogle Scholar, 677–82.

14 By ‘human rights-related concepts’ we refer to concepts stemming from international human rights instruments, which are often utilized by human rights-oriented audiences when framing their demands.

15 See, e.g., Hurd, I., ‘Legitimacy, Power, and the Symbolic Life of the UN Security Council’, (2002) 8 Global Governance 35 CrossRefGoogle Scholar, at 36–7.

16 D. Zaum, ‘International Organizations, Legitimacy, and Legitimation’, in D. Zaum (ed.), Legitimating International Organizations (2013), 3, 10.

17 See Stephen, M. D., ‘Legitimacy Deficits of International Organizations: Design, Drifts, and Decoupling at the UN Security Council’, (2018) 31 Cambridge Review of International Affairs 96 Google Scholar, 101.

18 Ibid.; on the problem of competing legitimacy demands see Zaum, supra note 16, at 16.

19 Of course, the representation of diverse interests, which may also involve a North-South divide, through NGO networks at the World Bank is itself not without problems; see, e.g., Nelson, P., ‘Conflict, Legitimacy, and Effectiveness: Who Speaks for Whom in Transnational NGO Networks Lobbying the World Bank?’, (1997) 26 Nonprofit and Voluntary Sector Quarterly 421 CrossRefGoogle Scholar, 427–8.

20 See, e.g., Heupel, supra note 3, at 243–50, mentioning the example of the US Congress in particular.

21 On the interactions between pressure by external actors and internal advocacy efforts by ‘socially inclined’ World Bank staff members see, e.g., Vetterlein, A., ‘Economic Growth, Poverty Reduction, and the Role of Social Policies: The Evolution of the World Bank’s Social Development Approach’, (2007) 13 Global Governance 513 CrossRefGoogle Scholar, 525–9.

22 F. Bignami, ‘Theories of Civil Society and Global Administrative Law: The Case of the World Bank and International Development’, in S. Cassese (ed.), Research Handbook on Global Administrative Law (2016), 325, 334. On the ambivalences of these dynamics see Rajagopal, B., ‘From Resistance to Renewal: The Third World, Social Movements, and the Expansion of International Institutions’, (2000) 41 Harvard Journal of International Law 529 Google Scholar, 532, 539.

23 See the contributions in D. Clark et al. (eds.), Demanding Accountability. Civil Society Claims and the World Bank Inspection Panel (2003).

24 A. McBeth, International Economic Actors and Human Rights (2010), 196–206.

25 See Güven, A. B., ‘Defending Supremacy: How the IMF and the World Bank Navigate the Challenge of Rising Powers’, (2017) 93 International Affairs 1149 CrossRefGoogle Scholar, 1152.

26 A key concern in this regard has been to increase representativeness of the Global South within the Bank’s institutional framework; see, e.g., Wade, R. H., ‘Emerging World Order? From Multipolarity to Multilateralism in the G20, the World Bank, and the IMF’, (2011) 39 Politics & Society 347 CrossRefGoogle Scholar, at 359–60.

27 An illustration of this is provided in the Board Paper on the ESF; World Bank, Review and Update of the World Bank’s Safeguard Policies Environmental and Social Framework (Proposed Third Draft). Strengthening the Effectiveness of our Safeguard Policies to Enhance the Development Outcomes of Bank Operations, August 2016. It should be noted that the borrowers’ demands for ownership are supported by a range of international policy documents, in particular the Paris Declaration on Aid Effectiveness of 2005, paras. 14–15, which have been responded to by the Bank through attempts to increasingly rely on borrower countries’ legal frameworks in the course of a pilot project. See Van Den Meersche, supra note 12, at 167.

28 See, e.g., Weaver, C., ‘The World’s Bank and the Bank’s World’, (2007) 13 Global Governance 493 CrossRefGoogle Scholar, 502.

