The UK’s support to the World Bank’s International Development Association (IDA)

1. Introduction


The International Development Association (IDA) is part of the World Bank Group and provides one of the most important multilateral channels for development assistance for the world’s poorest countries. IDA’s funds are replenished every three years from various sources, including bilateral donors, following high-level negotiations on the volume and terms of the replenishment. The 19th round of IDA’s replenishment (IDA19) was intended to cover the period from July 2020 to June 2023. However, a decision to fast-track IDA19 spending in order to respond to the COVID-19 pandemic meant that there would be a shortfall in the financing for the final year of IDA19. IDA shareholders therefore agreed to bring forward the IDA20 replenishment by a year. This process was completed in December 2021 and resulted in a $93 billion funding package being agreed for the period from July 2022 to June 2025.[1]

The UK was the largest bilateral donor to IDA during IDA15, IDA17, IDA18 and IDA19. It committed £2.52 billion for IDA18 (2017-20) and £3.06 billion for IDA19 (originally 2020-23), representing 13%[2] and 12%[3] of total donor funding respectively. For IDA20, however, the UK has reduced its funding commitment by 54% (to £1.41 billion) compared to that for IDA19, which means that it will be the third-largest donor to IDA during this funding round.[4] Nevertheless, the UK remains a major donor to IDA, which gives it an important voice in setting the priorities for each replenishment round. It seeks to ensure that IDA’s operations are aligned with UK development priorities, that it makes the most effective use of its resources and that the World Bank continues to build its own capacity as a development agency.[5]

The UK government is a strong supporter of the multilateral aid system, which it sees as an important channel for both international development finance and humanitarian aid. Just over a third of UK aid is provided as core funding to multilateral development organisations,[6] and IDA was the second-largest channel of this multilateral assistance, with 18.6% of the total.[7] UK government reviews highlight that the World Bank demonstrates a strong strategic fit with UK aid objectives.[8]

The Independent Commission for Aid Impact (ICAI) review assesses the value for money of the UK’s contribution to IDA and how well the UK has used its status as the largest bilateral donor to encourage continuous improvement in the World Bank’s performance. The review was carried out in parallel to the UK’s preparation for the IDA20 replenishment process and was intended to encourage the Foreign, Commonwealth and Development Office (FCDO) to clearly articulate its objectives, strategies and approach to IDA.

Purpose of the literature review

This literature review explores findings from the literature relating to the following questions:

  1. How well has IDA delivered its intended results (outputs and outcomes) through its operations?
  2. How well are IDA objectives on the selected focus themes of fragility and conflict, climate change and COVID-19 response delivered in practice?
  3. How well are IDA objectives on equity, inclusion and safeguarding delivered in practice?
  4. How well are the needs and voices of governments and poor people in partner countries reflected in IDA programmes?
  5. How well has IDA mobilised other sources of development finance, including from within the World Bank Group?

The literature review explores the evolving policy, functioning and effectiveness of IDA and debates about its model, strategic focus and development role. It covers World Bank and Independent Evaluation Group (IEG) publications, and academic and grey literature published by a range of organisations, including from the Global South. It addresses key themes including: IDA’s effectiveness in delivering development impacts, including in relation where possible to other multilaterals (through, for example, existing comparative analyses by the Quality of Official Development Assistance Index and the Multilateral Organisation Performance Assessment Network); the background, nature and impact of IDA’s work in relation to fragility and conflict, climate change adaptation/resilience, responses to COVID-19, equity and inclusion, and environmental and social safeguards; the evolution and impacts of IDA’s hybrid financing model; and the extent to which IDA programming reflects the needs and voices of governments and poor people in partner countries. The review primarily focuses on the overall evolution and effectiveness of IDA. The extent to which the UK contributes to and influences IDA’s work is covered in other components of the wider ICAI review and is therefore not discussed in the literature review.

Approach to evidence

The literature review has consulted 162 published documents, covering English-language corporate documents, policies, reports and evaluations produced by the World Bank Group, the IEG, other multilaterals and donor governments, as well as analyses and research studies undertaken by civil society and academic researchers, the majority since 2015. Relevant literature from the academic fields of economics and political economy has also been consulted. The methodology involved searches on the websites of international organisations and research institutes as well as social science databases and Google Scholar to identify further relevant literature. Search terms were derived from the core research questions and inclusion criteria were based on the extent to which publications provided useful and credible information against these.

The main limitation of the review is that independent analyses of the nature, effectiveness and impact of IDA’s work rely heavily on the results reported by its internal Results Measurement System (RMS). Much of the primary evidence regarding IDA’s effectiveness is restricted to the indicators included in the RMS, which are mostly quantitative and provide only a partial view of the breadth and depth of IDA’s development impact. Civil society and academic researchers have access to corporate documents, policies, reports and evaluations produced by the World Bank through the Bank’s Open Data repository, but lack the necessary access to undertake primary research into the effectiveness of IDA’s work. Most independent sources consulted for the literature review analyse and interpret evidence generated by the RMS, including those which compare IDA’s effectiveness to other international financial institutions (IFIs) and multilateral development banks (MDBs). Independent analysis on IDA, as opposed to aid more broadly, is also disproportionately generated from the Global North and is published by a narrow group of research institutions based either in Washington, DC (Centre for Global Development, Brookings Institution), London (Overseas Development Institute) or Paris (Organisation for Economic Cooperation and Development). As such, most of the evidence covered in this review is drawn either from these organisations or from the World Bank Independent Evaluation Group. While we consider this a limitation, many of the IEG evaluation reports referenced in this review use results frameworks other than the RMS and draw on wider sources of complementary evidence (for example, academic articles) when making assessments of IDA’s effectiveness.

Report structure

In Section 2, we provide an overview of IDA and briefly explain its history, mandate, geographic focus, financial model, and deployment model. We also highlight how IDA differs from other IFIs and MDBs as well as, more broadly, how multilateral aid differs from bilateral aid. In Section 3, we outline the evidence regarding the overall effectiveness of IDA in delivering development impacts and explore key debates about its strategic focus and development role. In Section 4, we explain the background and nature of IDA’s work in key focus themes relevant to our review questions (fragility and conflict, climate change, COVID-19, equity and inclusion, and environmental and social safeguards) and outline the evidence regarding the relative effectiveness and impact of IDA within each of these themes. In Section 5, we explain how IDA’s financial model has evolved during the review period and explore key debates about the impacts of leveraging capital market issuances and private investment.

[1] Additions to IDA resources: twentieth replenishment, World Bank, 2022, p. xii, link.

[2] Additions to IDA resources: eighteenth replenishment, World Bank, 2017, p. 143, link.

[3] Additions to IDA resources: nineteenth replenishment, World Bank, 2020, p. 165, link.

[4] Additions to IDA resources: twentieth replenishment, World Bank, 2022, p. 197, link.

[5] IDA18 business case, Department for International Development, undated, link.

[6] Statistics on international development: final UK aid spend 2020, Foreign, Commonwealth and Development Office, p. 4, link.

[7] Statistics on international development: final UK aid spend 2020, Foreign, Commonwealth and Development Office, p. 46, link.

[8]  Raising the standard: the Multilateral Development Review 2016, Department for International Development, December 2016, link.

2. Overview of IDA

Multilateral aid landscape

Aid flows between two governments or between two other official national sources are classed as bilateral aid, whereas aid flows from governments to international or ‘multilateral’ agencies are classed as multilateral aid (OECD, 2019). Multilateral organisations are diverse in their mandates, but collectively offer certain advantages to donors as a channel for aid, including their ability to concentrate resources and expertise, promote global goods, and address development challenges at a global scale (Greenhill & Rabinowitz, 2016). It has also been suggested that they provide a greater degree of political neutrality and insulation from donor economic interests than bilateral development organisations do (Gulrajani, 2016).

A third of global aid from Organisation for Economic Cooperation and Development (OECD) governments goes towards core funding to multilateral development organisations (OECD, 2020), but bilateral official development assistance (ODA) is also often channelled through multilaterals if allocated to specific earmarked purposes agreed with the bilateral donor. 15% of all multilateral contributions from OECD countries in 2018 were earmarked, rather than core, contributions (Bosch et al., 2020). This so-called ‘multi-bi’ funding is classified as bilateral ODA and is typically achieved through a variety of trust funds and similar instruments. Through such instruments the UK provided earmarked contributions to the World Bank in 2018 that eclipsed its core contributions to IDA by approximately a third (Barder et al., 2019).

Multilateral agencies comprise different types of organisations. Global funds and specialised agencies (for example, the United Nations Children’s Fund and the Global Fund to Fight Aids, Tuberculosis and Malaria) typically provide grants (non-repayable, interest-free) for developing countries to combat specific issues. Multilateral development banks (MDBs) and other international financial institutions (IFIs) (for example, the International Monetary Fund) provide loans and grants for a broader range of development programmes, which are provided on either commercial (ie ‘market-based’) or non-commercial (ie ‘concessional’) terms.

Individual IFIs differ in their geographic scope (that is, the countries and regions they provide financing to) and mandate (primarily, the extent to which they seek to generate a financial return on their investments and can operate without sovereign guarantees). Most IFIs have two major funds, often referred to as windows or facilities. One lending window is used to provide financing on market-based terms, typically in the form of loans but also through equity investments and loan guarantees. The other is used to provide financing on concessional terms, typically to middle-income and low-income governments and, in some cases, private sector firms.

The World Bank is the oldest and largest of the IFIs. Its creation at Bretton Woods in 1944 responded to an identified market failure in long-term capital flows into poor and war-torn countries (Birdsall & Morris, 2016). Today, the World Bank Group includes three IFIs that make loans and grants available to different classifications of developing countries and non-state actors, targeting differing degrees of financial return.[1] The International Bank for Reconstruction and Development (IBRD) provides non-concessional financing to middle-income country governments and some creditworthy low-income country governments. The International Finance Corporation (IFC) provides non-concessional financing to private sector firms, mostly in middle- and low-income countries. The International Development Association (IDA) provides concessional financing to governments in the world’s poorest and least creditworthy countries. In the following section, we briefly explain the history of IDA, its mandate and geographic focus, as well as the way in which it raises financial resources (financing model) and deploys them with the aim of achieving development impact (deployment model).

History, mandate and geographic focus

IDA was created by the members of the World Bank in 1960 as a tool to focus its efforts more explicitly on the economic development of the world’s poorest countries.[2]

“The purposes of the Association are to promote economic development, increase productivity and thus raise standards of living in the less-developed areas of the world included within the Association’s membership, in particular by providing finance to meet their important developmental requirements on terms which are more flexible and bear less heavily on the balance of payments than those of conventional loans, thereby furthering the developmental objectives of the International Bank for Reconstruction and Development (hereinafter called “the Bank”) and supplementing its activities.” (IDA Articles of Agreement, Article I, 1960)

Since 1960, IDA has grown to become the most important multilateral channel of concessional (low-cost) development finance for poor and fragile countries across the globe. It will provide finance of $93 billion in the three-year cycle beginning in 2022 (World Bank, 2022). No other single provider has committed to this level of concessional finance (roughly $31 billion per year) for IDA’s target group of poor and conflict-afflicted countries.

Some other significant providers of development finance, such as the African Development Bank, also focus on promoting economic development within countries with the highest level of need within their region, while others, like the European Investment Bank, make total amounts of finance available in a similar way to IDA. However, no single provider can match IDA for both the overall size of its finance package and the concessionality of the finance it provides.

IDA’s mandate specifically focuses on countries with the highest levels of need, which are often the hardest to reach. Eligibility for IDA support is determined firstly by relative poverty, defined as GNI per capita below an established threshold which is updated annually ($1,205 in FY2022). Eligibility is also defined by creditworthiness. IDA supports some countries above the income threshold which are deemed to lack the creditworthiness to access capital for development from other sources.[3] Countries which meet both thresholds are referred to as ‘blend’ countries. There are currently 74 IDA-eligible countries, 15 of which receive a blend of funding from both IDA and IBRD. Over its history, 46 countries have graduated from IDA funding – including 12 since 2010 and five since 2015[4] – although nine returned to eligibility due to economic or other shocks. The US Congressional Research Service (2020) reports that the top five recipients of loans and grants from IDA in FY2019 were Ethiopia ($1.3 billion), Pakistan ($1.2 billion), Mozambique ($1.2 billion), Côte D’Ivoire ($950 million) and the Democratic Republic of Congo (DRC, $760 million) (US Congressional Research Service, 2020). While there are other global IFIs, these mainly focus on middle-income and low-income countries which are deemed more creditworthy than IDA-eligible countries (for example, IBRD).

Financing model

Most IFIs mainly provide financing on non-concessional terms and these investments generate a financial return through loan interest charges, as well as economic and social benefits to varying degrees. They typically raise most of their financial resources from international capital markets (Humphrey, 2014). However, many IFIs also have concessional windows which usually do not leverage capital market borrowing and need to be regularly replenished by donors or shareholders, in addition to repayments of past loans on soft credit terms (Humphrey, 2014). IDA’s concessional loans and grants to the world’s poorest and least creditworthy countries, like similar terms offered on a smaller scale by some sister institutions like the African Development Fund, are less able to generate a financial return compared with harder, market-related windows.[5] It has therefore historically raised most of its additional resources through donor government contributions on a cash in, cash out basis, replenishing its funds every three years.

The four most recent replenishment rounds are IDA17, IDA18, IDA19 and IDA20. IDA17, or the 17th replenishment of IDA, made financing available from July 2014 to June 2017. IDA18 made financing available over the subsequent three years (from July 2017 to June 2020). IDA19 was intended to make financing available between July 2020 and June 2023, but its commitment authority will be exhausted by mid-2022, due to the provision of additional financing to support the COVID-19 response in poor countries in 2020 and 2021. IDA20 was therefore negotiated and agreed during 2021, and will provide fresh finance between July 2022 and June 2025.

In 2017, IDA introduced a new hybrid financial model by issuing bonds on capital markets against its own equity (ie loan receivables) for the first time. IDA now raises its resources through a combination of donor contributions, debt repayments, internal transfers from other parts of the World Bank Group and borrowing. More than 70% of the 19th replenishment of IDA was funded from resources other than donor contributions, compared to 65% under IDA18 and between 40 and 50% under the six previous replenishments (World Bank, 2020a). This has enabled it to remain the largest multilateral source of concessional finance for poor countries while reducing the share of donor contributions (Landers, 2021). In recent years, however, other sources of development finance pursuing their own agendas in IDA countries (for example, China) have become more prominent. As a result of this and the increased willingness of IDA countries to borrow on international capital markets, the share of IDA financing as a proportion of overall financial flows into IDA countries has declined over the last decade.

IFIs’ lending capabilities are constrained by their level of equity (that is, the amount of available funds they have to pay off bondholders in the event that borrowers are unable to repay their loans). Different proportions for this so-called ‘leveraging’ (or ‘gearing’) are invoked by market analysts and rating agencies, and for related reasons IFI managements and boards, to assess what prudent levels of borrowing and lending they should undertake. IDA’s donor contributions, transfers, outstanding loan repayments and accumulated reserves, along with its terms offered to borrowing countries, therefore jointly determine how much it can afford to lend, but the model it uses to calculate this is internal and complex. Some argue that other IFIs could safely leverage their equity more to increase the amount of financing available for development. Humphrey (2020) finds that, for every $10 in outstanding loans, IFIs generally hold up to four times more in equity (between $2 and $6) than most commercial banks ($1.50). Nevertheless, ‘soft’ (that is, largely concessional) funds such as IDA are not considered to be in a position to leverage equity to the degree of other financial institutions that lend on more commercial terms partly, because the former also require subsidies to bridge gaps between their market-related borrowing costs and softer lending terms.

