Please note that on our website we use cookies to enhance your experience, and for analytics purposes. To learn more about our cookies, please read our privacy policy. By clicking ‘Allow cookies’, you agree to our use of cookies. By clicking ‘Decline’, you don’t agree to our Privacy Policy.

No translations available
Batwa women in Uganda. Credit: MRG/Emma Eastwood

Comments regarding the World Bank’s Environmental and Social Framework

16 September 2016

After nearly two years’ engagement with the World Bank safeguards process, Minority Rights Group International (MRG), an international non-governmental organization working to secure the rights of minorities and indigenous peoples, and Lex Justi, a law practice with a business and human rights specialty, take this opportunity to outline our main reflections on the standards as finally adopted, namely the World Bank’s Environmental and Social Framework: Setting Environmental and Social Standards for Investment Project Financing.[1]  For brevity’s sake, we touch on only the outcomes of our key concerns during the first and second rounds of comments and consultations.

  1. Human Rights

The final language in the Vision, para. 3  appears to be stronger. Along with many other civil society groups and in fact some state representatives, MRG and Lex Justi were heavily critical of the language in the 2nd draft: ‘shares the aspirations of the Universal Declaration of Human Rights and helps its clients fulfill those apirations.’ We felt that the draft language represented a fundamental misreading of the Universal Declaration (UDHR) as an ‘aspirational’ document, as opposed to its true foundational nature. While the final version contains the following improved language: ‘the World Bank’s activities support the realization of human rights expressed in the Universal Declaration of Human Rights’, we remain concerned that the World Bank has not explictly expressed support for the UDHR or committed to promote respect for human rights. In particular,  the phrase ‘support the realization’ refers to the World Bank’s activities rather than the World Bank itself, and is therefore, strangely passive. Consequently, it gives the impression that the World Bank is not actively endeavouring to ensure that the implementation of investment projects it finances is carried out in a manner that respects and furthers human rights. Why does the text not just say that the World Bank will promote States’ respect of human rights? This would have brought the World Bank into line with other members of the UN family and would be consistent with Secretary-General Ban Ki-moon’s November 2013 ‘Human rights Up Front’ initiative.

We welcome the modification whereby  ‘in a manner consistent with its Articles of Agreement’ has been removed from the reference to the UDHR (previously ‘the promotion of such an approach’) to now more generally ‘will continue to support its member countires as they strive to progressively achieve their human rights commitments’.  This means that the UDHR’s applicability is quite rightly no longer contingent on the World Bank’s Articles of Agreement, another point which we and many others criticised.

We note that the phrase ‘progressively achieve’ actually reveals an incorrect understanding of human rights – some commitments (i.e. economic, social and cultural rights) can indeed be met progressively, whereas other rights should have immediate effect.  Overall,  the language is still improved, although we think that the World Bank should expressly state its support for the UDHR.

  1. Proportionality still worrying

In our written comments to the 2nd draft, MRG and Lex Justi criticised the number of references to proportionality, which we felt created a vagueness and opened up ways for Borrowers to evade obligations.

Unfortunately, we still find them in the final version, e.g. ‘proportional to the risks to and impacts on affected communities’ when talking about  consultation (ESP, para. 53). Surely, consultation should be conducted prior to and as part of a risk assessment, and the scale of consultation processes should not be pre-determined.  This wording is particularly worrying when read conjunctively with para. 12 of ESS7, which provides that the ‘Borrower will assess the nature and degree of the expected direct and indirect economic, social, cultural (including cultural heritage) and environmental impacts on Indigenous Peoples’ without any reference to input or consultation with affected indigenous communities. The views of affected indigenous peoples will only be taken into account, pursuant to the sentence that follows in paragraph 12, namely in connection with the Borrower’s ‘project design and implementation’ but not therefore in the design of the consultation process itself.

  1. Indigenous peoples

There are many aspects of the final version concerning indigenous peoples that are worrisome.

a) ‘Indigenous Peoples/Sub-Saharan African Historically Underserved Traditional Local Communities’

As others have stated, MRG and Lex Justi are deeply concerned about the introduction of ‘Sub-Saharan African Historically Underserved Traditional Local Communities’ in addition to ‘Indigenous Peoples’, as if the two groups should be treated similarly. We note that the Introduction to ESS7 lists some of the terms used by different countries, but the World Bank has only selected one of these, ‘Sub-Saharan African Historically Underserved Traditional Local Communities,’ to use alongside the international law term ‘Indigenous Peoples’. The addition of this terminology raises numerous problems.