29 These are notably the New Development Bank and the Asian Infrastructure Investment Bank. See on this Wang, H., ‘New Development Banks: Opportunities and Challenges for Global Governance’, (2017) 8 Global Policy 113 CrossRefGoogle Scholar.

30 See Güven, A. B., ‘The World Bank and Emerging Powers: Beyond the Multipolarity-Multilateralism Conundrum’, (2017) 22 New Political Economy 496 CrossRefGoogle Scholar, 499.

31 In addition to this, human rights concerns may interfere with what the Bank perceives to be its core objectives, which may give rise to an additional trade-off between the efficiency of its operations and the relevant legitimacy demands by civil society. On such trade-offs in general see Meyer, J. W. and Rowan, B., ‘Institutionalized Organizations: Formal Structure as Myth and Ceremony’, (1977) 83 American Journal of Sociology 340 CrossRefGoogle Scholar, 355.

32 In practice, the Bank has attempted to accommodate borrower countries’ demands by adjusting its operational practices where possible, notably by reducing the ‘policy burden’ of its lending arrangements; see Güven, supra note 30, at 497. This strategy is particularly important to the Bank given that the legitimacy deficits arising from a lack of representativeness of the Bank’s voting system have proven difficult to reform; see, e.g., Vestergaard, J. and Wade, R. H., ‘Still in the Woods: Gridlock in the IMF and the World Bank Puts Multilateralism at Risk’, (2015) 6 Global Policy 1 CrossRefGoogle Scholar, 7.

33 See Stephen, supra note 17, at 104.

34 See Oliver, C., ‘Strategic Responses to Institutional Processes’, (1991) 16 Academy of Management Review 145 CrossRefGoogle Scholar, 152–3, 163–4, and quote at 154.

35 See Stephen, supra note 17, at 104.

36 MacLean, T. L. and Benham, M., ‘The Dangers of Decoupling: The Relationship between Compliance Programs, Legitimacy Perceptions, and Institutionalized Misconduct’, (2010) 53 Academy of Management Journal 1499 CrossRefGoogle Scholar, 1500.

37 See for a similar definition Stephen, supra note 17, at 105, who focuses, however, on the divide between formal structures and practices.

38 Meyer and Rowan, supra note 31, at 357.

39 For examples from the empirical literature see MacLean and Benham, supra note 36, at 1500 with further references.

40 J. W. Meyer, ‘Reflections: Institutional Theory and World Society’, in G. Krücken and G. S. Dori (eds.), World Society: The Writing of John W. Meyer (2009), 36, 50.

41 Meyer and Rowan, supra note 31, at 357.

42 See Bromley, P. and Powell, W. W., ‘From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World’, (2012) 6 Academy of Management Annuals 483 CrossRefGoogle Scholar, 489, who refer to means-ends-decoupling in cases ‘when policies are implemented but the link between formal policies and the intended outcome is opaque’. This can also involve scenarios ‘where organizations adopt new ends that are not directly related to [the organization’s] core goals’. Ibid., at 497, referring to Corporate Social Responsibility policies as an example.

43 Dick, P., ‘From Rational Myth to Self-Fulfilling Prophecy? Understanding the Persistence of Means-ends Decoupling as a Consequence of the Latent Functions of Policy Enactment’, (2015) 36 Organization Studies 897 CrossRefGoogle Scholar, at 899.

44 This is the main difference with means-ends-decoupling, where a policy sets out objectives that it is, by its design, unable to achieve.

45 See the World Bank’s statement at consultations.worldbank.org/consultation/review-and-update-world-bank-safeguard-policies. This comprised three rounds of consultations between 2012 and 2016 during which an ‘Approach Paper’ as well as two consecutive drafts of the ESF’s text were discussed whose structure and content strongly draws on the IFC’s Sustainability Framework. All submissions by governments and stakeholders made in the run-up to the ESF’s adoption, as cited in this article, are available at consultations.worldbank.org/consultation/review-and-update-world-bank-safeguard-policies.