Deployment model

IDA has traditionally made finance available to eligible countries through a combination of programmatic support to national budgets (also referred to as Development Policy Financing (DPF) or budget support, now enhanced by some ‘results-based’ or ‘payment for results’ (P4R) disbursement arrangements) and development project loans or grants (also referred to as operations). Under IDA17, 78% of total IDA disbursements were allocated to operations, 12% to DPF and 10% to P4R (Miller et al., 2021). Under IDA18, the share of operational disbursements declined to 68% as DPF and P4R were scaled up to 18% and 14% of total disbursements respectively. This split remained the same in FY2021, the first year of IDA19.

IDA’s lending terms are concessional; they often include no or low interest charges and offer long maturity and grace periods to make IDA financing more affordable for countries facing development challenges and public finance shortages. The majority of IDA’s concessional funding is allocated to countries on the basis of Performance-Based Allocations (PBAs). PBAs are calculated according to the quality of countries’ policies and institutions – based on the results of the World Bank’s Country Policy and Institutional Assessments – as well as their population and level of income.

IDA also softens its financing terms to countries facing high levels of debt distress. The Debt Sustainability Framework (DSF) is the main tool used by the World Bank, the International Monetary Fund and other IFIs to assess countries’ debt-carrying capacities, and IDA translates DSF assessments into ‘traffic lights’ which determine financing terms available to each country. First introduced in 2005, the most recently revised DSF for low-income countries became operational in 2018.

Against the backdrop of the ‘bilateralisation of multilateral aid’ in recent years (that is, the increasing proportion of bilateral aid channelled through multilaterals for earmarked purposes agreed with the bilateral donor) (Tortora & Steensen, 2014), IDA is also one of the largest and most important sources of unearmarked multilateral finance for recipient governments. This is because IDA’s donors are unable to earmark contributions for specific purposes. For the most part, donor contributions to IDA are pooled together and allocated to countries according to their income levels and a performance-based incentive system designed to promote effective management of aid inflows and policy reforms. This allows recipient governments the flexibility to direct IDA financing to areas they consider to be in greatest need.

Under pressure from donors including the UK, recent replenishments have seen the proportion of finance allocated under the traditional allocation framework reduced in favour of ‘windows’. These windows set aside additional financing to support countries facing challenges that require additional resources more quickly, or for more specialised purposes, than possible under the regular allocation framework. Around 27% of funding during IDA18 was allocated to countries through special funding windows. These windows are the mechanisms through which IDA earmarks financing for specific purposes, although donors are not able to earmark contributions for specific windows. The Regional Window was established under IDA13 to finance regional integration objectives. The Crisis Response Window was set up under IDA15 to finance recovery from natural disasters, public health emergencies and economic shocks. The Refugee Sub-Window was introduced under IDA18 to support countries hosting large numbers of refugees. Around 27% of total funding during IDA18 was allocated to countries through special funding windows, envelopes or set-asides with their own allocation rules (World Bank, 2021g).

In 2017, IDA introduced the Private Sector Window (PSW) to mobilise private sector investments in IDA countries. The PSW is IDA’s first blended finance instrument that aims to mobilise private sector investments in underserved sectors and markets by reducing their financial risk. For example, it provides guarantees for the types of large-scale private development projects that other traditionally more private sector-focused arms of the World Bank invest in. Section 5.2 of this review discusses the PSW model in more detail, but the overarching rationale is that deploying limited amounts of concessional financing improves the risk-reward profile of development projects in poor countries. This enables the World Bank to increase the levels of private sector investments in countries which would otherwise be seen as too risky for private sector investors under strictly commercial terms. IDA allocated $2.5 billion for the PSW under IDA18, which the World Bank estimates has allowed IDA to leverage other World Bank Group investments by a ratio of approximately 1-to-6 in challenging markets (World Bank, 2020a). However, leveraging private investments into IDA countries is a more contentious ambition and has proved challenging in practice. Critics dispute the deployment of scarce, concessional donor resources to support profit-seeking private firms at the expense of the poorest and most vulnerable (Kenny, 2021), although supporters argue that private investment generates positive economic and social benefits that warrant the use of concessional funds to offset the costs of pioneering investments in poor and fragile country settings (Sacchetto et al., 2021).

In addition to levels of need and creditworthiness, there are broader policy conditions which recipient governments must adhere to in order to access IDA financing, which largely stem from the World Bank’s global strategy and, by extension, the priorities of its donors. In order to access IDA financing, recipient countries often need to enact structural reforms historically associated with ‘Western’ economic theories of effective economic management, historically referred to as the ‘Washington Consensus’ (Birdsall et al., 1993). The term Washington Consensus was coined by the economist John Williamson in 1989 and refers to a set of ten market-oriented, small-state policies (for example, deregulation, liberalisation, austerity and privatisation) for improving economic performance (Archibong et al., 2021). These policies were initially prescribed to Latin American countries by the Washington-based IFIs (that is, the World Bank and the International Monetary Fund) and later to African countries through ‘Structural Adjustment Programmes’, often as conditions for financing, including from IDA (Archibong et al., 2021). The legacy of conditions in these countries remains fiercely contested; many believe the economic reforms they were used to promote have turned out to be harmful (Easterly, 2000), and the World Bank claims to have adapted its allocation systems over time in response to early literature findings that Washington Consensus policies failed to improve economic conditions in recipient countries (Cormier & Manger, 2020). However, many popular observers claim that the World Bank has never truly abandoned the Washington Consensus, constraining national sovereignty (Elliott, 2016; Guven, 2018).

[1] The Multilateral Investment Guarantee Agency and the International Centre for Settlement of Investment Disputes are not IFIs but also form part of the World Bank Group.

[2] As measured by GNI per capita below IDA’s operational threshold ($1,185 for FY2021).

[3] Countries above the income threshold for over two years but not yet deemed creditworthy enough to manage the increases in lending costs associated with IBRD-only status are classified as ‘gap’ countries. A positive creditworthy assessment leads to reclassification as ‘blend’.

[4] Recent graduates have included China (1999), Indonesia (2008), India (2014), Bolivia, Sri Lanka and Vietnam (all

2017) and Moldova and Mongolia (both 2020).

[5] Until recently, the Asian Development Bank implemented a separate concessional window similar to IDA and the African Development Fund (the Asian Development Fund), but took the decision to merge this window with its standard lending facility into a single balance sheet in April 2015.

3. Effectiveness of IDA in delivering development impacts

Development impacts

The most common approach in the academic literature to measure aid effectiveness is to study the macroeconomic impacts of aid in countries which receive it in large quantities (Qian, 2015). Numerous studies have looked at the macroeconomic impact of aid within as well as across countries, but depending on the context of the study and the measures and empirical strategies deployed, results can vary widely from finding that aid can be impactful (that is, it lifts people out of poverty) to being harmful (that is, it keeps people in poverty). The author of The white man’s burden and the elusive quest for growth, William Easterly, is one the leading sceptics about foreign aid and claimed in 2006 that “[in the past forty years] $568 billion [has been] spent on aid to Africa, and yet the typical African country is no richer today than 40 years ago” (Easterly, 2006).

The lack of consensus in the academic literature is likely due to the tendency for some studies to look at the relationship between aid and aggregate economic growth (Dollar & Burnside, 2000; Easterly, 2003; Galiani et al., 2014), while others look at the relationship between aid and other indicators of development. The impact of aid on aggregate economic growth has been widely studied; it is common to divide this indicator by a country’s population to assess how developed an economy is and track this alongside total and, in some cases, particular kinds of aid flows to countries. However, the limitations of aggregate growth (for example, its failure to reflect how most people actually live or who benefits) are also widely acknowledged. Many studies of aid effectiveness therefore look at the effects of aid on other indicators of development or underdevelopment, such as conflict (Nunn & Qian, 2014b; Crost et al., 2014), corruption (Svensson, 2000) and democracy (Djankov et al., 2008). A related body of literature looks at the relationship between aid and recipient government spending, tax revenue and borrowing. Evidence from these studies suggests that aid results in higher public expenditure, but also in tax revenue decreases and increases in public sector debt (McGillivray & Morrissey, 2001). Overall, there is a broad but not universal consensus that the right types of aid in the right conditions work well to promote growth and social development, and vice versa.

IDA’s Results Measurement System (RMS) reports on IDA’s achievement of intended results at three levels, or ‘tiers’ (World Bank, 2020c). Tier 1 measures changes in high-level development indicators (such as poverty, human development, GDP per person employed) in IDA countries. The World Bank does not claim that changes in tier 1 indicators are attributable to its activities, given the many other causal factors in play, including regime changes (Mavrotas & Outtara, 2003; Mavrotas, 2002; Islam, 2005), fluctuations in the prices of natural resources (Harford & Klein, 2005) and global pandemics (Bottan et al., 2020). For example, the World Bank’s 2015 Sierra Leone completion and learning review states that GDP contracted by 21% in 2015, which the Bank’s Independent Evaluation Group (IEG) attributed to an Ebola outbreak and a fall in iron ore prices (IEG, 2020e). As noted by the RMS: “while it is relatively easy to identify an activity’s outputs, because they follow relatively immediately from the activity, [tier 1] outcomes are farther away on the results chain and also dependent on other factors not under the project’s control” (World Bank, 2020c, p. 16).

Due in part to the challenges of attribution (that is, establishing a convincing link between aid and changes in high-level development indicators), economists such as Abhijit Banerjee and Esther Duflo advocate a more detailed approach to studying aid effectiveness. Their influential book, Poor economics: a radical rethinking of the way to fight global poverty, calls for the use of experimental methods such as randomised controlled trials to study the effects of aid on specific outcomes within recipient communities (Banerjee & Duflo, 2012). While rigorous experimental and quasi-experimental evaluations of aid projects have proliferated in recent years,[1] not enough IDA projects are subject to rigorous evaluation to confidently attribute changes in development indicators to IDA support. This is often because there are ethical or feasibility constraints to the allocation of aid to recipients at random and it is often costly to deliver quasi-experimental alternatives. Across the multilateral aid landscape, relatively few ‘programmatic’ interventions are readily assessed by impact evaluations. Moreover, while high-level indicators may improve, disaggregated outcomes (that is, for particular groups) may stay the same or worsen (Coyle, 2014; Fioramonti, 2014; Stiglitz et al., 2010; Fleurbaey & Blanchet, 2013; Kharas, 2011) Niger’s 2018 Completion and learning review finds that, while poverty rates in Niger declined from 54% to 44% during the World Bank’s 2005-14 country strategy, poverty only declined in the capital city and other urban centres, remaining unchanged in rural areas (IEG, 2018d).

Development outcomes targeted by IDA

Tier 2 of the World Bank RMS measures changes in a range of development outcomes targeted by IDA-supported projects. Most tier 2 indicators are reach indicators (for example, the number of students reached by IDA-supported education operations or the number of people provided with a new or improved electricity service). Since the links between the different forms of IDA support and these types of tier 2 development outcomes are more plausible than the higher-level development impacts reported at tier 1, it is more often possible for researchers studying IDA’s effectiveness to attribute changes in tier 2 of the RMS to IDA’s activities.

However, it is difficult to determine whether changes reported by the RMS at tier 2 represent a good rate of return on IDA investment. Even if projects hit their tier 2 targets, it is difficult to establish their development effectiveness without a basis for knowing whether the targets were sufficiently ambitious in relation to outcomes. Moreover, the RMS captures quantifiable and aggregable results only if they are common across many projects, which leaves out quantifiable support that is unique to one or very few projects (such as the IDA-supported resilience programme following the 2015 Nepal earthquake). Although tier 2 results are reported for individual projects, it would be a significant undertaking to reconcile individual project-level results reported across the extensive array of themes and country contexts that IDA operates in. IDA’s important work in advocacy, analytics and policy dialogue is also left out of tier 2 because it is difficult to quantify the outcomes of these activities in a meaningful way.

Organisational and operational performance reviews

IDA’s RMS also tracks changes in indicators of organisational performance (tier 3), grouped into five categories: development outcomes and effectiveness, performance and quality, operational efficiency and responsiveness, financial sustainability and budget efficiency, and implementation of special themes. Under the development outcomes and effectiveness category, the RMS tracks IEG’s quality ratings for the outcomes of IDA country programmes and supported projects (operations) and collects data on perceptions of IDA’s effectiveness and impact on results from clients and beneficiaries through an annual survey. One way to measure IDA’s effectiveness is therefore to compare actual changes in tier 3 organisational performance indicators versus targets.

Multilateral performance reviews have been undertaken by some individual donors from the OECD’s Development Assistance Committee since 2010. They assess IDA’s effectiveness by reviewing RMS results reported across all tiers in conjunction with other qualitative evidence of organisational performance. The assessment methodology for the 2016 Multilateral aid review conducted by the UK’s Department for International Development (DFID) consisted of 16 criteria/assessment questions grouped into six components, three of which look at alignment with UK objectives and three at organisational strengths (DFID, 2016).

The Multilateral Organisation Performance Assessment Network (MOPAN) is a collective initiative of 19 donors whose secretariat is hosted by the OECD. It has been undertaking periodic assessments of multilateral aid providers since 2003. Its current methodology groups indicators into five dimensions of organisational effectiveness: strategic management, operational management, relationship management, performance management and results (MOPAN, 2016).

UK multilateral aid reviews and MOPAN assessments look at the World Bank Group as a whole.[2] Another comparative review, the Quality of Official Development Assistance (QuODA), specifically looks at IDA (CGD, 2021). QuODA was developed by the Centre for Global Development (CGD) and the Brookings Institution and consists of a series of indicators grouped into four dimensions: prioritisation, ownership, transparency and untying, and evaluation (CGD, 2021). The QuODA methodology is based on common principles for aid effectiveness agreed as part of landmark agreements such as the Paris Declaration on Aid Effectiveness (2005), the Accra Agenda for Action (2008) and the Busan Partnership Agreement (2011). The Busan Partnership, for example, stipulates that for aid to be effective: recipient countries should ‘own’ aspects of the development process; results should form the basis of resource allocation; all stakeholders should participate in and share responsibility for aid delivery; and giving should be transparent (OECD, 2011). Since the World Bank’s RMS and the results measurement systems of other international financial institutions (IFIs) and bilateral donors typically align with these sets of principles, QuODA is able to draw comparisons and rank the organisational performance of both IFIs and bilateral donors.

Despite the precision of QuODA’s methodology, it is inherently difficult to score and compare the organisational effectiveness of organisations with different mandates, geographic focus areas, financing and deployment models (MOPAN, 2020). For example, what is considered effective for an agency focused on providing support to refugees and displaced people in fragile countries may not be as effective for an organisation focused on leveraging private sector investments into middle-income countries. Two key changes in the population of IDA-eligible countries are believed to influence IDA’s relative effectiveness in delivering development impacts versus other providers. First, the number of IDA-eligible countries has decreased due to the ‘graduation’ of borrowing countries (for example, India in FY2014 and Vietnam FY2017) whose income per capita and creditworthiness have come to exceed the specified thresholds. At the same time, IDA resources have grown and therefore commitments per borrowing country have risen relatively sharply. Second, a higher proportion of the countries that remain eligible for IDA support are located in Africa and are classed as affected by fragility and conflict. The share of IDA disbursements directed to countries in Africa increased from 53% in 2015 to 64% in 2019 (World Bank, 2021c).

