Implications of the World Bank's environmental and social framework for biodiversity

The World Bank is the single largest source of development finance, with wide‐reaching influence. The Bank's safeguards aim to minimize the negative impacts of the projects it funds. These policies have recently been updated in a new Environmental and Social Framework. For conservation, the key changes include a mechanism for the use of biodiversity offsets and borrowers’ own frameworks to manage impacts. Concerns have been raised that these changes may weaken protections as there is substantial flexibility about when offsets or borrowers’ frameworks can be used, and uncertainty around the efficacy of offsets. The project‐by‐project nature of these mechanisms and the lack of clear criteria may also hinder future efforts to hold the Bank to account. Concerns about these changes were raised by conservation organizations during the consultation process, but the framework's formulation does not fully reflect recommendations made. Although elements of the new policy have the potential to benefit conservation, the flexibility presents a risk to biodiversity. It is vital for conservation organizations to engage effectively to ensure that any negative impacts arising do not go unchallenged.


INTRODUCTION
withhold funding unless reforms were implemented (Wade, 1997). These safeguards were widely considered thought-leading, comprehensive, and stringent and they formed the basis of many other organizations' policies (Park, 2010), although controversy over the negative environmental impacts of Bank funded projects remained (Rich, 2013;Sommer, Shandra, & Restivo, 2017).
In October 2018, the Bank replaced the safeguards with the Environmental and Social Framework (ESF) (World Bank, 2016). The ESF consolidates formally stand-alone policies into a single framework. It sets out first a broad vision statement, then requirements applying to the Bank, and finally 10 Environmental and Social Standards (ESS) imposing requirements on borrowers (World Bank, 2016). The primary focus of the ESF is on improving borrowers' experiences by, for example, improving the efficiency of decision-making and streamlining processes (World Bank, 2010Bank, , 2012. This is framed as the Bank responding to the preferences and needs of borrower countries and a changing lending landscape (World Bank, 2012) including the growth in alternative sources of financing (Bugalski, 2016). However, the ESF also recognizes the importance of improving the environmental and social outcomes of projects and expands the policy's coverage by including more types of risks and impacts (Himberg, 2015;World Bank, 2012). This reflects the Bank's "increased awareness of the value and vulnerability of the global commons" (World Bank, 2012), and may reflect the preferences of donor governments and civil society organizations (Van Den Meerssche, 2017).
Several of the changes in the ESF are of relevance to biodiversity conservation. For example, it uses the language of ecosystem services when discussing impacts and it expands the habitat classifications, requiring consideration of impacts in human "modified" habitats (the other types being "natural" or "critical" habitats, any of which may also be in the "legally protected and internationally and regionally recognized areas of biodiversity value" category, hereafter "protected") (World Bank, 2016). However, the adoption of the ESF has led to concerns about the weakening of protections and accountability by the Bank (Bugalski, 2016;Van Den Meerssche, 2017;World Bank, 2018c). The Bank's lending portfolio and its influence on other organizations makes understanding these changes important for biodiversity conservation. Here we consider two changes that we think are most likely to impact biodiversity conservation: biodiversity offsets and the delegation to borrowers' frameworks. Further, we discuss the implications of the ESF for the accountability of the Bank.