46 See World Bank, supra note 27, at 10–12. Apart from receiving written submissions, the World Bank held country-level consultations, ‘expert focus groups’ dealing with selected topics, including labour and indigenous peoples’ issues, as well as separate ‘regional dialogues’ with a total of 240 representatives of indigenous peoples’ organizations. See World Bank, World Bank Policies Review and Update, Dialogue with Indigenous Peoples October 2013-March 2014 Summary, available at consultations.worldbank.org/Data/hub/files/consultation-template/review-and-update-world-bank-safeguard-policies/ru/materials/final_summary_dialogue_with_ip_october_2013-march_2014.pdf.

47 World Bank, supra note 8, para. 3.

48 This sub-section draws on F. C. Ebert, ‘Labour Standards and the World Bank. Analysing the Potential of Safeguard Policies for Protecting Workers’, in H. Gött (ed.), Labour Standards in International Economic Law (2018), 273, at 284–7.

49 This includes highly specific requirements on occupational safety and health, a project-level grievance mechanism for workers and their organizations, as well as conditions of employment, the latter of which refer, however, to domestic standards; see World Bank ESS2, supra note 7, paras. 10–12; 21–3; 24–30. For an in-depth analysis see Ebert, supra note 48, at 279–80, 284.

50 See further C. La Hovary, Les droits fondamentaux au travail: origines, statut et impact en droit international (2009), 33 et seq.

51 See para. 2 of the ILO Declaration on Fundamental Principles and Rights at Work, adopted on 18 June 1998 (ILO 1998 Declaration). The respective rights are also protected under key international human rights treaties. See, e.g., F. C. Ebert and C. La Hovary, ‘International Labour Law’, in R. Wolfrum (ed.), Max Planck Encyclopedia of Public International Law (2013), paras. 23–39.

52 See Human Rights Watch, supra note 6, at 12; Bank on Human Rights, Key Human Rights Concerns and Recommendations Regarding the World Bank’s Proposed Social and Environmental Safeguards Framework (March 2015), 8.

53 See Building And Woodworkers International, Review and Update of the World Bank Safeguard Policies, Input from Building And Woodworkers International (undated), 5; Confederation of Indonesian Trade Unions et al., Indonesian Trade Unions’ and NGOs’ Response to the First Draft of the World Bank Safeguards Policy Review Related to the Labor Provisions (December 2014), 1; Global trade union associations insisted that the Bank’s labour safeguards ‘should be as strong as those of the other banks as regards compliance with the core labour standards’; see International Trade Union Confederation, Global Unions, and Trade Union Advisory Committee, A Robust World Bank Labour Safeguard and IFI Support for a Wage- and Public Investment-led Recovery (October 2014), para. 7.

54 ILO, ILO Statement to the Washington DC Consultation of the Review and Update of the World Bank’s Environmental and Social Safeguard Policies (February 2016), 2.

55 See Government of Germany, German Comments on the World Bank Safeguards Review (April 2015), 3. Cf. also Government of France, French Non-Paper on the Bank’s Safeguards Review (April 2015), 2, which even goes further by demanding that ESS2 ‘should be expanded to all ILO International Labor Standards’.

56 One global trade union association had criticized that the draft ESF contained ‘no reference of any kind to ILO conventions or the core labour standards’. Building and Woodworkers International, Comments on the World Bank Environmental and Social Safeguard on Labour (October 2015), 1.

57 See, e.g., Government of Indonesia, World Bank Safeguard Policies Review Consultation Phase 2 (December 2014), 1–2; Subhash Chandra Garg, Brief of statement made by the Executive Director for Bangladesh, Bhutan, India and Sri Lanka, at the Committee on Development Effectiveness on 24 June and 1 July, 2015 on ESF (2015), 1–2.

58 Government of China, Comments and Recommendations from the Chinese Side on the Bank’s Proposed New Safeguard Policies (undated), 1.