Comparative performance reviews also rely heavily on the internal approaches of the organisations they study. They do not review individual IDA-supported development projects and country programmes, but instead assess the overall performance of IDA based on the aggregation of results from individual projects and corporate performance data. The MOPAN (2016, p. 29) review points out that “the Bank’s approach to results and performance management does not allow for a detailed assessment of development effectiveness in terms of relevance, efficiency and sustainability… [and] the MOPAN framework depends on aggregate reporting, which means that a great deal of faith must be placed in the systems and processes that underpin IEG’s approach to results and performance reporting”. Organisational performance reviews are therefore useful for highlighting IDA’s relative strengths and weaknesses (because they measure attributes that tend to be associated with higher development impact) but, despite recent methodological developments, they do not yet provide rigorous evidence of development impact.

Independent analysis

IEG evaluations examine the work of the World Bank Group, often by reviewing the World Bank’s support to a particular development objective (Thematic Evaluations) or country (Country Programme Evaluations). Evaluation approaches vary, but include assessing outcomes against initial targets and expectations, or in some cases assessing what might have happened in the absence of World Bank or IDA support. This body of evidence is useful for generating qualitative insights around what works well and less well in particular contexts, and why. Independent research conducted by academics and civil society groups studying IDA’s effectiveness typically analyses, and draws conclusions from, the existing quantitative evidence of effectiveness reported publicly via IDA’s RMS and qualitative evidence of organisational and operational performance, primarily from IEG evaluations and completion and learning reviews sampled and audited by IEG. Primary research conducted on IDA’s effectiveness from sources outside the World Bank is mostly limited to IEG evaluations, meaning that it is difficult to comprehensively assess the effectiveness of IDA through independent sources. The World Bank and, by extension, IDA are heavily scrutinised by academics, activists and civil society groups. However, due to the lack of access to primary data, criticisms are often based more on secondary analysis, logic and opinion, than on rigorous primary evidence.

3.1 Evidence of IDA’s effectiveness

Development impact

The RMS finds evidence of IDA18 and IDA19 (before June 2021) achieving intended results at the development impact (tier 1) and development project outcome (tier 2) levels. The IDA18 final implementation and results report shows that the majority of expected tier 1 impacts were achieved (for example, an average annual GDP growth rate per capita in IDA countries of greater than 1%), as well as expected tier 2, IDA-related outcomes (for example, 5 million farmers adopting improved agricultural technology) (World Bank, 2020c). The most recent IDA19 results update report notes that tier 1 outcomes are expected to be negatively impacted by the COVID-19 crisis, but progress was still achieved in most key development indicators under tier 2 (World Bank, 2021c). However, while these may be indicative of effectiveness, we cannot confidently attribute aggregate changes in tier 1 and even tier 2 RMS-reported indicators solely to IDA. Therefore, evidence regarding IDA’s macroeconomic impact and development outcomes is necessarily inconclusive, even where highly plausible.

Operational effectiveness

There is some, though weak, evidence of IDA achieving intended results at the operational effectiveness level (tier 3). Under the tier 3 development outcomes and effectiveness category, the RMS tracks the percentage of IDA country programmes and supported projects (operations) which receive IEG outcome ratings of ‘satisfactory’ or above (‘moderately satisfactory’ and ‘highly satisfactory’).

IDA project outcome ratings trended upwards during the review period, with the steepest year-on-year increase occurring in FY2020, despite the COVID-19 pandemic. According to IEG’s Results and performance of the World Bank Group 2021 report (IEG, 2021b), around three-quarters (76%) of projects closed in FY2016 were rated by IEG as ‘satisfactory’ (41%) or ‘moderately satisfactory’ and above (35%), whereas a higher proportion (84%) of projects closed in FY2021 achieved ‘satisfactory’ (30%) or ‘moderately satisfactory’ or above ratings (54%). IEG also finds that “[R]atings increased for projects in all Practice Groups…improved in the most challenging places to operate, rising for projects in fragile and conflict-affected situations and increasing in IDA countries…increased notably for the World Bank’s largest projects – those valued at over $100 million.” (IEG, 2021b, p. X)

Under IDA19, a slightly higher share of commitments achieved at least satisfactory outcomes ratings compared to the share of projects, suggesting that larger projects tend to perform above average (the World Bank, 2021c).

Aggregate performance ratings for Country Partnership Frameworks (CPFs) are reported separately to operations. CPF ratings are consistently lower than project ratings, with the mid-term review of IDA19 reporting that, despite an improving trend, only around half (52%) of CPFs were rated by IEG as ‘satisfactory’ or better at exit (World Bank, 2021c).

However, IEG’s Results and performance of the World Bank Group 2021 report notes that outcome ratings are an imperfect indicator of whether or not intended outcomes are achieved. In reaching outcome assessments, IEG looks primarily at the results reported by projects against their own indicator frameworks, which vary substantially in terms of rigour. Although some projects apply more direct and robust measurement approaches, many select low-quality, output-based indicators or rely on weak, indirect or anecdotal evidence of results. The report identifies that despite higher project outcome ratings, only four[3] out of 16 possible outcome areas achieved higher ratings in FY2020. It considered these to be the outcome areas for which the World Bank had simply become better at measuring. A robust increase in monitoring and evaluation quality was found to have the strongest positive correlation with the recent increase in project-level outcome ratings (IEG, 2021b).

Organisational effectiveness

There is strong evidence of IDA’s relative overall effectiveness in terms of organisational performance indicators. In 2016, the DFID multilateral review rated the World Bank as ‘very good’ in terms of both organisational strengths and alignment with UK objectives (DFID, 2016). Also in 2016, the MOPAN assessment of the World Bank rated its performance in all five dimensions as either ‘highly satisfactory’ or ‘satisfactory’ (DFID, 2016). Moreover, IDA specifically ranks third overall in the most recent edition of QuODA, and achieved strong scores across all four dimensions of aid effectiveness. This consistent performance across all four dimensions is notable considering that some QuODA indicators are juxtaposed (CGD, 2021). A long-standing challenge for comparative performance reviews is the tension between needs and absorption capacity. Wood (2018) questions the 2018 Commitment for Development Index’s ranking of New Zealand as the world’s most effective donor, arguing that a higher share spent by development financial institutions in poorer countries is likely to increase prioritisation scores, whereas a higher share spent in well-governed, typically richer countries where local authorities could absorb it more easily is likely to increase ownership scores (Wood, 2018). This indicates that IDA’s strategies and operations are balanced well to achieve potentially conflicting development objectives.

Reach and depth of expertise

There is substantial evidence that IDA has strong global reach and deep in-house expertise which improves its overall effectiveness in delivering development impact. The DFID (2016) and MOPAN (2016) reviews find that the scale and breadth of the technical expertise of World Bank staff, as well as the range of countries and sectors the Bank operates in, set it apart from other IFIs and MDBs (DFID, 2016; MOPAN, 2016). The DFID Multilateral development review also stated that “As the largest development actor, the World Bank plays a leadership role which is central to achieving the UK’s ODA objectives and pivotal for the SDGs… Its strengths include its presence across almost all low- and middle-income countries, strong technical capacity enabling it to perform a leadership role in many areas, a wide range of financial instruments and generally good delivery against challenging development objectives” (DFID Multilateral Development Review 2016, p. 2).

Knowledge shaping

There is substantial evidence that IDA has strong analytical capabilities. The DFID (2016) multilateral assessment identifies the World Bank’s comparative advantage over bilateral aid as its ability to produce knowledge and shape the ways in which the development community thinks about issues. However, the MOPAN (2016) review identifies that, while the World Bank’s analytical capabilities are one of its primary strengths, it has a long-standing challenge regarding how it uses the knowledge it generates in practice. IEG’s evaluation of the World Bank Group’s self-evaluation systems finds that staff on the ground in some cases rarely use and struggle to find practical value in the many of the Bank’s analytical and diagnostic tools designed to support country analysis, results measurement, risk management and control throughout its operations (IEG, 2016a). IEG’s Results and performance (RAP) of the World Bank Group (2020) confirms that “for these and other reasons, there are trade-offs between outcome orientation and using results measurement systems for reporting and incentivising fulfilment of policy commitments” (IEG, 2020d).

Allocation of resources

There is substantial evidence that, compared to other IFIs and MDBs, IDA allocates resources well to countries with greatest levels of need. IDA scored higher than all bilateral providers (and third of all multilateral providers) on the prioritisation dimension of QuODA 2021, which measures how well allocations are targeted to respond to long-term development challenges. Palagashvili & Williamson (2014) find that bilaterals tend to direct aid towards better-governed countries, and therefore IDA’s greater emphasis on targeting countries with the highest levels of need reflects positively on its contribution within the overall aid ecosystem. Moreover, IDA is the largest source of (mostly) unearmarked multilateral finance for poor countries. Other IFIs provide more finance which is earmarked for specific purposes and many bilateral providers also tie aid to procurement conditions (that is, funding can only be used to procure goods and services from the providing country). Strong evidence from the wider literature suggests that both forms of earmarking have additional costs (OECD, 2019b).

However, emerging evidence suggests that allocating resources on a country-by-country basis, particularly through unearmarked loans to sovereign governments (Development Policy Financing, DPF), tends to lead to underinvestment in global public goods. Outside voices call for IDA to move away from its core, country-based allocation model and address the underprovision of so-called development-relevant global public goods, such as renewable energy technology and vaccines, and promote forms of international cooperation, such as managing externalities and strengthening transnational taxation systems (Gwin, 2002; Kaul, 2013; Birdsall & Morris, 2016; Kaul, 2017). Authors such as Birsdall & Diofasi (2016) argue that IDA’s country-based deployment model is less effective in addressing increasingly relevant cross-border challenges (such as global climate change, water scarcity, commodity-price volatility, pandemic diseases, cybercrime, corporate tax avoidance, and microbial resistance to antibiotics). The exposure of widespread underinvestment in public health services and social protection programmes in the wake of the COVID-19 crisis added yet more weight to calls for expanded provision of development-relevant global public goods. Moreover, as new sources of development finance for poor-country governments have proliferated, the risk of multilateral financing duplicating or supplanting domestic financing has increased. Civil society organisations such as Oxfam (2021), Save the Children (2021) and the International Trade Union Confederation (2021) support calls for the World Bank to leverage IDA resources to create a global fund for public services.

IDA’s proponents argue that a balance must be struck; the allocation of further concessional IDA resources into earmarks for the provision of particular development-relevant global public goods would require the introduction of more funding ‘windows’ and would lead to trade-off consequences for poor countries which rely heavily on core IDA finance.

Debt sustainability

Some evidence suggests that some of IDA’s sovereign lending is not sustainable. Over a third of IDA-recipient countries are rated as being at high risk of or in debt distress, more than twice the number in 2015 (IEG, 2021a). These trends predate the COVID-19 pandemic, but the ongoing impacts of the COVID-19 crisis have almost certainly led to further increases in debt vulnerabilities. The Bretton Woods Project, a leading critic of the World Bank and other Bretton Woods Institutions, estimated that, between 2008 and 2018, the median level of publicly owed or guaranteed foreign debt for sub-Saharan African countries increased from 38% to 56% of GDP (Bond & Brown, 2020). As debt levels increase, IDA-country governments that are often already in chronic fiscal and balance of payments deficits tend to direct funding away from key public services (for example, education, health and infrastructure) to keep up with debt service payments (Bond & Brown, 2020). However, IDA lending is only part of the reason for this debt increase, since many IDA countries also increased borrowing from private or government sources (most notably, China) during this period. Moreover, IDA financing is typically made available on much softer terms than alternatives (for example, average Chinese government loans or loans from commercial lenders). For countries assessed as already facing serious debt distress, IDA provides funding on a 100% grant basis in any case.

In the past, the World Bank has provided large-scale debt relief to countries.[4] However, more recently it has emphasised supporting and incentivising IDA countries to better manage their debt. In 2020, the World Bank and the International Monetary Fund introduced the Debt Service Suspension Initiative (DSSI), which enabled IDA countries in or at risk of debt distress to request a postponement of their debt payments until the end of 2020. The World Bank also implements a range of initiatives designed to enhance debt transparency, strengthen debt management capacity and improve the analysis of debt management policies, which together form part of its Sustainable Development Financing Policy (IDA, 2020). However, much of the recent literature as well as popular criticism of IDA from the Global South claims that these measures do not go far enough (Atingi, 2019; Bretton Woods Project, 2019; Were & Mollel, 2020). One of the key weaknesses identified in IDA’s debt sustainability measures is the focus on building the capacity of countries to manage future debts at the expense of working with creditors to alleviate existing debt burdens. Moreover, the DSSI merely suspends rather than cancels the debts, and does not apply to the World Bank’s own debt service. Watts (2020) estimates that IDA-eligible countries paid $45 billion in total loan repayments to the World Bank in 2020, despite the DSSI.

Conditionality and its critics

IDA’s lending conditions are controversial and evidence regarding their effectiveness in promoting improved economic performance in recipient countries is inconclusive. During the 1980s, Washington Consensus-based policy adjustments were widely considered to promote effective economic management and sound policies (that is, deregulation, liberalisation, austerity and privatisation) that were perceived to carry the potential for sustained economic development and, by extension, poverty alleviation in poor countries. From the 1990s, however, the Washington Consensus model began to be viewed not only as obsolete but correlated in certain contexts with uneven growth, boom-and-bust cycles, excessive corporate welfare and industrial enclave-style development. Many authors argued that the introduction of Structural Adjustment Programmes in the 1980s contributed to balance of payments deficits and underinvestment in public services (Stein, 1992; Geo-JaJa & Mangum, 2001). In Bolivia, for example, structural adjustment policies imposed as part of World Bank lending conditions from 1985 led to rising unemployment, reductions in public revenues and countrywide riots over water privatisation (Kohl, 2002). South Korean economist Ha-Joon Chang’s influential book Kicking away the ladder argued that economic growth and prosperity are more closely associated with the kinds of proactive state interventions that Washington Consensus reforms eschew (Chang, 2003). Chang’s work follows on from a book published in 2002 by the former World Bank Chief Economist Joseph Stiglitz, entitled Globalisation and its discontents, which accuses the Washington-based IFIs of imposing policy prescriptions that oppose those that are considered to have delivered prosperity historically in many Western market economies (Stiglitz, 2002).

The World Bank claims that it has adapted over time in response to the evidence of Washington Consensus policies failing to improve socioeconomic conditions. Officials contend that IDA has evolved to provide financing according to borrower needs. ‘Country ownership’ and ‘context-specific lending’ have been official goals of the World Bank for over two decades and it now uses the terminology of ‘prior actions’ rather than ‘conditions’. However, some question whether the World Bank has truly abandoned the ‘one-size-fits-all’ Washington Consensus for more progressive, context-specific conditionality. In 2020, the United Nations Conference on Trade and Development (UNCTAD) noted that the World Bank’s prescribed policy mix of austerity, inflation targeting, trade and investment liberalisation, innovative finance and labour market flexibility “led to a world of growing economic inequalities, arrested development, financial fragility, and unsustainable use of natural resources before the pandemic hit” (UNCTAD, 2021, p. 17). Landers & Aboneaaj (2021) find evidence of a recent uplift in conditionality under new World Bank President David Malpass and an IEG study into the impact of DPF in IDA countries finds that conditions attached to DPF constrain recipient countries’ responses to the COVID-19 crisis. On the other hand, some recent evidence suggests that conditionality may deliver on improved economic performance in the long term. Archibong et al. (2021, p. 1) find that “following initial declines in per capita economic growth over the 1980s and 1990s, the countries that adopted the reforms experienced notable increases in per capita real GDP growth in the post-2000 period”.