BIODIVERSITY OFFSETS
The ESF commits borrowers to the aim of no net loss, and preferably a net gain, in biodiversity for projects located in natural, critical, and protected habitats (World Bank, 2016). It applies a mitigation hierarchy, stipulating that borrowers must first avoid, then minimize, then mitigate, and finally offset the negative impacts of a project. In modified habitats there is no commitment to no net loss and impacts need only be avoided, or minimized and mitigated "as appropriate" (World Bank, 2016). For offsets to be used, the area affected must not be ". . . unique and irreplaceable from a biodiversity standpoint," and all "technically and financially feasible" prior steps in the mitigation hierarchy must have been taken (World Bank, 2016). The adoption of the mitigation hierarchy, the prominence of offsets, and the commitment to no net loss are all changes in policy from the old safeguards.
The mitigation hierarchy has been proposed as an effective means of dealing with negative impacts to biodiversity (Arlidge et al., 2018). No net loss policies and biodiversity offsets are increasingly prominent globally (zu Ermgassen, Utamiputri, Bennun, Edwards, & Bull, 2019) and the ESF mirrors the policies of the IFC, the private-sector lending arm of the World Bank Group, which are widely considered best practice (IFC, 2012;Narain et al., 2020). Biodiversity offsets aim to bring benefits to biodiversity by counterbalancing the losses from development; however, they remain controversial (Benabou, 2014;. There is currently insufficient empirical evidence to determine whether or not offsets are effective in delivering the required outcomes (zu Ermgassen et al., 2019). Their success is likely heavily dependent on their design, such as governance and funding arrangements, and local factors, such as land availability and biodiversity trends Sonter et al., 2020). Even well-designed offsets will likely entail some residual uncompensated negative impacts (zu Ermgassen et al., 2019).
During consultation on the new policy, the inclusion of the mitigation hierarchy was welcomed by some conservation organizations, but others argued that discretionary and inconsistent language weakens the policy (see Appendix S1: Table S1 for details). For example, ESS1 ("Assessment and Management of Environmental and Social Risks and Impacts") requires offsets only when they are "technically and financially feasible" (World Bank, 2016a). This caveat is not repeated in ESS6 ("Biodiversity Conservation and Sustainable Management of Living Natural Resources"), which requires offsets when residual impacts remain and "where appropriate," (Table S1; World Bank, 2016a) and is weaker than the IFC's equivalent policy (Performance Standard 6, hereafter PS6) (IFC, 2012). Concerns were also raised that the lack of a requirement not to proceed if residual impacts cannot be offset risks creating a perception that all risks can be managed (Table S1). Three submissions also argued that the ESF was overreliant on offsetting, a controversial management tool (Table S1).
The ESF has few restrictions on offsets' use and design. Instead decisions are made on a project-by-project basis by a "qualified expert," following "Good International Industry Practice" (World Bank, 2016). Examples of good practice that could be followed are given in the policy's guidance notes (World Bank, 2018b), mirroring PS6 (IFC, 2012). For critical habitats and any habitat in the protected category, experts must be independent and internationally recognized, but for other habitats they can be Bank staff (World Bank, 2016) who may be incentivized to approve projects whenever possible (Weaver, 2007), a potential conflict of interest. In PS6, experts for offsets must always be external (IFC, 2012). Experts' bias could be an issue even if they are independent, as they may be reliant on Bank staff for future work (Benabou, 2014).
The principle of additionality is key to offsetting, meaning that the biodiversity gains must be directly caused by the offset and would not have happened anyway, otherwise there is no net improvement in biodiversity . The ESF states that an offset must be "designed and implemented to achieve measurable, additional, and longterm conservation outcomes" (World Bank, 2016). However, there are no specific protections against cost shifting, a situation where offset funding replaces rather than supplements national funding commitments for conservation (IUCN, 2016;Ledec, Johnson, Lovei, Warner, & Parker, 2016;Pilgrim & Bennun, 2014). Cost shifting undermines additionality and could lead to conservation becoming dependent on money from offsets, creating a perverse incentive for development to ensure protection (IUCN, 2016;Pilgrim & Bennun, 2014). This is a particular issue given that the borrowers are national governments with existing commitments to achieve conservation targets, who therefore could use offset funding to achieve those goals instead of undertaking new actions (Maron, Gordon, Mackey, Possingham, & Watson, 2016). This problem could be addressed with policy mechanisms  but none are included within the ESF.
The ESF allows for offsetting to be used when projects impact IUCN class I-IV protected areas (those with strict protections). This is in conflict with the limits on offsetting proposed by the IUCN's guidance (IUCN, 2016). Permitting offsetting on strict protected areas is recognized as a source for concern by the Bank (Ledec et al., 2016), and was raised during the consultation process (Table S1). Two arguments may be made in support of offsetting in these contexts. First, projects impacting protected areas are required to be consistent with the legal protection status and government recognized management plans (World Bank, 2016). Second, development might be seen as inevitable, and it is therefore better to have offsets than not (Ledec et al., 2016). Offsetting of impacts to strict protected areas is recommended by the relevant IUCN resolutions, however development should be only permitted in these areas "in the most exceptional circumstances where overwhelming public interest (such as maintenance of traditional livelihoods) requires it" (IUCN, 2008). Nevertheless, the uncertainty about whether offsets are effective means that their use in these highly sensitive and important areas represents a significant risk and is a reasonable cause for concern (Table S1). Furthermore, given the Bank's wide-reaching influence, by permitting offsetting in these areas, the ESF could drive an increase in their use by legitimizing the practice.