59 See World Bank, supra note 27, at 22. The Guidance Note to ESS2 more modestly states that ‘ESS2 is in part informed by several International Labour Organization (ILO) and United Nations (UN) Conventions’, among which the eight ILO conventions underpinning the core labour standards; see World Bank, Guidance Note for Borrowers. ESS2. Labor and Working Conditions (June 2018), para. 2.2.

60 See World Bank, supra note 27, at 22.

61 See also Van Den Meersche, supra note 12, at 171. Some references to ILO instruments and documents are contained in the Guidance Note for Borrowers. ESS2, paras. 2.2, 16.1, and Annex on Additional References.

62 Compare ESS2, para. 20 to Art. 2(1) of the ILO Forced Labour Convention No. 29 of 1930.

63 Compare ESS2, para. 17 to Art. 2(3) of the ILO Minimum Age Convention No. 138 of 1973.

64 Ibid., Art. 2(4).

65 See World Bank ESS2, supra note 7, para. 13 (emphasis added).

66 Cf. ILO Equal Remuneration Convention as well as ILO Discrimination (Employment and Occupation) Convention No. 111 of 1958.

67 See Recital 7 and Art. 2 of the ILO 1998 Declaration and more generally E. De Wet, ‘Governance through Promotion and Persuasion: The 1998 ILO Declaration on Fundamental Principles and Rights at Work’, (2008) 9 German Law Journal 1429, 1437.

68 IFC, Performance Standard 2. Labour and Working Conditions (January 2012), paras. 13–14.

69 See World Bank, First Draft of the ESS2 (Washington, 30 July 2014), para. 11.

70 See, e.g., International Trade Union Confederation, Global Unions, and Trade Union Advisory Committee, supra note 53, at 2; see along similar lines Government of Germany, supra note 55, at 3.

71 See Government of the United States, United States Comments on World Bank Safeguards Review – Phase 2 (March 2015), 11.

72 See World Bank, Second Draft of the ESS2 (Washington, 1 July 2015), para. 16.

73 See, e.g., Building and Woodworkers International, supra note 56, at 1; see in the same vein Government of the United States, supra note 71, at 11.

74 World Bank ESS2, supra note 7, para. 16.

75 See on the latter Ebert, F. C., ‘The Integration of Labour Standards Concerns into the Environmental and Social Policy of the International Finance Corporation’, (2014) 47 Law and Politics in Africa, Asia and Latin America 229 Google Scholar, 234–9.

76 World Bank ESS2, supra note 7, para. 16.

77 In essence, where the domestic laws of the host country constrain the establishment or activities of trade unions, borrowers are only under an obligation not to hamper workers from setting up alternative arrangements and to refrain from ‘discriminat[ing] or retaliat[ing]’ against those ‘participat[ing], or seek[ing] to participate’ in such arrangements. See World Bank ESS2, supra note 7, para 16.

78 See already Van Den Meersche, supra note 12, at 171.

79 See ITUC, Survey of Violations of Trade Union Rights (undated), available at survey.ituc-csi.org/China.html?lang=en#tabs-2.

80 In 2017, China was the World Bank’s largest borrower with commitments amounting to US$2.42 billion; see World Bank, Annual Report 2017 (2017), 77, 79.

81 Azzarello, S., ‘The World Bank’s Environmental and Social Standards on Indigenous Peoples: A Case Study on Global Governance’, (2017) 50 Law and Politics in Africa, Asia and Latin America 135 Google Scholar, 137.

82 World Bank ESS7, supra note 7, para. 5

83 Brazil noted that ILO Convention No. 169 is a poorly ratified and ‘ambitious’ document. See World Bank, Review and Update of the World Bank’s Environmental and Social Safeguard Policies. Phase 3. Feedback Summary, Brasilia (1 March 2016), 6.