Bureaucracy and speed

Some evidence suggests that World Bank bureaucracy limits IDA’s effectiveness in responding quickly and flexibly to the needs of poor countries. The MOPAN (2016) review finds that the World Bank’s rigorous control environment has trade-off consequences for speed and flexibility of implementation and considers the extensive array of units, processes and tools to support risk management, corporate reporting and donor accountability heavy and bureaucratic. Birdsall & Morris (2016) find that bureaucracy is a common challenge for older MDBs which, at least before the COVID-19 crisis, were not considered effective crisis response tools (an assertion based largely on the sluggish MDB response to the 2008 global financial crisis). Critics of MDBs argue that they place an administrative burden on developing-country governments, rather than responding flexibly to their needs (US Congressional Research Service, 2020). However, MDB operations are subject to intense scrutiny by activists, academics and civil society groups, and the donors upon which IDA relies for regular replenishments are highly politically sensitive to stories of corruption, waste, human rights abuses, and environmental injustices (US Congressional Research Service, 2020). Given these circumstances, it is unsurprising that IDA is nestled within a large, complex and heavily rules-constrained bureaucracy, at the behest of its donors. While this is considered by critics to limit its effectiveness in delivering development impact, proponents argue that it necessary to reduce risks of operational corruption, injustice and exploitation.

Innovation and adaptation

There is emerging evidence of trade-offs between the World Bank’s rigorous control environment and innovation (Honig, 2018; Honig, 2020). Critics argue that, at the operational level, the World Bank is more focused on disbursements than on delivering results; staff emphasise short-term ‘box-ticking’ outputs (for example, reports and frameworks) and engage less in innovative activities that carry the potential to deliver results in the long term (US Congressional Research Service, 2020). Indeed, IEG’s evaluation of the World Bank’s self-evaluation system finds that demands for extensive corporate monitoring shift staff focus away from project implementation and may even stifle motivation to pursue positive development impact (IEG, 2016a). IEG’s Results and performance (RAP) of the World Bank Group 2020 report concludes that, organisationally, it must improve on how its systems are set up to support a more practical culture that moves beyond meeting individual project targets and demands for corporate monitoring (IEG, 2020d).

Evaluation and learning

Some evidence suggests that IDA’s evaluation and learning has improved from a position of relative weakness. The MOPAN (2016) and DFID (2016) reviews identified a need for the World Bank to strengthen its monitoring and evaluation frameworks at the project level and that its learning and feedback loops could be improved. In response, the Bank committed to multiple improvements to IDA’s RMS, many in line with the recommendations of donors including the UK. As a result, IDA’s RMS is now considered to be one of the strongest and most comprehensive multilateral monitoring and evaluation systems. QuODA 2021 – which used the MOPAN assessment to assess the World Bank’s approach to evaluation – ranked IDA 12th-highest of 49 multilateral and bilateral aid providers on this dimension, and emphasised that IDA performs well on its evaluation and learning systems, but below average for its results-based management.

[1] The Development Impact Evaluation programme and Strategic Impact Evaluation Fund are key sources of funding for impact evaluations of World Bank projects.

[2] Forthcoming MOPAN assessments for publication in 2021 and 2022 will cover IDA/IBRD and IFC separately.

[3] Expanded access to services, increased human capital, improved enterprise and sector performance, and enhanced equity and inclusion.

[4] Debt relief has been provided under the Heavily Indebted Poor Countries Initiative and the Multilateral Debt Relief Initiative.

4. Effectiveness of IDA in selected focus themes

In this section, we explain the policies, practices, actions and reforms enacted by IDA in relation to five focus themes relevant to our review questions: fragility and conflict, climate change (focused on resilience and adaptation), COVID-19, equity and inclusion, and environmental and social safeguards. We outline the available evidence from the World Bank Results Measurement System (RMS), multilateral reviews of IDA, Independent Evaluation Group (IEG) evaluations and the wider literature on the applications and effectiveness of IDA policies within each theme.

4.1        Fragility and conflict

The World Bank’s poverty analysis tool showed in 2018 that poverty rates are around 10% higher in IDA countries categorised as fragile and conflict situations (FCS) than in IDA non-FCS countries, and also that extreme poverty is stagnant in IDA FCS countries (World Bank, 2019, p. 1). The World Bank therefore recognises that achieving its twin goals of reducing extreme poverty and promoting shared prosperity will require tackling both the drivers and the impacts of fragility, conflict and violence, and channels much of this effort via IDA.

Policies in fragility, conflict and violence

Evidence suggests that IDA’s volume of support to fragile states has historically been lower than other multilaterals, although during IDA18 and IDA19 this has started to change. IEG’s 2014 synthesis report Learning from IDA experience noted that one-sixth of IDA resources in 2011 were directed towards FCSs compared to 40%, on average, for other multilateral development banks (MDBs) (IEG, 2019c). The Department for International Development’s (DFID) 2018 annual review of IDA18 advised the UK delegation to encourage IDA management to increase the number of commitments and expertise of staff deployed to FCV countries (DFID, 2018).

Fragility and conflict became a special theme under IDA16 and, in 2020, the World Bank developed a dedicated strategy for fragility, conflict and violence (FCV) (World Bank, 2020d). This strategy sets out a differentiated approach to operations in FCV countries based on identified key underlying drivers of conflict and fragility (such as climate change, demographic shocks, gender inequality, patterns of discrimination, economic and social exclusion, and perceptions of grievances and injustice). In more recent replenishment rounds, the World Bank has been challenged to scale up its efforts to support countries in situations of FCV.

IDA support to FCV situations has gradually increased in recent replenishment rounds. Total commitments to FCV countries (and four countries identified through the Risk Mitigation Regime to be at risk of becoming FCV) were $9.6 billion under IDA18, over double that of IDA17 (World Bank, 2019). $1.5 billion of this was financed through the Regional Window Programme (RWP), the Refugee Sub-Window (RSW) and the Crisis Response Window (CRW). Approximately 40% of RWP resources, for example, were allocated to FCV countries under IDA18 (IEG, 2019b). During the first 14 months of IDA19, FCS countries accounted for 39% of IDA commitments, “up to $14.5 billion in the first 14 months of IDA19, compared to $5.8 billion during the corresponding period for IDA18” (World Bank, 2021c, p. ii).

As the flow of financial resources to IDA FCV countries has increased, the World Bank has also increased the number of staff deployed to FCV countries (FCDO, 2018; FCDO, 2019; World Bank, 2019). Against an original target of 150, IDA18 saw a net increase of 123 staff working in FCV countries, which was achieved through strengthening the package of benefits available for experienced staff to relocate to FCV (World Bank, 2019). Under IDA19, the World Bank aimed to deploy a further 150 staff members to FCV, but did not expect to achieve this target and revised the target to 100 in 2021 (World Bank, 2021c). The Bank has also enacted reforms to strengthen its approach to engagement in fragile situations, for example through introducing Risk and Resilience Assessments (RRAs), which inform IDA country strategies and development plans under IDA19 (World Bank, 2020d).

Effectiveness of delivery in fragility, conflict and violence

Despite reforms and more overall staff and financial resources being directed to FCV countries, strong evidence suggests that operational effectiveness in FCV continues to lag behind the institutional average (MOPAN, 2016; DFID, 2018; DFID, 2019; IEG, 2020c; World Bank, 2017; World Bank, 2021). The IDA18 RMS shows that the share of project outcomes rated ‘moderately satisfactory’ or better by IEG in IDA FCV countries declined between FY2015 and FY2017, although it notes that part of this decline was driven by the performance of portfolios in Afghanistan and, to a lesser extent, Yemen (World Bank, 2017). Furthermore, while there was strong demand for the RWP under IDA18, the RSW, CRW and Private Sector Window (PSW) did not fully exhaust their allocated resources. On the other hand, comparatively weaker performance in FCV is unsurprising considering the higher risks and limited absorption capacity of recipient governments in FCV countries. Other international financial institutions (IFIs) also struggle to perform effectively in FCV settings due to the complex political economy challenges associated with working in conflict zones. It is therefore important to consider the relevance of IDA’s deployment model and strategic approach to the needs of FCV countries.

There is emerging evidence that IDA’s model is ill-suited to tackling the drivers and impacts of FCV. IEG evaluations of IDA operations in FCV countries find that IDA support during times of crisis overtaxes country capacity and should be simplified to be more effective in these settings. The 2020 IEG review of the Sierra Leone completion and learning review finds that, “While administrative capacity was developed – for example, for local councils to have integrated development plans and budgets – capacity still remained a significant constraint for programme implementation. The results framework was too complex, and the programme was not selective considering the limited IDA envelope and the constraints on domestic absorptive and implementation capacity” (IEG, 2020e, p. 3).

Furthermore, while the World Bank’s analytical work on FCV issues is strong, some evidence suggests that it has not sufficiently translated into IDA country strategies and frameworks. IEG’s World Bank Group engagement in situations of fragility, conflict and violence review finds that IDA’s country strategies to assist FCV countries often lack realism and tailoring to the unique challenges present in FCV settings (IEG, 2016c). Concerns around the limited practical use of FCV analysis have been mirrored at the operational level. IEG reviews have found that IDA operations rarely adequately address the issues that, according to the World Bank’s FCV strategy, can lead to an escalation of conflict and violence. For example, the World Bank recognises that social discontent arising from the lack of inclusiveness of the growth process can lead to conflict and violence, but it lacks a realistic, coordinated framework for supporting inclusive growth and creating jobs that is based on addressing private sector constraints in FCV (IEG, 2016b).

Although the World Bank has enhanced its analytical capacity to engage in situations of FCV since the introduction of RRAs, early analysis of their use suggested that country strategies do not always cover the fragility drivers which the RRAs identified (IEG, 2020c). A 2020 IEG stocktaking of the World Bank’s experience with RRAs and operational programming in FCV countries finds that Country Partnership Frameworks (CPFs) for FCV countries do not differ significantly from CPFs for non-FCV countries (IEG, 2020c). In addition, the World Bank RMS lacks fragility indicators, and project-level outcome indicators are rarely disaggregated by geographic region or social groups (IEG, 2018).

However, IEG’s 2021 World Bank engagement in situations of conflict review (IEG, 2021f) suggests that RRAs are starting to have a more significant impact in promoting the conflict sensitivity and responsiveness of World Bank operations. It finds that the existence of an RRA encourages ‘leaning in to’ a conflict “because it makes the design of subsequent investment projects more likely to integrate conflict sensitivity” (IEG, 2021f, p. 28-29).

4.2        Climate change

Climate change policies

The World Bank recognises that IDA-eligible countries are among the most vulnerable to climate impacts. Under IDA19, the Bank developed a Climate Change Action Plan (CCAP) for 2021-25, which introduces Climate and Development Reports (CCDRs) to inform and support IDA countries to transform key sectors in alignment with climate goals (World Bank, 2021f). Climate was identified as a special theme for the first time under IDA17, and the World Bank updated the RMS to include a number of climate change and resilience indicators at the tier 1 impact level (World Bank, 2014). The Bank also began to monitor a range of tier 2 project outcomes and outputs related to climate, such as lifetime energy savings and renewable energy generation (World Bank, 2014). Additionally, the Bank began to screen all IDA operations for potential climate risks under IDA17, recognising the potential for projects across the breadth of its portfolio to produce climate costs as well as benefits (World Bank, 2014).

Climate finance investments typically entail community-based green infrastructure development and reinforcement, such as financing for wastewater infrastructure ecosystem restoration and sustainable management of forests and marine resources, and decentralised energy solutions, such as providing access to affordable, reliable and sustainable energy or accelerating renewable energy efficiency improvements (World Bank, 2021c). IDA has also supported multi-country climate initiatives through the RWP, for example by providing investment to the Great Green Wall Initiative through the Sahel and West Africa programme, which aims to improve environmental conditions, increase economic opportunity, and reduce the vulnerability of poor rural populations to climate change impacts (IEG, 2019b).

Following the mainstreaming of climate change into IDA country and sector strategies under IDA17, IDA provided increased levels of finance to support both climate adaptation and mitigation projects, with a greater emphasis on adaptation. Between 2016 and 2020, an average of 26% of total IDA disbursements were allocated to ‘climate finance’ (covering both climate adaptation and resilience). The total of $21.4 billion delivered by the World Bank for climate finance in 2020 makes it the largest single provider of climate finance for low-income countries (World Bank, 2021e). Support for climate adaptation also increased between 2016 and 2020, from $6 billion to $9 billion in total funding. However, critics often note that the share of overall World Bank investment for climate-related projects is lowest (and increasingly so) in IDA countries (IEG, 2019c).

Moreover, the World Bank’s CCAP has been criticised for lacking detail regarding how CDDRs will influence investment decisions and how it defines whether an investment is classified as ‘climate finance’. A recent investigation into donor reporting of climate adaptation finance suggests that the World Bank’s climate adaptation finance figures are over-reported (Hattie, 2020). The report concluded that a $500 million Earthquake Housing Reconstruction Project in Nepal over-reported the amount of finance provided to adaptation by $328 million, and a $600 million Rural Productive Safety Net loan in Ethiopia in 2017 over-reported by $106 million. Some voices outside the World Bank would therefore like to see greater transparency in the RMS regarding climate finance (Bretton Woods Project, 2021b).

For IDA19, the World Bank further scaled up ambitions on climate change in the form of increasing the target proportion of projects with climate co-benefits from 30% to 35% and agreeing to allocate no less than 50% of climate finance to adaptation (Bretton Woods Project, 2021b). It also aims to complete “up to” 25 CCDRs in the year following the publication of the CCAP on 22 June 2021 (World Bank, 2021f, p. 12). The mainstreaming of climate into IDA’s RMS helped the World Bank to better capture where and to what extent IDA operations generated climate co-benefits (MOPAN, 2016). IEG’s IDA synthesis report finds that, between FY2008 and FY2010, and between FY2015 and FY2017, the share of project components in the IDA portfolio with potential climate change benefits increased by 50% (IEG, 2019c). However, DFID noted, in its 2019 annual review of IDA, that while mainstreaming climate into IDA operations had achieved positive impact, more climate finance and more ambitious targets for projects with climate co-benefits were needed to address the scale of climate vulnerabilities facing IDA-eligible countries (DFID, 2019).

Effectiveness of delivery on climate objectives

Some evidence suggests that IDA projects are less successful in achieving intended climate objectives compared to non-climate projects. The World Bank revised several of its expected climate targets for IDA19, as it did not expect them to be achievable within the shortened IDA19 period (World Bank, 2021c). The expected target for the proportion of IDA-supported projects with climate-related results was revised from 100% to 90%, while the target for renewable generation was revised from 10 gigawatts to 5 gigawatts  (World Bank, 2021c). Other targets (for example, the share of projects with climate co-benefits) were met ahead of time, suggesting that the shortened IDA19 period did not have a disproportionate impact on IDA’s achievement of climate-related objectives.