BORROWERS' FRAMEWORKS
The ESF introduces a mechanism that allows borrowers to use their own frameworks to assess and manage the environmental and social impacts of a project (World Bank, 2016). Borrowers' frameworks are described as the elements of a country's laws, regulations, policies, and systems that are relevant to the risks and impacts of the project (World Bank, 2016). The assessment of the adequacy of a borrower's framework is carried out on project-by-project basis by Bank staff (World Bank, 2018a). If they determine that a borrower's framework is adequate the ESF does not apply further (World Bank, 2016). This is a significant shift from the existing policy where projects were assessed against the requirements set out in the safeguards (Van Den Meerssche, 2017). During the consultation, submissions from individuals representing borrowers welcomed the introduction of borrowers' frameworks (Table S1). However, others raised a number of concerns (Table S1). For example, some questioned the capacity of borrowers to offer the same level of rigorous assessment and monitoring (Table S1). The strength of national policies dealing with impacts to biodiversity varies, many low-income countries, a focus of Bank lending, have much weaker requirements as can be seen in their limited uptake of no net loss policies (zu Ermgassen et al., 2019). Four submissions, from a range of organizations, argued that the delegation of responsibility for environmental and social standards was risky and therefore should not be allowed for high-risk projects either indefinitely or for an initial period after the ESF is adopted (Table S1).
To meet the standard required for delegation, the borrower's framework must be judged "likely to address all relevant risks and impacts," and must "enable the project to achieve objectives materially consistent with the ESS" (World Bank, 2016). However, despite calls during the consultation for a clear definition and rigorous criteria for assessing frameworks (Table S1), no definition of "materially consistent" is provided. Instead the Bank has argued that what constitutes being "materially consistent [. . . ] is context-specific," and therefore a standard definition cannot be given (World Bank, 2018c). Typically, projects funded by the Bank have greater requirements for assessment, consultation and compensation than those implemented under national laws, so they are more costly and take longer to be approved (Buntaine, 2016;World Bank, 2010). This creates a strong preference amongst borrowers to use their own frameworks, but it also creates a conflict of interest for the Bank staff who are charged with assessing the frameworks and are incentivized to approve projects as part of their performance metrics (Weaver, 2007). These conditions risk encouraging staff to subvert policies to speed up the approval processes (Rich, 2013), a phenomenon that has previously been reported by Bank staff in anonymous interviews (Buntaine, 2016). Indeed, the problem of the Bank's internal organizational culture has been raised as a barrier to safeguard effectiveness (Buntaine, 2016;Rich, 2013;Weaver, 2007;World Bank, 2002), but the ESF does not address this issue; rather it increases the ability of Bank staff to exercise discretion (Passoni, Rosenbaum, & Vermunt, 2016).
The ESF's delegation criteria are less stringent than in the piloting of borrowers' frameworks, which required "equivalence" with principles that matched the Bank's safeguards (Bugalski, 2016;World Bank, 2011). Although that pilot recommend moving away from an equivalence measure, it also concluded that clear assessment criteria, communicated to all stakeholders and benchmarked against existing international agreements, are vital (World Bank, 2011). Calls for these kind of assessment criteria were made during the consultation process (Table S1) but are lacking in the ESF. Historically, the Bank has justified involvement in controversial projects on the basis that it can improve standards through its policies (Rich, 2013;Wade, 1997). Although the ESF has mechanisms to assist countries in strengthening their systems by including training and capacity building as part of a project, this is intended to achieve "material consistency" so the borrower's own framework can be used in the future (World Bank, 2016). The ability of the Bank to improve standards in development has therefore been diminished by the ESF and the adoption of borrowers' frameworks could ultimately result in weaker standards.