Firstly, the term ‘Sub-Saharan African Historically Underserved Traditional Local Communities’ does not exist in international law, nor is it underpinned by any international instrument in the same way as the term ‘Indigenous Peoples’. For instance, we wonder how the principle of self-identification, which is key to the understanding of both indigenous peoples’ rights and minority rights would be applicable.

Secondly, the phrase ‘Historically Underserved Traditional Local’ can encompass communities belonging to majorities, which may be underserved because of remote geographic locations or past marginalisation. Such communities may then have important development needs  that should be addressed, but these would be different from the structural discrimination and other challenges which continue to affect indigenous peoples and other marginalised groups. Alternatively, if all of the criteria in para. 8 of ESS7 are met, then one can wonder why the term has been introduced at all. Essentially, the introduction of this phrase undermines a standard specifically relating to indigenous peoples.

Thirdly, we cannot help but wonder why the geographic specificity of ‘Sub-Saharan African’ has been introduced here. Surely, if one accepts use of the term  ‘Historically Underserved Traditional Local’ (the intention and applicability of which we question in this context), there would be communities falling within this concept elsewhere? At the very least, the division drawn by ‘Sub-Saharan African’ feels peculiar, given that there are presumably such communities in the Sahara region and North Africa.

Finally, we must criticise the fact that such a radical change to the text was introduced into the final version without any consultation or chance for civil society to provide feedback. This is especially true when we had welcomed the open-ended terminology in para. 1 of the 2nd draft and final version of ESS7, which we feel represents good practice in including a variety of terminology that can be deemed equivalent to ‘Indigenous Peoples’, depending on local useage. Why then did ‘Sub-Saharan African Historically Underserved Traditional Local Communities’ even need to be added?

b) ‘Collective attachment’

We feel that it is positive that the World Bank includes ‘or have collective attachment to’ in para. 54 of the ESP, concerning consultation and participation. This is in recognition of the fact that there may well be indigenous communities which have previously been displaced or otherwise had to migrate, but which still remain strongly attached to their ancestral lands.  However, we are concerned that the definition of ‘Collective attachment’ in footnote 6 of ESS7 is too narrowly drafted in requiring ‘economic ties’ in addition to ‘physical presence’. The text in the footnote acknowledges the significance of ‘sacred sites,’ but the definition of ‘collective attachment’ would require these sacred sites to have economic as well as cultural importance  to indigenous communities, thus, setting a standard that in many cases could be difficult for them to meet.

The concept of ‘collective attachment’ is also elaborated on in para. 9 of ESS7 (namely that ESS7 also applies to indigenous communities that may, for a variety of reasons, have been displaced), which we feel is largely good practice and reflective of indigenous realities in many parts of the world.  We would however criticise the exception in the footnote: ‘it does not apply to individuals or small groups migrating to urban areas in search of economic opportunity.’ As we stated at the 2016 Brussels feedback session on the 2nd draft, marginalised groups migrate for a complex range of reasons, and the search for economic opportunity may well be induced by the discrimination faced back home.

c) Effects on indigenous peoples outside the project area

Moreover, the final version still does not acknowledge that indigenous communities living downstream of development projects may suffer devastating impacts, despite being outside the immediate project area. No relevant language applies in such instances, since ESS7 consistently uses ‘the project area’ to define its scope (e.g. in paras. 7, 9 & 10).

  1. Risk of total loss of culture

MRG and Lex Justi are disappointed that modifications were not made, in order to better protect against the risk of the total loss of culture, pursuant to our comments on the 1 July 2015 draft.  Thus, the final version of the Framework, in paras. 26 and 27 of ESS5, provides compensatory remedies that are formulated in purely economic terms and do not recognise that for  many indigenous communities removal from land intrinsically linked to their traditional cultures and livelihoods would essentially mean the end of those cultures within a couple of generations (e.g. pastoralists, fisher peoples, forest peoples).

We appreciate that the wording of para. 7 in ESS8 has been clarified, but the resulting language still does not adequately capture the risk to cultural heritage. In fact, we find that the language diminishes the protection of cultural heritage in providing: ‘The requirements of ESS8 apply to intangible cultural heritage only if a physical component of a project will have a material impact on such cultural heritage’.  The presence of a project can have other quite well documented consequences that affect indigenous communities, for instance through a large-scale influx of workers belonging to other communities. Thus, the reference to a ‘physical component’ narrows the application of the protections of ESS8.