However, IEG’s learning from the IDA experience synthesis paper (2019c) reported that the results of IDA-supported projects with climate-specific objectives were lower than for other special themes (IEG, 2019c). Projects with climate objectives received ratings of ‘mostly achieved or higher’ for under half (44%) of all objectives, whereas projects with gender and jobs and economic transformation objectives achieved ‘mostly achieved or higher’ ratings for more than half of all objectives (IEG, 2019c). Several recent IEG thematic evaluations examine the relative performance of IDA projects focused on climate outcomes, such as pollution management (IEG, 2017), carbon market creation (IEG, 2018a), renewable energy electrification (IEG, 2020a), and natural resource management (IEG, 2021c). These reviews find that IDA-supported climate-focused projects achieve a lower rate of ‘satisfactory’ or greater IEG development outcome quality ratings compared to non-climate-focused projects supported by IDA.

The World Bank argues that climate target revisions were necessary under IDA19 because of the disruptive impact of COVID-19, but climate change-related IEG thematic evaluations attribute the lower rates of success of IDA climate interventions to issues of service provider capacity, political support and private sector participation in-country. Waste management and wastewater treatment infrastructure interventions “were successful less often, depending on the ability to recover costs and institutional capacity of the municipalities in charge” (IEG, 2017, p. XIII), while projects focused on emissions reductions struggle to secure commitments from the necessary political and private sector actors to achieve emissions reductions in key industries such transport, forestry and agriculture (IEG, 2021c). Substantial evidence therefore suggests that IDA struggles to secure the engagement from political and private sector stakeholders necessary for projects to achieve climate targets such as pollution management or emissions reductions.

There is also the question of whether IDA is striking the right balance between supporting climate adaptation and resilience, vis-à-vis supporting climate mitigation and other environmental challenges. Adaptation has increasingly taken up a larger percentage of IDA’s climate finance, increasing from 40% in 2016 to 52% to 2020, and further to 61% (as of June 2021) (World Bank, 2021c, p. 10), which is in line with the IDA19 agreement that adaptation should be more than 50% of IDA’s climate finance (World Bank, 2020a, p. XIII). This reflects increases in financing for disaster risk management works (for example, flood defences) and climate-resilient infrastructure (for example, transport infrastructure). However, IEG finds that, between FY2007 and FY2017, the World Bank decreased support for other important environmental challenges such as biodiversity conservation and water resource management (IEG, 2017).

Since adaptation can be understood as the process of adjusting to the current and future effects of climate change, whereas mitigation is associated with making the impacts of climate change less severe, outside critics argue that IDA is not doing enough to support mitigation. IEG (2017) finds that only a fraction (2.5%) of World Bank investments in electricity are for off-grid electrification, while research conducted by German environmental lobby group Urgewald (2021) estimates that the World Bank has provided $12 billion in finance to fossil fuels since the Paris Agreement was signed (Urgewald, 2020). Bank officials contend that, because IDA countries are most vulnerable to the impacts of climate change but least responsible for global emissions to date, supporting mitigation over adaptation unfairly risks exacerbating the climate- and non-climate-related challenges faced by IDA countries in the short term. For example, withdrawing investments in non-renewables risks exacerbating energy poverty in IDA countries.

However, emerging evidence undermines the argument that IDA’s continued investments in non-renewables are necessary from an equity perspective. Research undertaken by The Big Shift Global (a multi-stakeholder campaign coordinated by civil society organisations from the Global North and South) estimates that relatively few (13%) of the World Bank’s energy investments support improved electricity access (Mainhardt, 2017). Other MDBs such as the European Investment Bank, and bilateral donors such as the UK and the US, have committed to ending support for fossil fuel investments.

4.1        COVID-19

The COVID-19 pandemic has delivered a protracted economic shock of enormous proportions (Gavas, 2020). Half a billion people are estimated to have entered poverty due to the pandemic (Hoy & Sumner, 2020), 100 million to have entered extreme poverty (World Bank, 2021a), and 255 million full-time jobs are estimated to have been lost worldwide (ILO, 2021). While the International Monetary Fund (IMF) is the largest international crisis lending organisation, COVID-19 had such major impacts on health and education systems that many poor countries turned to IDA for additional support when the crisis affected their short-term financing needs (Duggan et al., 2020). IDA is therefore widely considered to play an important role in preventing a significant reversal of decades of development progress post-COVID.

Financing to support the COVID-19 response

Implementation of the 19th replenishment of IDA was dominated by the response to COVID-19 (World Bank, 2021c). Early on, the World Bank frontloaded and re-directed more than 40% of IDA19 resources to support the COVID-19 crisis response in poor countries. This funding included budget support (Development Policy Financing, or DPF) to protect crisis-hit countries’ fiscal circumstances, as well as operational financing to support a wide array of COVID-19 relief efforts including the supply of personal protective equipment, implementation of vaccination programmes, remote delivery of education, and protection of households and small and medium-sized enterprises suffering income losses. To manage the emerging finance gap for crisis recovery, the World Bank successfully advanced the 20th IDA replenishment by working closely with donors. This was widely perceived as a necessary response given the significantly increased financing needs of poor countries during the early stages of the pandemic. The World Bank also complemented additional lending with measures designed to address IDA countries’ growing debt vulnerabilities. Alongside the IMF, the Bank supports the implementation of the Debt Service Suspension Initiative (DSSI), which provides liquidity relief and debt management support for IDA-eligible countries (World Bank and IMF, 2021).

Speed, scale and effectiveness of COVID-19 response financing

Few studies have been published at the time of writing that assess the effectiveness of the complex array of IDA-supported COVID-19 relief activities. However, a number of studies have looked at IDA’s role as a crisis financier (Duggan et al., 2020; Gavas, 2020; Humphrey, 2020; Landers & Aboneaaj, 2021). Emerging evidence suggests that actual IDA disbursements were released too slowly to enable countries to respond effectively.

Gavas (2020) finds that new IDA loan commitments rose 118% year on year in the first seven months of 2020 and actual disbursements rose by 31%. Duggan et al. (2020) compare the World Bank’s lending response to the COVID-19 crisis to the Global Financial Crisis (GFC) 12 years earlier. They find that average actual monthly increases in World Bank disbursements during the first seven months of 2020 were less than half those observed during the GFC. Furthermore, as a percentage of GDP, World Bank disbursements in 2020 rose by just 0.3%, compared to an estimated 4.1% decline in median low-income country growth (Gavas, 2020).

On the other hand, IDA specifically has expanded its COVID-19 financing significantly faster than the International Bank for Reconstruction and Development (IBRD) and the International Finance Corporation (IFC) (the World Bank’s lending arms for richer countries and the private sector). This is partly because IBRD and IFC have less flexible financing models and are more sensitive to market risk and overexposure issues. The introduction of the Crisis Response Window under IDA15 helped to improve IDA’s ability to respond to crises which, following the 2014 Ebola outbreak in Central Africa, also included pandemics.

However, critics also argue that the impact of IDA’s budget support during the crisis was constrained by significant policy conditionality. Landers & Aboneaaj (2021) find that, throughout the COVID-19 crisis, the average IDA DPF operation (that is, budget support) required the recipient government to enact eight policy reforms to secure funding, of which only one-third were relevant to the COVID-19 crisis. This places the World Bank out of step with the IMF, which deployed a significant portion of its COVID-19 financing with little to no conditionality (Landers & Aboneaaj, 2021). Not only does conditionality impede the speed and flexibility of IDA budget support during crisis times (Landers & Aboneaaj, 2021), but some evidence also suggests that it detrimentally affects the effectiveness of operations. An IEG study into the impact of DPF in IDA countries finds that, in times of crisis, DPF operations that adapt to support the government’s existing response priorities yield better results than operations with conditions that seek to promote longer-term policy goals (IEG, 2018c).

Morris (2020) credits the World Bank’s efforts to alleviate the debt vulnerabilities faced by IDA countries in the form of promoting the DSSI, but contends that it has largely failed to secure sufficient commitments from national and private creditors to relieve the debt distress experienced by a growing number of IDA countries (Morris, 2020). Due to the absence of key stakeholder groups, the DSSI is estimated to have suspended payments on as low as 10% of the total debt service flowing out of IDA countries. IDA-eligible countries, IDA/IBRD blend countries and former IDA graduate countries were estimated to have paid $45 billion in total loan repayments to the World Bank in 2020, in spite of the DSSI (Watts, 2020). Moreover, the World Bank institutionally, although arguing in favour of the DSSI, has not agreed to participate directly out of fear that its credit ratings would be harmed (Humphrey et al, 2021).

In terms of IDA’s operational support to the COVID-19 response, Oxfam (2020) analyses IDA19’s emergency health funding to 71 countries and finds that 89% of projects did not plan to support preventative action, such as removing financial barriers that exclude people from accessing healthcare (Oxfam, 2020). In addition, two-thirds of the projects studied lack any plans to increase the number of healthcare workers (Oxfam 2020). Oxfam calls for IDA20 to do more to institutionalise preparedness into its country programmes, and support countries in financing their public health services, by strengthening their employment offers to attract more healthcare workers and removing financial barriers to healthcare access (Oxfam, 2021).

4.2        Equity and inclusion

In the World Bank Group’s 2006 report on equity and development, equity is defined in terms of two basic principles (World Bank, 2006). The first is equal opportunity, meaning that a person’s economic, social and political outcomes should reflect their efforts and talents, rather than their background or predetermined circumstances such as gender, race, birthplace or family origins. The second principle is the need for interventions to protect the most vulnerable and neediest members of society. This definition of equity included differences in opportunity as well as outcomes, which widened the focus of equity and inequality analyses that had been traditionally focused on economic outcomes such as income distribution and consumption per capita (Bourguignon et al., 2006).

Inclusion is defined by the World Bank as the process of improving the terms on which individuals and groups take part in society. This involves increasing the ability, opportunity and dignity of those who are disadvantaged because of their identity. These disadvantages are often based on social identity, which may be based on characteristics such as gender, age, location, occupation, race, ethnicity, religion, citizenship status, disability, and sexual orientation and gender identity (SOGI), among other factors (Bourguignon et al., 2006).

Equity and inclusion policies

To promote greater inclusion in development aid, countries pledged that ‘no one will be left behind’ under the 2030 Agenda for Sustainable Development. IFIs including the World Bank officially commit to the ‘leave no one behind’ principle, but the ways in which individual IFIs conceptualise and monitor their commitment to the principle varies. The World Bank Group is strategically aligned with the ‘leave no one behind’ commitment through its focus on poverty reduction; the second of the Group’s twin goals requires the acceleration of per capita income growth for the ‘bottom 40%’. IDA has also placed an increased emphasis on human capital in recent replenishments, identifying it as a cross-cutting issue under IDA19 and as a special theme for IDA20. The World Bank claims that human capital-focused lending (that is, financing for health, education, social protection systems and disability inclusion) to IDA countries has increased – from $6.7 billion per year (FY2011 to FY2019 average) to $11.6 billion a year (FY2020) (World Bank, 2021d). Greenhill & Rabinowitz (2017) find that the core social sectors of health, education and social protection are the key to improving the living conditions of the most vulnerable groups. However, part of the increase in IDA lending to these sectors may simply be explained by the frontloading of IDA resources originally allocated for FY2021 to FY2022 in response to the COVID-19 crisis.

Critics of the World Bank contend that its approach to promoting equity and inclusion lacks an explicit focus on addressing the inequalities and discrimination facing marginalised groups identified in the 2030 agenda (for example, women and girls, refugees, indigenous peoples and people with disabilities). Independent analysis suggests that more could be done to direct national and subnational IDA allocations to the places and populations most at risk of being left behind (Lucci et al., 2019). Many critics consider the focus on the ‘bottom 40%’ and on national-level social service provision too broad to truly address the issues facing these groups. Karr et al. (2017), for example, find that only 5% of the World Bank’s active social protection portfolio explicitly included people with disabilities as target beneficiaries.

In response, the World Bank’s reform processes have seen the gradual introduction of policies and strategic initiatives designed to institutionalise a focus on addressing the unique challenges faced by traditionally excluded or marginalised groups including women and girls, youth, LGBTI people, disabled people and, recently, ethnic minorities.

Recognising that gender inequality and exclusion based on gender have adverse impacts on poverty reduction, security and shared prosperity, the World Bank first introduced a gender strategy in 2002 (World Bank, 2002). This strategy supported IDA practitioners to incorporate a gender lens in the design and implementation of country programmes and operations, and established a framework for monitoring gendered impacts within the RMS. In 2015, a new gender strategy was developed (World Bank, 2015b), based on the conceptual framework introduced in the 2012 World development report on gender equality and development (World Bank, 2012), and focuses on addressing gender inequalities in access to jobs and assets, health and education, and enhancing women’s voice and agency.

To implement the strategy, the World Bank established a new methodology and targets for monitoring progress via ‘gender tags’ for World Bank operations, including IDA. The strategy requires that IDA country programmes and operations, supported by extensive analytical work to diagnose gender gaps, work towards objectives that help close identified gender gaps. Gender-focused IDA commitments typically entail initiatives that reduce occupational segregation and improve access to economic opportunities, mitigate the impacts of sexual and gender-based violence (GBV), increase women’s usage of digital tools, and remove barriers to women’s entrepreneurship. This work often involves exploiting links with the other IDA special themes such as Jobs and Economic Transformation (JET), Climate Change, FCV and Governance and Institutions. Gender was identified as a special theme in its own right under IDA18 (World Bank, 2017b) and leading civil society groups support IDA’s continued commitments to invest in gender-related issues such as tackling GBV (Oxfam, 2021). DFID recognised that the focus on gender as a special theme was promising, but highlighted the need for an increase in the number of IDA operations on gender equality (DFID, 2019). Regarding disability, DFID highlighted that disability should be reflected as a cross-cutting issue in four of the five special themes, as opposed to two (DFID, 2019).

Although gender is the cornerstone of the World Bank’s group-based inclusion agenda, it has also introduced similar strategies and policies for promoting a strategic and operational focus on other excluded and marginalised groups. In 2018, the Bank launched its first Disability Inclusion and Accountability Framework, which aimed to integrate disability into its policies, operations and analytical work, as well as to support the implementation of disability-inclusive IDA country programmes and operations (World Bank, 2018a). The Bank made ten commitments to foster disability-inclusive development in areas including education, digital development, data collection, transport, and private sector investments (World Bank, 2018a). Among these are commitments to ensure that, by 2020 and 2025 respectively, all public facilities in disaster risk management projects, and all urban transport projects, for example, are disability-inclusive (IEG, 2019a). DFID’s 2019 annual review of IDA proposed that disability should be reflected as a cross-cutting issue in four of the five special themes under IDA20, as opposed to two under IDA19 (DFID, 2019).

Moreover, the World Bank established a task force focused on SOGI issues and has conducted analytical research into the economic impact of discrimination against the LGBTI community and other groups. Further to this dedicated task force, the World Bank appointed a global adviser on SOGI, responsible for promoting LGBTI inclusion throughout the Bank’s work in 2016. Recognising evidence that ethnic minorities are disproportionately affected by crises such as the COVID-19 pandemic, the Bank recently expanded its inclusion agenda to address the exclusion of ethnic minorities (World Bank, 2019).