84 See, for example, Inter-American Court of Human Rights, Yayke Axa Indigenous Community v. Paraguay, Judgment of June 17, 2005, para. 127; Committee on the Elimination of Racial Discrimination, Concluding Observations: Canada, CERD/C/CAN/CO/21-23 (13 September 2017), para. 18. For further analysis on the use of these instruments in the human rights context see B. Saul, Indigenous Peoples and Human Rights: International and Regional Jurisprudence (2016), especially Chs. 3 and 4. Notably, the World Bank Inspection Panel has stressed the importance of ensuring that projects sponsored by the Bank be consistent with ILO Convention No. 169 in countries that have ratified this Convention. See World Bank Inspection Panel, Investigation Report, Honduras: Land Administration Project (June 12, 2007), paras. 256–8.

85 World Bank, Guidance Note for Borrowers. ESS7: Indigenous Peoples/ Sub-Saharan African Historically Underserved Traditional Local Communities (June 2018), para. 6.3.

86 World Bank ESS7, supra note 7, para. 23.

87 ILO Convention No. 169, Art. 6; UNDRIP, Art. 19.

88 Report of the Tripartite Committee Set Up to Examine the Representation Alleging Non-observance by Mexico of the Indigenous and Tribal Peoples Convention, 1989 (No. 169), Made under Article 24 of the ILO Constitution by the Union of Academics of the National Institute of Anthropology and History (GB.289/17/3), para. 102.

89 World Bank ESS7, supra note 7, para. 23 (a).

90 World Bank Guidance Note for Borrowers. ESS7, supra note 85, para. 25.1 (emphasis added).

91 UNDRIP, Art. 18.

92 World Bank ESS7, supra note 7, fn. 14.

93 See Plattner, M. F., ‘Populism, Pluralism and Liberal Democracy’, (2010) 21 Journal of Democracy 81 CrossRefGoogle Scholar, 84; H. Quane, ‘The UN Declaration on the Rights of Indigenous Peoples: New Directions for Self-Determination and Participatory Rights?’, in S. Allen and A. Xanthaki (eds.), Reflections on the UN Declaration on the Rights of Indigenous Peoples (2011), 259, 267; Yashar, D., ‘Democracy, Indigenous Movements, and the Postliberal Challenge in Latin America’, (1999) 52 World Politics 76 CrossRefGoogle Scholar, 77; Tockman, J., ‘The Hegemony of Representation: Democracy and Indigenous Self-government in Bolivia’, (2017) 9(2) Journal of Politics in Latin America 121 CrossRefGoogle Scholar, 131.

94 See United Nations Development Group’s Guidelines on Indigenous Peoples’ Issues (adopted on 1 February 2008, published in 2009), 27, available at undg.org/wp-content/uploads/2016/10/UNDG_guidelines_EN.pdf.

95 World Bank ESS7, supra note 7, para. 24.

96 World Bank, World Bank Safeguard Policies Review and Update, Dialogue with Indigenous Peoples, supra note 46, at 3.

97 See J. Anaya, Report of the Special Rapporteur on the Rights of Indigenous Peoples, James Anaya, Extractive industries and indigenous peoples, UN Doc. A/HRC/24/41 (1 July 2013), para. 26.

98 See HRC, Study of the Expert Mechanism on the Rights of Indigenous Peoples, ‘Free, Prior and Informed Consent: a Human Rights-Based Approach’, UN Doc. A/HRC/39/62 (10 August 2018), 4.

99 See M. Barelli, ‘Free, Prior and Informed Consent in the UNDRIP. Arts. 10, 19, 29 (2), and 32 (2)’, in J. Hohmann and M. Weller (eds.) The UN Declaration on the Rights of Indigenous Peoples: A Commentary (2018), 247, 249.

100 See, in general, Schilling-Vacaflor, A., ‘Who Controls the Territory and the Resources? Free, Prior and Informed Consent (FPIC) as a Contested Human Rights Practice in Bolivia’, (2017) 38 Third World Quarterly 1058 CrossRefGoogle Scholar.