HOLDING THE BANK TO ACCOUNT
A wider consequence of the adoption of the ESF may be a reduction in the ability of civil society to hold the Bank to account. The delegation mechanism within the ESF has the potential to dilute the authority of the Inspection Panel (the Bank's independent grievance mechanism), making legal proceedings against the Bank more difficult (Bugalski, 2016;Van Den Meerssche, 2017). The lack of strict requirements within the ESF and the use of project-byproject agreements make it difficult to prove the Bank has failed to meet its obligations (Bugalski, 2016). It is no longer sufficient to show that a project has failed to meet the requirements of an element of Bank policy, and now complainants would have to show that the project could never have been expected to meet the policy's standards in an acceptable timeframe and manner (Van Den Meerssche, 2017). For the Bank to have met its requirements under the ESF it need only demonstrate proficient assessment, which could be argued even in cases where a project failed to comply with all of the policy's environmental standards (Van Den Meerssche, 2017). The Inspection Panel has historically used a standard of reasonableness to judge if the Bank has met its duty of care, regardless of the specific language in policies (Passoni et al., 2016). Causal links between the Bank's failure on due diligence and the negative impact of a project caused by a borrower's actions have previously been identified (Passoni et al., 2016). Therefore, if the Panel is allowed to continue as it has done before, delegation might not impede accountability (Passoni et al., 2016). However in recent years the Panel has been unwilling to launch formal investigations, giving Bank management greater leeway (Bugalski, 2016). Additionally, given the significant shift in policy, Bank management and the board may look to reduce the remit of the Panel when borrowers' frameworks are used (Passoni et al., 2016). In the absence of empirical evidence, the ultimate impact of the ESF on accountability of the Bank remains to be seen.

CONCLUSIONS AND CONSERVATION'S RESPONSE
The ESF is a more comprehensive policy than the previous safeguards, requiring consideration of a greater number of impacts, but it is also a very flexible policy that focuses on outcomes rather than strict requirements and allows for project-by-project agreements. It represents a significant shift in approach for the Bank. Elements of its policy mirror those widely accepted as current best practice and its flexibility may be desirable for some stakeholders, particularly borrowers. However, it has also raised concerns from several conservation organizations (Table S1). The policy's requirements for the use of biodiversity offsets and delegation to borrowers' frameworks have drawn criticism for being vague and overly discretionary. If the mitigation hierarchy is well-applied and no net loss achieved, and if borrowers' frameworks provide equivalent or better environmental safeguards, then the broader scope of the ESF could bring significant benefits for biodiversity. However, without a sufficiently clear policy, these potential benefits remain uncertain and there is a risk that they could result in the Bank funding more projects that negatively impact biodiversity.
The ESF also represents missed opportunities to strengthen protections. In line with the previous policy, the ESF only applies to investment projects and not to Development Policy Lending (formally Structural Adjustment Loans) which represent around half of all Bank lending, and have historically had significant environmental impacts (Sommer et al., 2017). A global infrastructure boom is predicted in coming decades, with initial evidence suggesting much of this infrastructure will be constructed without strong environmental safeguards (Narain et al., 2020;zu Ermgassen et al., 2019). Because of this it is important, now more than ever, for publicly funded institutions such as the World Bank to drive up standards in development. Several elements in the ESF are a positive contribution to this aim but others appear to fall short.
What impact the ESF will actually have remains to be seen, but several elements are a cause for concern for biodiversity. Looking forward, those stakeholders who share these concerns need to continue to engage effectively with the Bank, its donors and the recipients of its funds in order to ensure that the highest standards of environmental protection are maintained in projects supported by the Bank. Such engagement by civil society groups has previously been a driving force of reform at the Bank. The projectspecific nature of the ESF will require project-specific responses, led by the relevant local stakeholders, to ensure that any harm caused does not go unchallenged.
To support these efforts, there is a need for interdisciplinary and transdisciplinary research that brings together an understanding of the political and legal mechanisms that can most effectively influence Bank stakeholders, with new methods for identifying geographical areas of concern, anticipating risks, and evaluating the impact of development on conservation outcomes. By better understanding these conflicting demands and opportunities, conservation organizations can improve their ability to identify the most important risks and engage constructively with the Bank's stakeholders to achieve positive outcomes for biodiversity and people.

A C K N O W L E D G M E N T S
JM is supported through the Natural Environment Research Council's E3 Doctoral Training Partnership (grant NE/L002558/1) in partnership with the Royal Society for the Protection of Birds.
We would like to thank the two anonymous reviewers for their comments which improved the manuscript.

A U T H O R C O N T R I B U T I O N S
JM conducted the review of policy documents and literature and wrote the initial manuscript. JM, GB, and AK conceived ideas and planned the paper. All authors assisted in developing the ideas, contributed to the writing of the final manuscript, and gave final approval.