Many IDA country programmes and operations aim to create jobs and promote inclusive economic transformation in poor countries. JET was identified as a special theme under IDA18 (World Bank, 2017b). The World Bank argues that this work supports the inclusion agenda because it raises productivity and earnings and improves pathways to better jobs, delivering poverty reduction, female empowerment and social cohesion objectives (World Bank, 2016; World Bank, 2019; World Bank 2021c). However, critics argue that IDA’s JET interventions focus on promoting productivity and job creation at the expense of promoting equitable working conditions and equal opportunity in recruitment in the private sectors of poor countries. The International Trade Union Confederation (ITUC) supports the inclusion of JET as a special theme. However, it points out that IDA replenishments typically focus on increasing productivity rather than promoting equitable international labour markets (ITUC, 2021).

Application of equity and inclusion policies

Emerging evidence suggests that, although the World Bank’s analytical and strategic work in gender is strong, it has achieved limited progress in addressing gender gaps in IDA countries. According to the most recent IDA19 results and implementation update, the implementation of IDA19 gender commitments and interventions is on track, but support for increased use of digital tools to prevent and respond to GBV will need to be revised and updated to mitigate the effect of the truncated IDA19 period (World Bank, 2021c). Bank officials argue that the closing of gender gaps in the world’s poorest countries is inherently challenging, but recent gender-related IEG evaluations point to the need for more effective implementation of the gender strategy at the operational level.

IEG’s mid-term gender strategy review (2021) finds that the gender ‘flag and tag’ approach institutionalises gender into IDA country strategies and frameworks and that IDA management buys into the strategy (IEG, 2021e). However, it also notes that IDA operational staff report that this evidence can be difficult to access, overly technical, and of limited operational relevance to low-income or fragile country contexts. GBV emerges as one specific area where operational teams find it difficult to translate the World Bank’s analytical work on gender into practical operations. Moreover, staff dedicated to working on gender issues are limited by uncertain responsibilities, time pressures and limited career development opportunities because a gender specialism is not a recognised career path within the Bank.

There is mixed evidence from evaluations of non-gender-focused IDA-supported projects regarding the extent to which these effectively consider gender issues. A 2019 IEG evaluation of World Bank support for urban resilience reports that all but one of the Social, Urban, Rural and Resilience General Practice’s urban flood projects included dedicated activities for gender and that three-quarters of the projects used indicators that tracked gender benefits (IEG, 2019a). On the other hand, IEG’s evaluation of a Community Action Programme and a Community-Based Integrated Ecosystem Management Project in Niger (2020) revealed a distinct lack of gender inclusion (IEG, 2020b). Despite efforts to integrate women in decision-making roles, the project lacked cultural sensitivity, resulting in the exclusion of women from the cash-for-work activities. Childcare options were also not offered during project hours and in project locations and this presented a barrier to the inclusion of women.

Mixed evidence regarding the effectiveness of IDA’s implementation of inclusion commitments is mirrored across other ‘left behind’ groups. IEG’s evaluation of World Bank support to urban resilience finds that people with disabilities and the elderly were neglected in the design of IDA projects (IEG, 2019a). Moreover, research conducted by the Uganda Initiative for Social and Economic Rights finds that a $17.3 million IDA-supported Reproductive Health Voucher Project did not reach the poorest mothers, as 68% of beneficiaries were either middle-class or rich (ISER Uganda, 2019). The study attributes this to the fee attached to the voucher system, which proved to be a significant barrier to the poorest women (ISER Uganda, 2019). A World Bank research paper reviewed the extent to which social inclusion was addressed in Systematic Country Diagnostics (SCDs), finding that almost all SCDs that identify excluded groups contain analyses of some of the domains in which these groups are excluded, but very few SCDs address the question of why certain groups are excluded (Das, 2016).

Disaggregation of results data

One of the key limitations of World Bank support for inclusion is the lack of disaggregated data. Lucci et al. (2019) find that the World Bank RMS does not disaggregate data for vulnerable groups sufficiently to monitor the ‘leave no one behind’ commitment, although it credits the Bank’s increased focus on fragility for including plans to improve disaggregation for refugees and internally displaced people. DFID’s 2016 multilateral review of the World Bank corroborates the finding that there are no comprehensive data to assess the extent to which World Bank programmes are inclusive. An independent analysis by Save the Children UK highlights the limitations of the IDA RMS Human Capital Index methodology and related reporting, which uses national averages and neglects disparities between groups within countries (Leigh et al., 2019). Save the Children supports further disaggregation of data under IDA20 to help improve the World Bank’s understanding of the equity and inclusion of IDA country programmes and operations (Watts, 2020).

Citizen engagement

Citizen engagement – the term used by the World Bank and other IFIs to denote the interactions between ‘citizens’ (also referred to as ‘stakeholders’, ‘communities’ or ‘beneficiaries’) and government or private sector actors within the scope of development interventions – is a key element of efforts to ensure that development programmes promote equity and inclusion. Historically the MDBs, and in particular the World Bank and the IMF, have been criticised for either being ‘top-down’ or ‘one-size-fits-all’ in their approaches to development, with poor countries and their citizens structurally excluded from decision-making processes (Bretton Woods Project, 2019) and analysis of World Bank poverty reduction strategies showing remarkable similarities across partner countries (World Development Movement, 2005). IDA has consistently been challenged by activists and civil society groups to engage more with citizens during the design and implementation of projects and to give borrowing governments more voice and influence over country programmes and frameworks.

In 2013, the World Bank made a corporate commitment to achieve 100% mainstreaming of citizen engagement in operations where beneficiaries can be clearly identified. It introduced corporate indicators to track progress and developed a more comprehensive strategic framework for mainstreaming citizen engagement into its operations, which came into effect in 2014.

IEG’s Results and performance of the World Bank Group (2015) finds that, in the years before this commitment, “in the majority of cases, there was no clear evidence that [participatory] consultations had a meaningful effect on the design or implementation of Bank strategies. In some instances, consultations appeared to be more of a ‘box ticking’ exercise” (IEG, 2015).

To address this issue, the 2014 strategy introduced a new country programme model replacing the country assistance strategy. The new model introduced two new instruments for engaging with partner countries. Initially, a Systematic Country Diagnostic (SCD) diagnoses the most critical constraints to poverty reduction in partner countries, prepared in mandatory consultation with the authorities and stakeholders in these countries. Next, a Country Partnership Framework (CPF) articulates focus areas for IDA support, based on the constraints diagnosed in the SCD and aligned with the country’s own development agenda. The CPF serves as the main strategic tool for IDA support of partner countries’ development programmes, as well as other World Bank activities in the country. Lastly, the World Bank conducts performance and learning reviews (PLRs) and completion and learning reviews (CLRs) every two years or at mid-term during a country strategy period. PLRs and CLRs outline key changes to the country context and lessons learned from programme implementation.

As described in official World Bank statements, this new model is “designed to tailor policies and programmes to the needs and priorities of individual countries”. Other meaningful reforms in relation to partner country and citizen engagement include the 2016 Environmental and Social Framework, which introduces an Environmental and Social Standard on stakeholder engagement and information disclosure (see Section 4.5). A more recent IEG evaluation of the Bank’s approach to citizen engagement (2018) finds that the mainstreaming of citizen engagement into the World Bank RMS generated management awareness and buy-in and increased the percentage of projects with citizen engagement mechanisms – a “major achievement” (IEG, 2018b). From a sample of 299 randomly selected World Bank investment projects, the percentage of projects with dedicated citizen engagement mechanisms increased from 67% to 76% between FY2011 and FY2013, and FY2014 and FY2016 (IEG, 2018b).

Emerging evidence suggests that although corporate indicators have helped to strengthen buy-in and increase the number of project-level engagement indicators in results frameworks, their application in the design and implementation of projects and country programmes remains limited (IEG, 2018b). Birdsall & Morris (2016) argue that World Bank governance reforms fail to satisfy borrower demands for more voice and influence over their policies and operations. However, IEG (2018) notes that high-quality stakeholder engagement is resource-intensive and requires dedicated skills, time and resources that often cannot be applied in poor and fragile countries where government and citizen capacity is limited.

Government engagement

Where partner country governments pursue development priorities that respond to the needs of citizens, greater efforts by donor agencies such as IDA to engage governments on national development priorities can help to promote equity and inclusion. The MOPAN (2016) assessment identified the World Bank’s engagement with borrowing or recipient country officials as one of the Bank’s key strengths compared to other IFIs and MDBs. Moreover, it notes that country engagement “has been enhanced further by the recent reform process, which has resulted in the development of a renewed approach to country engagement that promises to be more evidence-based, selective and responsive to client needs” (MOPAN, 2016, p. 38). Surveys of stakeholders in partner countries undertaken under tier 3 of the World Bank RMS consistently suggest that the Bank is highly valued by partner country officials as a long-term development partner.

The most useful source of comparative evidence on IDA’s response to government priorities comes from the ownership dimension of Quality of Official Development Assistance (QuODA) (2021). IDA was ranked eighth of 48 providers on the ownership dimension, below three regional development banks – the Asian Development Bank (first), the African Development Fund (second), and the Islamic Development Bank (seventh) – and two global funds – the International Fund for Agricultural Development (fourth) and The Global Fund to Fight AIDS, Tuberculosis and Malaria (sixth). QuODA (2021) also finds that aid providers operating in fragile states appeared not to be disadvantaged on the ownership dimension, challenging a long-held view that high-quality stakeholder engagement is unrealistic in FCV countries.

4.3        Environmental and social safeguards

According to the World Bank, safeguard policies are designed to prevent and mitigate undue harm to people and their environment in the development process (World Bank, 2016b). IFI safeguard systems have evolved iteratively over recent years and typically seek to ensure that relevant measures, capacities and processes are in place which, when applied, identify, avoid and minimise harms of development projects and investments to people and the environment.

Over the years, World Bank-supported projects, including IDA-supported projects, have attracted negative media attention due to infringements and in some case violations of human rights (Bretton Woods Project, 2019). Some examples include mass evictions and forced displacement, and child and forced labour (Bretton Woods Project, 2019). The World Bank’s safeguard system has evolved iteratively to protect against these risks.

Safeguard system structure

In general, IFI/MDB safeguard systems entail: an overarching policy statement or strategy; operational requirements which are often mandatorily applied to lending operations; environmental and social procedures which are often mandatorily applied through the preparation, approval, monitoring and supervision of projects; access to information policies; and guidebooks or manuals to support the carrying out of ‘best-practice’ safeguarding procedures. Before 2016, the World Bank’s safeguard policies were mostly horizontally structured through operational policies (OPs) and bank procedures (BPs), without an overarching policy statement. World Bank OPs apply to borrowers, whereas BPs apply primarily to the Bank’s internal operations. An independent Inspection Panel also provides a mechanism for people and communities who believe that they have been, or are likely to be, adversely affected by a World Bank-funded project to raise their concerns and lodge complaints. The World Bank was the first IFI to incorporate an inspection mechanism into its safeguard systems. Since the Inspection Panel was introduced in 1997, other IFIs have copied its model more or less closely and introduced their own mechanisms (Schulte, 2017).

A 2010 IEG evaluation of safeguards and sustainability policies finds that public perceptions of the World Bank’s ability to mitigate social and environmental risks improved during the 2000s, with three-quarters of non-governmental organisation respondents to an IEG survey rating World Bank Group performance better than in the 1990s (IEG, 2010). The evaluation also finds that safeguards and sustainability policies introduced during the 2000s have helped to mitigate social and environmental risks associated with World Bank-financed projects, although it identifies possible improvements in specific areas.

Before 2016, the World Bank lacked an overarching policy statement and principles governing its approach to safeguards, putting it out of step with other IFI/MDB safeguard systems (Himberg, 2015). To address this issue, the World Bank board approved a new Environmental and Social Framework (ESF) in 2016 and expanded protections for people and the environment in World Bank-financed investment projects (World Bank, 2016b).

Application of safeguards

While the 2015 comparative review of MDB safeguarding systems finds that most are relatively similar in terms of overall structure, it also finds important differences between IFIs/MDBs regarding the ways issues are addressed in the respective systems (Himberg, 2015). The major difference between the World Bank and other IFIs/MDBs is that the World Bank applies more formal distinctions to cover project lending versus policy-based lending (which includes IDA’s DPF) and private sector lending (which includes IDA’s PSW), with each instrument governed by a separate OP (Himberg, 2015).

An OP for policy-based financing requires officials to determine whether the country policies supported by DPF are likely to cause significant detrimental effects on the country’s environment and other natural resources, or on poor people and vulnerable groups. It also directs Bank staff to address shortcomings in borrower systems if they are unable to reduce the negative effects of DPF (World Bank, 2017a). Another OP, Performance Standards for Private Sector Projects, is applied to private sector lending (World Bank, 2013). The World Bank’s 2016 ESF, which covers more themes than the array of OPs that pre-date it, therefore only applies to project lending. This has raised concerns about whether the Bank’s safeguard system is best suited to protect the poorest and most vulnerable. Other IFIs/MDBs often apply the same operational safeguards to all three types of lending, but adjust their procedural requirements to particular circumstances (Himberg, 2015). Conservationists have raised concerns that the World Bank’s flexibility in the use and application of safeguards presents a risk to biodiversity and may also compromise future attempts to hold the Bank accountable (Morley et al., 2020). Moreover, it is important to note that officials in recipient countries carry the responsibilities for assessing, managing and monitoring the environmental and social risks and impacts associated with each stage of projects supported by IDA.

According to data from the Inspection Panel’s website, from 1994 to September 2021 the Panel received 153 requests for inspection. While relatively little is known about complaint outcomes and factors that enhance the likelihood of success, one recent independent analysis finds that roughly 50% of Inspection Panel cases result in at least one positive outcome, usually in the form of formal acknowledgements of harm, project change or termination, and compensation (Zvobgo & Graham, 2020). It notes that punishment of perpetrators (that is, officials in recipient countries responsible for applying safeguards) is exceedingly rare. The most frequently raised issues in Inspection Panel cases are for environmental or economic damage and involuntary resettlement.

Thematic coverage of safeguards

Before the introduction of the 2016 ESF, the World Bank’s safeguard coverage was criticised for not including several key risks, such as “biodiversity, ecosystem services, natural resource management”, “climate change” and “disability”. Himberg (2015) notes that “other MDBs provide thematic and/or more detailed coverage” of these impacts and risks. The 2016 ESF expanded the number of risks included for protections, which largely brought the World Bank in line with the coverage of other IFIs/MDBs, with the exception of LGBTI protection. Larsen (2016) finds that the ESF’s definition of indigenous people is narrowed for certain ethnicities, which may introduce barriers to support for some indigenous people. In the consultation period of the ESF, some partner countries expressed a need for explicit protection for the LGBT community, while other partner countries opposed this. The opposition is reflected in the omission of LGBT protection in the framework (Larsen, 2016).

5. Evolution of IDA’s financing model

5.1        Leveraging capital market issuances

As discussed in Section 2, IDA has historically raised most of its resources through donor government contributions and, increasingly, from repayments on previous loans issued (Manning, 2017). However, in 2017, IDA also began issuing debt against its equity to supplement these traditional sources of finance. IDA’s blending of capital market issuances with donor funds is widely perceived as a positive evolution of its financing model, although there is an ongoing debate regarding exactly how much IDA could afford to lend above its available equity and reserves.