101 Papillon, M. and Rodon, T., ‘Proponent-Indigenous Agreements and the Implementation of the Right to Free, Prior, and Informed Consent in Canada’, (2017) 62 Environmental Impact Assessment Review 216 CrossRefGoogle Scholar, 220. See also Tomlinson, K., ‘Indigenous Rights and Extractive Resource Projects: Negotiation over the Policy and Implementation of FPIC’, (2017) 21 International Journal of Human Rights 1 Google Scholar.

102 World Bank, The World Bank’s Safeguard Policies Proposed Review and Update, Approach Paper, Executive Summary (October 10, 2012), 11, para. 35.

103 World Bank, World Bank’s Safeguard Policies Review and Update Expert Focus Group on the Emerging Area Free, Prior, and Informed Consent of Indigenous Peoples, Manila, Philippines (21 March 2013), 3.

104 World Bank, supra note 83, at 6.

105 World Bank, Review and Update of the World Bank’s Environmental and Social Safeguard Policies. Phase 2. Feedback Summary, Nairobi (25 February 2015), 4.

106 World Bank, supra note 27, at 37.

107 World Bank Guidance Note for Borrowers. ESS7, supra note 85, para. 24.2.

108 Ritter et al. refer to these impacts as the expected impacts from the building of hydroelectric dams, highways and from mining project in the Brazilian Amazonia. See in general Ritter, C. et al., ‘Environmental Impact Assessment in Brazilian Amazonia: Challenges and Prospects to Assess Biodiversity’, (2017) 206 Biological Conservation 161 CrossRefGoogle Scholar.

109 UNDRIP, Art. 19.

110 This issue was raised by the Minority Rights Group International and Lex Justi in their comments on the draft Guidance Note 7. See Letter from the Minority Rights Group International and Lex Justi to the World Bank Safeguards Team with their comments on the draft Guidance Notes (15 December 2017), available at minorityrights.org/advocacy-statements/comments-regarding-world-banks-environmental-social-framework/.

111 World Bank ESS7, supra note 7, para. 12.

112 ILO Convention No. 169, Art. 15(2).

113 ILO Convention No. 169, Art. 6; UNDRIP, Arts. 19, 32(2). See M. Barelli, supra note 99, at 250; M. V. Cabrera Ormaza, The Requirement of Consultation with Indigenous Peoples in the ILO: Between Normative Flexibility and Institutional Rigidity (2017), 119.

114 See J. Anaya, Promotion and Protection of all Human Rights, Civil, Political, Economic, Social and Cultural Rights including the Right to Development. Report of the Special Rapporteur on the situation of Human Rights and Fundamental Freedoms of Indigenous Peoples, James Anaya, UN Doc. A/HRC/12/34 (15 July 2009), para. 48.

115 UN/OHCHR, Review and Update of the World Bank’s Safeguards Policies, Comments and Recommendations of UN/OHCHR in Relation to the Draft Environmental and Social Framework (March 2016), 14.

117 World Bank ESS7, supra note 7, para. 26.

118 Ibid., para. 25(d).

119 World Bank Guidance Note for Borrowers. ESS7, supra note 85, para. 25.3

120 Report of the Committee Set Up to Examine the Representation Alleging Non-observance by Colombia of the Indigenous and Tribal Peoples Convention, 1989 (No. 169), Made under Art. 24 of the ILO Constitution by the Central Unitary Workers’ Union (CUT), GB.282/14/3, para. 79.

121 This concern was raised by Oxfam in its comments to the Draft Guidance Note 7: See Oxfam, Proposed Gender Actions and Recommendations on the World Bank’s Environmental and Social Framework Guidance Notes, 9 October 2017, available at pubdocs.worldbank.org/en/412901515533419775/Draft-ESF-Guidance-notes-comments-OXFAM.pdf, at 17. See also D. Brown, ‘Participation in poverty reduction strategies: Democracy strengthened or democracy undermined?’, in S. Hickey and G. Mohan (eds.), Participation: From tyranny to transformation? Exploring new approaches to participation and development (2004), 241. Notably, in the past, the World Bank Inspection Panel had observed that, in various cases, the Bank’s staff wrongly assumed that an agreement to discuss the project and an early interest in it constituted a broad community support. See World Bank Inspection Panel, Emerging Lessons Series No. 2 (2016), 8.