  1. Free, prior and informed consent

MRG and Lex Justi are relieved that the principle of Free, Prior and Informed Consent (FPIC) as applied to indigenous peoples[2] remains in the final document (ESS7, para. 24), despite repeated threats by states that it should be removed.

However, we still feel that it is unfortunate that the final version retains: ‘There is no universally accepted definition of FPIC’ in para. 55 of the ESP. We find that this sends a negative signal, does not reflect the ever-growing international consensus and undermines the contents of the rest of the paragraph.  This sentence is particularly worrisome to us given that para. 26 in ESS7 now provides that ‘consent refers to the collective support… for the project activities that affect them, reached through a culturally appropriate process. It may exist even if some individuals or groups object to such project activities…’ The term ‘collective support’ has been criticised by indigenous peoples’ organisations for not being sufficiently rigorous, specifically, for not ‘ensuring respect for the results of affected indigenous peoples’ independent and collective decision-making processes.’[3] Thus, while the World Bank states that there is no ‘universally accepted definition’, at the same time it diminishes the consensual nature of FPIC.

In addition, a close comparison of the 2nd draft and the final version shows that there has been significant watering down of the circumstances in which FPIC applies. For instance, the word ‘adverse’ has been introduced into para. 24(a) of ESS7, as in ‘have adverse impacts on land and natural resources subject to traditional ownership or under customary use or occupation.’ Yet, indigenous peoples’ rights also can be affected by impacts that the government of the country views as contributing to development, and thus, as positive impacts. The language in the 2nd draft was simply, ‘have impacts on…’. The introduction of the word ‘adverse’ limits application of FPIC in a way that is not in line with current standards, besides which  it raises the issue: who gets to decide whether an impact will be ‘adverse’ and will affected indigenous communities have a say in that process, as they of course should?

The third category of circumstances in which FPIC applies, pursuant to para. 24(c) of ESS7, also has been weakened.  Specifically, the qualifier ‘material’ has been added to modify the relevance of cultural heritage to the community: ‘….cultural heritage that is material to the identity and/or cultural, ceremonial, or spiritual aspects of the affected Indigenous Peoples/Sub-Saharan African Historically Underserved Traditional Local Communities lives.’ This addition introduces an additional determination without the involvement of the community, thereby affecting the community’s right to FPIC.

Moreover, MRG and Lex Justi are disappointed that our feedback on the 2nd draft was not taken on board, namely concerning continued use of the word ‘significant’ earlier in para. 24(c) concerning conditions where FPIC should apply. We criticised the term ‘significant impacts’ on cultural heritage, when the word ‘significant’ is not used in the other sub-paragraphs stating when FPIC should apply. We continue to wonder how such a vague term will be interpreted  and by whom. The term allows for a determination without community involvement that an impact will not be significant, thereby blocking application of FPIC. (Similar language concerning ‘significant impacts’ is contained in para. 32 of ESS7).

More positively in para. 55 of the ESP, the following language was retained from the 2nd draft which we feel is a clear, positive statement:  ‘When the Bank is unable to ascertain that such consent is obtained… the Bank will not proceed further…’ This is reflected in para. 27 of ESS7 – again welcome.  Yet, with a diminished notion of FPIC and more limited circumstances in which FPIC applies, we are extremely concerned that the rights of indigenous peoples will not be respected consistent with international human rights law standards in this area.

  1. Individual ownership rights

The 2nd draft version of para. 21 of ESS7 was heavily criticised by us as well as by indigenous peoples’ organisations for opening up the possibility of conversion of customary land rights into individual ownership rights (in fact the first draft went even further and blocked application of ESS7 to those holding individual land titles). The reason we are alert to this issue is that such individual titling processes have repeatedly led to a weakening of customary land tenure and ultimately the displacement of communities.

While much of the final version (para. 29 of ESS7) of the paragraph is positive in that it paves the way for the legal recognition of customary land rights, Borrowers may still opt for ‘conversion of customary usage rights to communal and/or indvidiual ownership rights’. There is a caveat in a footnote, namely, ‘conversion of customary usage rights to individual ownership rights will only be an objective following consultation with the [communities] concerned and assessment of the impacts of such conversion on the communities and their livelihoods’.