Some argue that IDA could and should leverage its equity more to increase the amount of financing available for development in poor countries (Humphrey, 2020), particularly in the wake of the COVID-19 crisis. Civil society groups such as Oxfam and Save the Children call for IDA to make a supplementary budget of roughly equivalent value to the $35 billion frontloading amount available to eligible countries under IDA20 (Oxfam, 2021; Save the Children, 2020). World Bank officials contend that IDA cannot leverage more resources without threatening the Bank’s AAA bond rating (World Bank, 2020b), which some authors argue is essential for IDA to be able to raise funds on capital markets in the long term at reasonably low cost (which can then be transferred to borrowers in the form of low interest rates). Humphrey (2018, p. i) finds that the diverse, frequently changing and often conservative and opaque approaches credit agencies’ take to evaluating MDBs “restricts their operational capacity” as they “face uncertainty on how to assess potential lending headroom”.

On the other hand, emerging evidence suggests that IDA could leverage additional resources by optimising its balance sheet without jeopardising the World Bank’s AAA bond rating (Humphrey & Prizzon, 2020; Humphrey, 2021). Humphrey (2020) estimates that IDA lending could increase by as much as 160% above current levels without threatening the World Bank’s AAA status. Moreover, this could rise to as much as 300% if the Bank were willing to accept a one-level downgrade to AA+ (Humphrey, 2020).

Although IDA has more equity than any other MDB soft fund, it leverages significantly less than the International Bank for Reconstruction and Development (IBRD). Gavas (2020) finds that IDA’s outstanding capital market debt is less than 10% of its equity base, compared to 600% for IBRD. Part of this is because IDA has a lower-quality loan portfolio (that is, it has lent to borrowers who are less likely to repay, at lower interest rates and with longer maturity periods). IDA also runs a built-in deficit between its borrowing costs, which are market-determined, and its fixed on-lending terms, including outright grants as well as credits well below market terms. This structural gap needs to be funded continuously by capital and capital-like injections from its supporting governments and constitutes a further source of risk if market terms worsen.

Overall, there is inconclusive evidence on whether IDA could leverage additional resources without undermining the long-term sustainability of its financial model. The unique characteristics of international financial institutions (IFIs) and multilateral development banks (MDBs) (for example, their preferred credit status[1]) make evaluating their capital adequacy compared to commercial banks particularly difficult (Humphrey, 2021). The IDA Balance Sheet Optimisation paper discusses possible adjustments to the model IDA uses to calculate how much it can afford to lend, but is sensitive to assumptions about: potential arrears and/or formal loan forgiveness arrangements, if any; increases in future market rates; and changes in the mix of maturities and interest terms offered to IDA borrowers.

5.2        Leveraging private investment

Another heavily discussed evolution of IDA’s financing model (or rather, its deployment model) has been the deployment of concessional resources and the creation of an incentive framework to leverage or mobilise additional private investment into IDA countries. In recent years, the World Bank and other IFIs have been challenged to increase the flow of private investments into the world’s poorest and most fragile countries, based on the idea that additional private investments generate economic growth and create jobs and thus lift people out of poverty (Attridge & Engen, 2019). Investments are also seen as a less volatile source of external capital than debt for poor countries and therefore help to stabilise current accounts and exchange rates (Gallagher & Zarsky, 2007).

Poor countries are generally seen as high-risk by commercial investors due to weak infrastructure, uncompetitive domestic firms and low levels of technical and vocational educational skills. Additionally, countries affected by fragility, conflict and violence (FCV) also suffer from high levels of political risk (Mayer, 2018; Ginsberg, 2013). Despite average rates of return on investment into poor and FCV countries being considered high (Mayer, 2018), these risks deter private sector investments in IDA countries. While the World Bank estimated that 5% to 8% of total foreign investments into developing countries between 2005 and 2010 went to FCV countries, which was commensurate with their share of the developing world’s overall GDP, this finance was concentrated in a small number of resource-rich economies (MIGA, 2011). The Organisation for Economic Cooperation and Development (2015) finds that the majority of foreign investments into fragile states in 2012 went to just six countries[2] for use primarily in the extractives sector (OECD, 2015).

Finance provided by IFIs is traditionally seen as a replacement for private sector investment, (that is, IFIs provide development finance to poor and fragile countries underserved by risk-averse private investors). However, in recent years, the idea emerged that the actions of IFIs could and should be additional to the private sector, rather than replacing it. Leading World Bank economists argued in 2015 that the reorientation of IFIs from lenders for major development projects to brokers for private sector investments carries the potential to expand the total stock of development finance into the world’s poorest countries from “billions to trillions” (World Bank, 2015a).

Benn, Sangare & Hos (2016) conceptually define the terms ‘mobilise’ and ‘leverage’ within the context of aid additionality as referring to the ways in which mechanisms stimulate the allocation of additional financial resources to particular objectives. ‘Leverage’ is usually associated with a quantitative indicator such as a leverage ratio, while ‘mobilise’ refers to a causal link between the private finance made available for a specific project and the official flows that were used to incentivise this finance.

The core mechanism through which IFIs mobilise and leverage private sector investments is blended finance, where concessional finance is used to leverage additional private investment. The Private Sector Window (PSW) is IDA’s first blended finance instrument, introduced under IDA18, that aims to mobilise private sector investments in underserved sectors and markets by reducing their financial risk. Its rationale is that deploying limited amounts of concessional financing improves the risk-reward profile of development projects in poor and fragile countries which, due to high levels of commercial and/or political risk, would not otherwise go ahead. IDA allocated $2.5 billion for the PSW under IDA18 (World Bank, 2017b).

The World Bank’s blended finance works by allocating concessional IDA funds to support investments made and insurance products provided by other more private sector-focused arms of the Bank, namely the International Finance Corporation (IFC) and the Multilateral Investment Guarantee Agency (MIGA). IFC makes investments in private sector enterprises which contribute to the economic development of member countries, whereas MIGA provides guarantees to protect private sector investors against non-commercial risks. Blending IDA financing with IFC financing enables IFC to undertake investments in riskier private sector projects (that is, projects which are not yet able to meet fully commercial financing terms but have the potential to deliver development impact). Blending IDA financing with MIGA guarantees and other risk mitigation insurance products enables MIGA to provide protection to investments in riskier, but potentially more impactful, private sector projects. The overall objective of blended finance within the World Bank is therefore to crowd greater amounts of financing into private sector projects in IDA countries. This, in turn, increases the flows of foreign investments into IDA countries through reducing the commercial and non-commercial risks attached to potential investment projects.

IEG (2021d) assessed the early implementation of the PSW under IDA18, from FY2018 to FY2020, and focused on usage, governance, additionality (that is, the extent to which IDA co-financed other World Bank and private sector investments in IDA-eligible countries) and concessionality (that is, the level of concessional financing required from IDA to make investments commercially viable). It was too early for IEG to evaluate the development outcomes of investment projects supported by blended finance (IEG, 2021d).

The overall findings of this evaluation were that the World Bank’s approaches to mobilising private investments are relevant and have delivered results, with approximately $32 billion in low- and middle-income countries mobilised in 2018 alone. The World Bank also achieved its aim to focus blended finance on FCV countries; 24% of all PSW approvals and 45% of all PSW-supported projects from FY2018 to FY2020 were in FCV countries. However, overall usage of the PSW under IDA18 was found to be well below originally allocated amounts, with only $537 million of the original $2.3 billion allocation committed by the end of FY2020.

IEG (2021d) attributes this sluggish start-up to a limited pipeline of viable investment projects, strict eligibility criteria for investment projects and a protracted approval process for investment projects which requires involvement from multiple arms of the World Bank in addition to IDA. It is also important to note that IDA does not blend finance with private sector lending windows of other IFIs/MDBs, only those within the World Bank Group, which could limit deal flow.

In terms of the additionality of blended finance-supported investments, IEG (2021d) finds that the share of new IFC commitments in IDA-eligible countries did not increase and the share of new MIGA guarantee issuance in IDA-eligible countries increased only marginally.[3] For every dollar of concessional IDA funding committed through blended finance between FY2018 and FY2020, $2.20 in funds from other arms of the World Bank and about $3.60 in funds from private investors were leveraged. The $537 million committed by IDA therefore mobilised $1,949 million in private sector investment, a relatively small proportion of the overall private sector investment mobilised by the World Bank ($32 billion in 2018 alone).

This is perhaps unsurprising given that financial risks are just one of many constraints to private investments in poor and FCV countries. The financial risk mitigation that blended finance provides only addresses only one of the factors deterring private sector investments in IDA countries. Attridge & Engen (2019) find that the average leverage ratios of IFIs and MDBs in low-income countries are relatively small (1:0.37). Moreover, Lee & Cardenas (2020) find that the overwhelming majority of IFI blended finance goes to middle-income rather than low-income countries, and that infrastructure transactions with private participation in poor countries have stagnated despite significant efforts by IFIs to mobilise private investment in these projects. Kenny (2021) calls for IDA to leave the PSW out of the next replenishment entirely or, failing that, for funds to be rapidly returned to the main IDA financing pool if they are not disbursed quickly enough or do not have a strong enough focus on the most vulnerable.

[1] This means that debt-distressed countries borrowing from IDA are expected to give priority to meeting their obligations to IDA over those to other creditors.

[2] Nigeria, Iraq, Democratic Republic of Congo, Egypt, Congo and Sudan.

[3] The share of IFC investments in PSW-eligible countries averaged 14% between FY2010 and FY2017 and 13% during the implementation of the IDA PSW from FY2018 to FY2020. MIGA’s new guarantee issuance averaged 20% between FY2010 and FY2017 and 22% between FY2018 and FY2020.

6. Bibliography

Archibong, B., Coulibaly, B. & Okonjo-Iweala, N. (2021) Washington Consensus reforms and economic performance in sub-Saharan Africa: lessons from the past four decades. Washington, DC: Brookings.

Atingi, M. (2019) Paper presented at the plenary session during the AERC (African Economic Research Consortium) biannual research workshop. Cape Town, African Economic Research Consortium.

Attridge, S. & Engen, L. (2019) Blended finance in the poorest countries: the need for a better approach. London: ODI.

Baker, J. (2000) Evaluating the impact of development projects on poverty: a handbook for practitioners. Washington, DC: World Bank.

Banerjee, A. & Duflo, E. (2012) Poor economics. New York: Public Affairs.

Barder, O., Ritche, E. & Rogerson, A. (2019). “Contractors or collectives?” Earmarked funding of multilaterals, donor needs and institutional integrity: the World Bank as a case study. Washington, DC: CGD.

Benn, J., Sangare, C. & Hos, T. (2016) Measuring private finance mobilised for sustainable development. Paris: OECD Publishing.

Birdsall, N. et al. (1993) The East Asian miracle: economic growth and public policy. Washington, DC: World Bank Publications.

Birdsall, N. & Diofasi, A. (2015) Global public goods for development: how much and what for. Washington, DC: CGD.

Birdsall, N. & Morris, S. (2016) Multilateral development banking for this century’s development challenges: five recommendations to shareholders of the old and new multilateral development banks. s.l.: s.n.

Bond, P. & Brown, D. (2020) World Bank’s rating obsession will negate debt justice. London: Bretton Woods Project.

Bosch, E., Fabregas, A. & Fischer, F. (2020). Earmarked funding to multilateral organisations: how is it used and what constitutes good practice? Paris: OECD.

Bottan, N., Hoffman, B. & Vera-Cossio, D. (2020) The unequal impact of the coronavirus pandemic: evidence from seventeen developing countries. Washington, DC: IADB.

Bourguignon, F., Ferreira, F. & Walton, M. (2006) Equity, efficient and inequality traps: a research agenda. KSG Working Paper.

Bretton Woods Project (2019) What are the main criticisms of the World Bank and the IMF? London: Bretton Woods Project.

Bretton Woods Project (2021a) Open letter to World Bank Group (WBG) President David Malpass and WBG leadership. London: Bretton Woods Project.

Bretton Woods Project (2021b) World Bank’s new Climate Change Action Plan fails to deliver much-needed transformative agenda. London: Bretton Woods Project.

CGD (2021) QuODA 2021. London: CGD.

Chang, H.-J. (2003) Kicking away the ladder. s.l.:Anthem Press.

Cormier, B. & Manger, M. (2020) The evolution of World Bank conditionality: a quantitative text analysis. s.l., 13th annual conference on the political economy of international organisation.

Coyle, D. (2014) GDP: a brief but affectionate history. Princeton: Princeton University Press.

Crost, B., Felter, J. & Johnston, P. (2014). Aid under fire: development projects and civil conflict. American Economic Review, 1833-56.

Das, M. B. (2016) Social inclusion in macro-level diagnostics. Washington, DC: World Bank.

DFID (2016) MDR 2016: World Bank. London: DFID.

DFID (2018) 2018 annual review of IDA. London: DFID.

DFID (2019) 2019 annual review of IDA. London: DFID.

Djankov, S., Montalvo, J. & Reynal-Querol, M. (2008) The curse of aid. Journal of Economic Growth, 169-94.

Dollar, D. & Burnside, C. (2000) Aid, policies, and growth. American Economic Review, 90, 847-68.

Duggan, J., Morris, S., Sandefur, J. & Yang, G. (2020) Is the World Bank’s COVID-19 crisis lending big enough, fast enough? New evidence on loan disbursements. Washington, DC: CGD.

Easterly, W. (2000) The effect of IMF and World Bank programs on poverty. Washington, DC: World Bank.

Easterly, W. (2003) Can foreign aid buy growth? Journal of Economic Perspectives, 17, 23-48.

Easterly, W. (2006) Why doesn’t aid work? Cato Unbound blog, 2 April.

Elliott, L. (2016) ‘The World Bank and the IMF won’t admit their policies are the problem’. The Guardian, 9 October.

Engel, L. & Prizzon, A. (2018) A guide to multilateral development banks. London: ODI.

Fioramonti, L. (2014) How numbers rule the world. The use and abuse of statistics in global politics. London: Zed Books.

Fleurbaey, M. & Blanchet, D. (2013) Beyond GDP. Measuring welfare and assessing sustainability. Oxford : Oxford University Press.

Galiani, S., Knack, S., Xu, L. & Zou, B. (2014). The effect of aid on growth: evidence from a quasi-experiment, Washington, DC: World Bank Working Papers.

Gallagher, K. & Zarsky, L. (2007) The enclave economy: foreign investment and sustainable development in Mexico’s Silicon Valley. Cambridge, MA: MIT Press.

Gavas, M. (2020). Strategic reset: how bilateral development agencies are changing in the COVID-19 era. Washington, DC: CGD.

Geo-JaJa, M. & Mangum, G. (2001) Structural adjustment as an inadvertent enemy of human development in Africa. Journal of Black Studies, 32(1), 30-49.

Ginsburg, R. (2013) Political risk insurance and bilateral investment treaties. Journal of World Investment and Trade, 943-977.

Greenhill, R. & Rabinowitz, G. (2016) Why do donors delegate to multilateral organisations? A synthesis of six country case studies. London: ODI.

Greenhill, R. & Rabinowitz, G. (2017) Financing ‘leave no one behind’: policy options in the social sectors. London: ODI.

Gulrajani, N. (2016) Bilateral versus multilateral aid channels: strategic choices for donors. London: ODI.

Guven, A. (2018) Whither the post-Washington Consensus? International financial institutions and development policy before and after the crisis. Review of International Political Economy, 25(3), 392-417.