122 See, e.g., P. Dann and M. Riegner, Safeguard-Review der Weltbankgruppe. Ein neuer Goldstandard für das globale Umwelt- und Sozialrecht? (2017), 27; Van Den Meersche, supra note 12, at 178–9.

123 See in a similar vein Passoni, C., Rosenbaum, A. and Vermunt, E., ‘Empowering the Inspection Panel: The Impact of the World Bank’s New Environmental and Social Safeguards’, (2016) 49 New York University Journal of International Law and Politics 921 Google Scholar, 926.

124 See in detail Van Den Meersche, supra note 12, at 180–1. For a critical perspective see Bugalski, N., ‘The Demise of Accountability at the World Bank?’, (2016) 31 American University International Law Review 1 Google Scholar, at 30–1; G. Lehane, ‘Human Rights at the World Bank Group’, in Y. Radi (ed.), Research Handbook on Human Rights and Investment (2018), 186, 205.

125 See Dann and Riegner, supra note 122, at 27; Passoni, Rosenbaum and Vermunt, supra note 123, at 927.

126 See World Bank, Environmental and Social Policy, para. 23 and, for further analysis, Van Den Meersche, supra note 12, at 168.

127 See World Bank, Environmental and Social Policy, supra note 5, para. 37.

128 See further Ebert, supra note 48, at 298.

129 World Bank ESS2, supra note 7, para. 3(a) in conjunction with para. 4. On the specific case of civil servants involved in World Bank-funded projects see ibid., para. 8.

130 Both the ‘project proponent’ as well as the ‘project implementing agencies’ are covered; ibid., para. 3(a).

131 See also Government of Germany, supra note 55, at 3.

132 See International Trade Union Confederation, Global Unions, and Trade Union Advisory Committee, supra note 53, at 2; Government of Germany, supra note 55, at 3.

133 World Bank ESS2, supra note 7, para. 3b.

134 Ibid., paras. 31, 32.

135 Ibid., paras. 32–3, quote at para. 32.

136 Ibid., para. 3c, fn. 5.

137 See World Bank ESS2, supra note 7, para. 40.

138 Ibid., para. 41. In cases where steps to make the primary supplier comply with the relevant requirements are unsuccessful, the borrower must switch to a supplier which is able to show that it is observing the standards at hand; see World Bank ESS2, supra note 7, para. 42.

139 Ibid., paras. 3d and 34.

140 Ibid., para 35.

141 Ibid., para. 37.

142 Ibid., paras. 40, 41.

143 Ibid., para. 3b (emphasis added).

144 Ibid., para. 3b, fn. 4.

145 Cf. World Bank Guidance Note for Borrowers. ESS2, supra note 61, paras. 3.4, 3.5, 5.1.

146 World Bank ESS2, supra note 7, para. 3c, fn. 5 (emphasis added).

147 World Bank, First Phase of Consultations to Update Safeguards Policies, Meeting with Civil Society, Academia, and the Private Sector, Dakar, Feedback Summary, (1 February 2013), at 2; World Bank, Safeguard Policies Review and Update: Consultation Phase 1 Feedback Summary. Meetings, Expert Focus Group, Paper Submissions, Consultations with Project-Affected Communities, and Dialogue with Indigenous Peoples, (March 2014), at 12.

148 F. Mukwiza Ndahinda, Indigenousness in Africa: A contested legal framework for Empowerment of ‘Marginalized’ communities (2011), 10.