We note however that this language risks falling outside the scope of FPIC obligations contained in para. 24, specifically sub-para. (a) given its new ‘adverse impacts’ language (please see above). In other words, Borrowers may determine that conversion to individual ownership rights only requires consultation because they do not view the impact to be ‘adverse’. We find the continued inclusion of possible individual ownership rights without stronger safeguards extremely worrying.

  1. An opt out remains

We are dismayed that a key opt out remains in the text,
despite repeated critique from MRG, Lex Justi and others. This is namely in the fourth level of the mitigation hierarchy identified in the Objectives and in para. 27(b) of ESS1, which states, ‘Where significant residual impacts remain, compensate for or offset them, where technically and financial feasible.’ Given that we are talking potentially about violations of Borrowers’ human rights obligations, we find it unacceptable that actions to support affected communities may depend on their technical or financial feasibility.

We also note that the term ‘significant’ has been inserted in this paragraph before ‘residual impacts,’ which introduces another subjective determination. As noted above under point 5, the use of the term ‘significant’ allows  a determination about what is ‘significant’ without community involvement thereby avoiding the application of FPIC.

Footnote 26 in ESS1 does appear to be a little stronger, however: ‘the rationale for this determination (including the options that were considered) will be set out in the environmental and social assessment.’ At least this provides for some oversight, but we worry that Borrowers may lean on this language to evade obligations to affected communities.

  1. No detailed listing of marginalised groups

Prior drafts had extensive lists of ‘disadvantaged or vulnerable groups’ (language which we have repeatedly criticised – we would have prefered the less dated and more currently  accepted term: ‘marginalised’). These lists have now largely been removed and become more open-ended, e.g. footnote 28 in ESS1 which states ‘those who may be more likely to be adversely affected…and/or more limited than others in their ability to take advantage… ‘.  The only specificity is provided by: ‘age, including the elderly and minors…’.  Similar language is contained in footnote 2 of ESS10 and the Glossary.

We note firstly that this represents a considerable departure. In prior drafts, there were lists of specific groups. In our comments, we asked that the lists be made more uniform as well as comprehensive and in line with other international standards. We feel that it is a definite step backward that the attempt to list marginalised groups has been dropped. We query the dropping of gender as a mentioned category, for instance, in the cited footnotes mentioned in the previous paragraph.

The dropping of the detailed lists has not solved the inconsistency problem. For example, particular attention to women and/or gender is paid in paras. 18 and 26 of ESS5 and para. 23 of ESS7, for instance, which makes the absence of specific mention of women in footnote 28 in ESS1, footnote 2 of ESS10 and the Glossary both puzzling and not really acceptable. Also, in para. 54 of the ESP, the term ‘Indigenous Peoples’ is used followed by the paranthetical ‘(or as they may be referred to in the national context)’, and thereby provides a different forumulation.  Further, in para. 17 of ESS5, the term ‘displaced Indigenous Peoples’ is used in the context of consultations.  (We had criticised this particular wording as restricting and undermining the provisions in ESS7.)

  1. Design of grievance mechanisms

In our comments on the 2nd draft, we noted that in para. 26 of ESS10 the design of the grievance mechanism is up to the Borrower, with no mention of consultation with the communities for whose use it is intended. The language remains unchanged in the final version (also para. 26).

More generally, there does appear to be a gap concerning consultation regarding grievance mechanisms throughout the text. No mention of it can be found in para. 34 of ESS7 (‘The Borrower will ensure that a grievance mechanism is established for the project… which is culturally appropriate and accessible….’)  or para. 21 of ESS10 (‘meaningful consultation …on project risks, impacts, and mitigation measures….’). This is frankly not current good practice, given that for example Principle 31 of the UN Guiding Principles on Business and Human Rights states that non-judicial grievance mechanisms at the operational level should be ‘[b]ased on engagement and dialogue: consulting the stakeholder groups for whose use they are intended on their design and performance….’

  1. Concluding remarks

As a last comment, given the length and complexity of the final document, we hope that the World Bank will consider the suggestion we made during the Phase 3 Consultation to formulate a guide or summary of the standards, which would include a chart visualising the interconnections between them. We feel that this is crucial in order to ensure the accessibility and understanding of the standards by not only borrowers, but also persons and groups potentially affected by World Bank funded investment projects, including indigenous and minority communities, which are of particular concern to our organisations.

[2] And ‘Sub-Saharan African Historically Underserved Traditional Local Communities’.

[3] Asian Indigenous Peoples Pact in