Gwin, C. (2002) IDA’s Partnership for Poverty Reduction : an independent evaluation of fiscal years 1994-2000. Washington, DC: World Bank.

Hartford, T. & Klein, M. (2005) Aid and the resource curse: how can aid be designed to preserve institutions? Washington, DC: World Bank.

Hattie, A. (2020) Climate adaptation finance: fact or fiction? Amsterdam: CARE Netherlands & CARE Denmark.

Himberg, H. (2015) Comparative review of multilateral development bank safeguard systems. Washington, DC: World Bank.

Honig, D. (2018) Navigation by judgment: why and when top-down management of foreign aid doesn’t work. New York: Oxford University Press.

Honig, D. (2020) Actually navigating by judgment: towards a new paradigm of donor accountability where the current system doesn’t work. Washington, DC: CGD.

Hoy, C. & Sumner, A. (2020) Growth with adjectives: global poverty and inequality after the pandemic. Washington, DC: CGD.

Humphrey, C. (2014) The politics of loan pricing in multilateral development banks. Review of International Political Economy, 21(3), 611-639.

Humphrey, C. (2017). Six proposals to strengthen the finances of multilateral development banks. London: ODI.

Humphrey, C. (2018) The role of credit rating agencies in shaping multilateral finance: recent developments and policy options. s.l.: Inter-Governmental Group of 24.

Humphrey, C. (2020) All hands on deck: how to scale up multilateral financing to face the Covid-19 crisis. London: ODI.

Humphrey, C. (2021) New shareholders for multilateral banks: a viable approach to increase development finance? London: ODI.

Humphrey, C. (2021) The case for an external review of multilateral development bank capital adequacy. London: ODI.

Humphrey, C. & Prizzon, A. (2020) Scaling up multilateral bank finance for the Covid-19 recovery. London: ODI.

ICAI (2018) Analysis of references to safeguarding in DFID’s existing guidance on programming. s.l.: unpublished.

IEG (2010) Safeguards and sustainability policies in a changing world: an independent evaluation of World Bank Group experience. Washington, DC: World Bank.

IEG (2015) Results and performance of the World Bank Group (RAP) 2015. Washington, DC: World Bank.

IEG (2016a) Behind the mirror: a report on the self-evaluation systems of the World Bank Group, Washington, DC: World Bank.

IEG (2016b) Industry competitiveness and jobs: an evaluation of World Bank Group industry-specific support to promote industry competitiveness and its implications for jobs. Washington, DC: World Bank.

IEG (2016c) World Bank Group engagement in situations of fragility, conflict, and violence. Washington, DC: World Bank.

IEG (2017) Toward a clean world for all: an evaluation of the World Bank Group’s support to pollution management. Washington, DC: World Bank.

IEG (2018a) Carbon markets for greenhouse gas emission reduction in a warming world. Washington, DC: World Bank.

IEG (2018b) Engaging citizens for better development results. Washington, DC: World Bank.

IEG (2018c) Maximizing the impact of development policy financing in IDA countries. Washington, DC: World Bank.

IEG (2018d) Niger – completion and learning review. Washington, DC: IEG.

IEG (2019a) Building urban resilience: an evaluation of the World Bank Group’s evolving experience (2007-2017). Washington, DC: World Bank.

IEG (2019b) IDA Regional Window Program 2003-17. Washington, DC: World Bank.

IEG (2019c) Learning from IDA experience: lessons from IEG evaluations, with a focus on IDA special themes and development effectiveness. Washington, DC: World Bank.

IEG (2019d) World Bank Group support in situations involving conflict-induced displacement. Washington, DC: World Bank.

IEG (2019e) World Bank support for irrigation service delivery: responding to new challenges and opportunities. Washington DC: World Bank.

IEG (2020a) Evaluation of the World Bank Group’s support for electricity supply from renewable energy resources, 2000–2017. Washington, DC: World Bank.

IEG (2020b) Niger – Community Action Program and Community-Based Integrated Ecosystem Management Project Phase I and II (English). Washington, DC: World Bank.

IEG (2020c) Recent World Bank experience with RRAs & operational programming in FCV countries. Washington, DC: World Bank.

IEG (2020d) Results and performance of the World Bank Group 2020. Washington, DC: World Bank.

IEG (2020e) Sierra Leone – completion and learning review. Washington, DC: IEG.

IEG (2021a) Early-stage evaluation of the International Development Association’s Sustainable Development Finance Policy. Washington, DC: World Bank.

IEG (2021b) Results and performance of the World Bank Group 2021. Washington, DC: World Bank.

IEG (2021c) The natural resource degradation and vulnerability nexus: an evaluation of the World Bank’s support for sustainable and inclusive natural resource management (2009–2019). Washington, DC: World Bank.

IEG (2021d) The World Bank Group’s experience with the IDA Private Sector Window: an early-stage assessment. Washington, DC: World Bank.

IEG (2021e) World Bank Group gender strategy mid-term review. Washington, DC: World Bank.

IEG (2021f) World Bank engagement in situations of conflict. Washington, DC: World Bank.

ILO (2021) ILO monitor: COVID-19 and the world of work. Seventh edition. Geneva: ILO.

International Development Association (1960) Articles of Agreement, Washington, DC: World Bank.

International Development Association (2020). Sustainable Development Finance Policy: promoting sustainable borrowing and lending practices in IDA countries. Washington, DC: World Bank.

ISER Uganda (2019) Failing to reach the poorest? Assessment of the World Bank funded Uganda Reproductive Health Voucher Project. Kampala: ISER Uganda.

Islam, M. (2005) Regime changes, economic policies and the effect of aid on growth. Journal of Development Studies, 41(8), 1467-1492.

ITUC (2021) Quality jobs and sustainable economic transformation: recommendations for the 20th replenishment of the World Bank’s International Development Association. Brussels: ITUC.

Karr, V., van Edema, A., Sims, A. & Brusegaard, C. (2017) No one left behind: a review of social protection and disability at the World Bank. Disability in the Global South, 4(1), 1112-1142.

Kaul, I. (2012) Global public goods: explaining their underprovision. Journal of International Economic Law, 15(3), 729-750.

Kaul, I. (2017) Providing global public goods: what role for multilateral development banks? London: ODI.

Kenny, C. (2021) Go big on IDA, but leave out the Private Sector Window. Washington, DC: CGD.

Khan, A. (2019) Is the World Bank leaving no one behind? London: ODI.

Kharas, H. (2011) Measuring aid effectiveness effectively: a quality of official development assistance index. Washington, DC: Brookings.

Kohl, B. (2002) Stabilizing neoliberalism in Bolivia: popular participation and privatization. Political Geography, 21(4), 449-472.

Landers, C. (2021) IDA-20: donors must go big, and IDA must too. s.l.: ODI.

Landers, C. & Aboneaaj, R. (2021) World Bank budget support in the time of COVID: crisis finance… with strings attached. Washington, DC: CGD.

Larsen, G. (2016) The World Bank just overhauled its safeguards for the first time in 30 years. s.l.: World Resources Institute.

Lee, N. & Cardenas, M. (2020) The future of private finance for development in poor countries. Washington, DC: CGD.

Leigh, C. et al. (2019) Follow the money: equitably financing child survival. London: Save the Children UK.

Lucci, P., Khan, A., Turner, J. & Sims, J. (2019) Is the World Bank leaving no one behind? London: ODI.

Mainhardt, H. (2017) Funding clean energy access for the poor: can the World Bank meet the challenge? s.l.: Bank Information Center Europe, the Swedish Society for Nature Conservation and The Big Shift Global.

Manning, R. (2017) Multilateral development aid: assessing the major replenishments of 2016. Tokyo: UNU-WIDER.

Manuel, M. & Rogerson, A. (2021) “Two birds, no stones”: a global initiative to combine rural social protection with climate resilience. London: ODI.

Mavrotas, G. (2002) Foreign aid and fiscal response: does aid disaggregation matter? Weltwirtschaftliches Archiv, 534-59.

Mavrotas, G. & Outtara, B. (2003) Aid disaggregation, endogenous aid and the public sector in aid-recipient economies: evidence from Côte d’Ivoire. Tokyo: United Nations University.

Mayer, H. (2018) Political risk insurance and its effectiveness in supporting private sector investment in fragile states. Oxford: LSE-Oxford Commission on State Fragility, Growth and Development.

McGillivray, M. & Morrissey, O. (2001) A review of evidence on the fiscal effects of aid. Nottingham: Centre for Research in Economic Development and International Trade, University of Nottingham.

MIGA (2011) World investment and political risk report 2010. Washington, DC: World Bank.

Miller, M., Prizzon, A., Pudussery, J. & Roger, L. (2021) How the allocation of IDA20 should shift to ease the recovery from the Covid-19 crisis in lower-income countries. London: ODI.

Mitchell, I. & Caitlin, M. (2018) How do you measure aid quality and who ranks highest? London: CGD.

MOPAN (2016) The World Bank institutional assessment report. Paris: The MOPAN Secretariat.

MOPAN (2020) MOPAN 3.1 methodology manual. Paris: OECD.

Morley, J., Buchanan, G., Mitchard, E. & Keane, A. (2020) Implications of the World Bank’s environmental and social framework for biodiversity. Journal of the Society for Conservation Biology.

Morris, S. (2020) Harder times, softer terms: assessing the World Bank’s new Sustainable Development Finance Policy amidst the COVID crisis. Washington, DC: CGD.

Nunn, N. & Qian, N. (2014). ‘The determinants of food provision to Africa and the developing world’. In: African successes: sustainable growth. Chicago: University of Chicago Press.

OECD (2011) The Busan Partnership for Effective Development Co-operation. Paris: OECD.

OECD (2015) States of fragility 2015: meeting post-2015 ambitions. Paris: OECD Publishing.

OECD (2019a) Development finance standards. s.l.: s.n.

OECD (2019b) Untied aid. [Online]
Available at:
[Accessed 17 October 2021].

OECD (2020) Multilateral development finance 2020. Paris: OECD Publishing.

Oxfam (2020) From catastrophe to catalyst: can the World Bank make COVID-19 a turning point for building universal and fair public healthcare systems? Oxford: Oxfam UK.

Oxfam (2021) Oxfam position paper on IDA20 replenishment. Oxford: Oxfam UK.

Palagashvili, L. & Williamson, C. (2014). Ranking foreign aid agency best practies: new donors, new findings. SSRN Electronic Journal.

Qian, N. (2015) Making progress on foreign aid. Annual Review of Economics, 7, 277-308.

Sacchetto, C., Collier, P., Kriticos, S. & Logan, S. (2021) Strengthening development finance in fragile contexts: policy brief. Oxford, UK: International Growth Centre.

Schulte, S. S. (2017) ‘Sustainable development priorities in World Bank Inspection Panel decisions’. In: M. Cordonier Segger (ed.) Sustainable development principles in the decisions of international courts and tribunals. London: Routledge, 30.

Severino, J.-M. & Moss, T. (2012) Final report of the future of IDA working group. Washington, DC: CGD.

Stein, H., 1992. Deindustrialization, adjustment, the World Bank and the IMF in Africa. World Development, 20(1), 83-95.

Stiglitz, J. (2002). Globalisation and its discontents. New York: W.W. Norton & Company.

Stiglitz, J., Sen, A. & Fitoussi, J.-P. (2010) Report by the Commission on the measurement of economic performance and social progress. Brussels: European Commission.

Svensson, J. (2000) Foreign aid and rent-seeking. Journal of International Economics, 437-61.

Tortora, P. & Steenson, S. (2014) Making earmarked funding more effective: current practices and a way forward. Paris: OECD Publishing.

UNCTAD (2021) Out of the frying pan … into the fire? – Trade and development report 2020 update (March 2021., Geneva: UNCTAD.

Urgewald (2020) World Bank annual meeting: bank invested over $10.5 billion in fossil fuels since Paris Agreement. Berlin: Urgewald.

US Congressional Research Service (2020) Multilateral development banks: overview and issues for Congress. Washington, DC: s.n.

Watts, R. (2020) The case for a supplementary IDA budget – a lifeline for children. London: Save the Children.

Were, M. & Mollel, L. (2020) Public debt sustainability and debt dynamics. World Institute for Development Economics Research working paper series 2020-112.

Wood, C. (2018) Is New Zealand really the world’s best donor? Canberra: Development Policy Centre.

World Bank (2002) Integrating gender into the World Bank’s work: a strategy for action. Washington, DC: World Bank.

World Bank (2006) World development report 2006: equity and development., Washington, DC: World Bank.

World Bank (2012) World development report 2012: Gender equality and development. Washington, DC: World Bank.

World Bank (2013) Performance standards for private sector activities. Washington, DC: World Bank.

World Bank (2014) IDA17: maximizing development impact. Washington, DC: World Bank.

World Bank (2015a) From billions to trillions: MDB contributions to financing for development. Washington, DC: World Bank.

World Bank (2015b) Gender equality, poverty reduction, and inclusive growth. Washington, DC: World Bank.

World Bank (2016a) Special theme: jobs and economic transformation. Washington, DC: World Bank.

World Bank (2016b) The World Bank Environmental and Social Framework. Washington, DC: World Bank.

World Bank (2017a) Bank policy: Development Policy Financing. Washington, DC: World Bank.

World Bank (2017b) Report from the Executive Directors of the International Development Association to the Board of Governors: additions to IDA resources – eighteenth replenishment (English). Washington, DC: World Bank.

World Bank (2018a) Disability Inclusion and Accountability Framework. Washington, DC: World Bank.

World Bank (2018b) IDA17 retrospective – maximizing development impact. Washington, DC: World Bank.

World Bank (2018c) PovcalNet (online analysis tool). [Online]
Available at:

World Bank (2019) Special theme: fragility, conflict & violence. Washington, DC: World Bank.

World Bank (2020a) Additions to IDA resources: nineteenth replenishment – ten years to 2030: growth, people, resilience (English). s.l.: s.n.

World Bank (2020b) IDA balance sheet optimization paper. Washington, DC: World Bank.

World Bank (2020c) IDA18 Results Measurement System. Washington, DC: World Bank.

World Bank, 2020d. World Bank Group strategy for fragility, conflict, and violence 2020-2025, Washingon, DC: World Bank.

World Bank (2021a) 2021 year in review in 11 charts: the inequality pandemic. Washington, DC: World Bank.

World Bank (2021c) IDA19: implementation of policy commitments and results update. Washington, DC: World Bank.

World Bank (2021d) IDA20 special theme: human capital. Washington, DC: World Bank.

World Bank (2021e) PRESENTATION: WBG Climate Change Action Plan 2021-2025. Washington, DC: World Bank.

World Bank (2021f). World Bank Climate Change Action Plan 2021-25 – supporting green, inclusive development. Washington, DC: World Bank.

World Bank (2021g) IDA18 retrospective: investing in growth, resilience, and opportunity through innovation. Washington, DC: World Bank.

World Bank (2022) What is IDA? Washington, DC: World Bank.

World Bank and the International Monetary Fund (2021) World Bank Group and International Monetary Fund support for debt relief under the Common Framework and beyond. Washington, DC: World Bank.

World Development Movement (2005). One size for all: a study of IMF and World Bank poverty reduction strategies. s.l.: Global Justice Now.

Zvobgo, K. & Graham, B. (2020) The World Bank as an enforcer of human rights. Journal of Human Rights, 19(4), 425-448.