149 The World Bank Inspection Panel, supra note 121, at 6. See also: Tammy Kim, E., ‘Unlikely formation: Contesting and Advancing Asian/African “Indigenousness” at the World Bank Inspectional Panel’, (2008–2009) 41 N.Y.U. Journal of International Law and Policy 131–58Google Scholar; Sarfaty, G. A., ‘The World Bank and the Internationalization of Indigenous Rights Norms’, (2005) 114 Yale Law Journal 1791 Google Scholar, 1804–6.

150 World Bank, Review and Update of World Bank’s Environmental and Social Safeguard Policies, Phase 3, Feedback Summary, 5–6 November 2015, New Delhi, 20–21; Review and Update of World Bank’s Environmental and Social Safeguard Policies, Phase 3, Feedback Summary, 16–17 November 2015, Lima, 6.

151 World Bank, Review and Update of World Bank’s Environmental and Social Safeguard Policies, Phase 3, Feedback Summary, 16–17 November 2015, Lima, 6.

152 MacLaren, O. W. and Parisseau, J., ‘The New World Bank Safeguard Standard for Indigenous Peoples: Where do we start?’, (2017) 45 Syracuse Journal of International Law and Commerce 35 Google Scholar, 44.

153 World Bank ESS7, supra note 7, para. 6.

154 Ibid., paras. 8, 10; World Bank Guidance Note for Borrowers. ESS7, supra note 85, para. 8.1. This solution proposed by the Bank was subject to criticism. In its comments to the second draft of the ESF, the Office of the High Commissioner for Human Rights noted that the use of new terminology could result in the dilution of the protection given to indigenous peoples under international law; UN/OHCHR, Review and Update of the World Bank’s Safeguards Policies, Comments and recommendations of UN/OHCHR in relation to the draft Environmental and Social Framework (15 March 2016), 14. Such a concern appeared to reflect the position of some indigenous groups from Africa who wished to obtain support from the Bank to advance their claims for legal recognition; World Bank Regional Dialogue with Indigenous Peoples in Africa, Cape Town, 4–6 December 2013, Summary, 3.

155 Compare ESS7 para. 8 to Operational Policy 4.10 Indigenous Peoples (2005), para. 4.

156 ILO Convention No. 169, Art. 1, para. 2.

157 World Bank. ESS7, supra note 7, para. 8a.

158 Ibid., fn. 6. The Guidance Note reaffirms this approach stating that ‘collective attachment signifies that the groups generally consider their lands and resources to be collective assets’; World Bank Guidance Note for Borrowers. ESS7, supra note 85, para. 8.2.

159 Art. 13 of ILO Convention No. 169 and Art. 25 of the United Nations Declaration on the Rights of Indigenous Peoples.

160 World Bank. ESS7, supra note 7, at 10.

161 See Mutua, M., ‘Savages, Victims, and Saviors: The Metaphor of Human Rights’, (2001) 42 Harvard International Law Journal 201 Google Scholar.

162 Determinacy is understood by Thomas Franck as the quality of a rule that conveys a clear message, and which constitutes an indicator of rule-legitimacy. See T. Franck, The Power of Legitimacy among Nations (1990), 49, 52.

163 T. Broude and Y. Shany, ‘The International Law and Policy of Multi-Sourced Equivalent Norms’, in T. Broude and Y. Shany (eds.), Multi-Sourced Equivalent Norms in International Law (2011) 1, 8.

164 C. Charters, ‘Multi-Sourced Equivalent Norms and the Legitimacy of Indigenous Peoples’ Rights under International Law’ in T. Broude and Y. Shany (eds.), Multi-Sourced Equivalent Norms in International Law (2011) 289, 300.

165 For a similar, albeit more abstract, argument see MacLean and Benham, supra note 36, at 1515.

166 On the challenges arising with regard to ‘repairing legitimacy’ see Suchman, supra note 9, at 597.

167 See G. A. Sarfaty, Values in Transition. Human Rights and the Culture of the World Bank (2012), 79–81.

168 Often finance provided by these actors comes with less policy strings attached than the World Bank’s financial services. See, e.g., Wang, supra note 29, at 116; Weaver, supra note 28, at 502.