What we do



Follow ODI

How the allocation of IDA20 should shift to ease the recovery from the Covid-19 crisis in lower-income countries


Written by Mark Miller, Annalisa Prizzon, Jessica Pudussery, Lionel Roger

As a response to the scale and the length of the Covid-19 crisis in lower-income countries, shareholders decided to bring the negotiations of the replenishment of the International Development Association (IDA) forward by a year. This decision averted a major drop in IDA grants and loans from next year. In a previous Insight, we argued that IDA would have been forced to reduce its lending to a projected $22.5 billion each year in FY22, from $35 billion in the current FY21.

The increase in IDA grants and loans from 2022 is welcome news indeed. But how should these IDA commitments be deployed across countries, sectors and instruments to ease the recovery from the Covid-19 crisis and help address the multiple shocks lower-income countries are facing? Here we argue that the current country allocation model is not fit and meant for this purpose. The Crisis Response Window (CRW) should be far larger than in previous replenishment rounds. Its allocation criteria should also reflect the multiple shocks lower-income countries are experiencing on their health systems, societies and economies. Furthermore, overall IDA assistance should be deployed with the most flexible instruments available in its toolkit to reach countries as fast as possible to boost local economies.

The country allocation of core IDA resources

The country allocation of core IDA resources is based on two main components. First, the country’s governance and ability to spend resources well (i.e. the country performance rating, a combination of the Country Policy and Institutional Assessment and the portfolio performance): countries with better institutions and policies are expected to receive a larger allocation of grants and loans, all other things being equal. This is the largest component in the IDA allocation formula. Second, the needs of the recipient country, proxied by the income per capita (but inversely correlated – the lower the income per capita, the higher the IDA allocation) and the country’s population.

The correlation between the IDA country allocation – per capita – is stronger with the Country Performance Rating (CPR) than with the average income per capita, as illustrated in Figure 1a and 1b, once small island developing states (SIDS) are excluded. In principle, it is not surprising as the CPR has greater weight in the allocation formula. In Figure 1b, per capita IDA tends to be higher, on average, in countries with higher gross national income (GNI) per capita – the lower the income per capita, the greater the allocation should be, but that refers to the total country’s allocation rather than per capita in the allocation formula.

Figure 1: 1a: Country Performance Rating (2018) vs IDA per capita country allocation (log) | 1b: Figure 1b: GNI per capita (Atlas Method) vs IDA per capita country allocation (log)

(To view the figures in the accompanying PDF, please see p. 4.)

These two broad criteria do not capture the impact of major macroeconomic shocks. For example, at the time of the Global Financial Crisis in 2008–2009, the World Bank Independent Evaluation Group found that ‘most new lending in response to the crisis reflected pre-crisis lending patterns and had a low correlation with the severity of the crisis impact’, especially in IDA countries where flexibility was more limited. Solely looking at the economic impact of the crisis, those economies that have been the most affected by the Covid-19 crisis would not be the main beneficiaries of IDA assistance, should core allocation rules apply.

The correlation between the average per capita IDA allocation in the IDA18 cycle and the impact of the crisis on some macroeconomic indicators is rather mild or non-existent, true whether including or excluding small island developing countries. Among these indicators we considered, for example, the recovery to pre-crisis levels of economic activity (the difference between gross domestic product (GDP) per capita forecast in 2024 and its pre-crisis value in 2019 as a proxy) (Figure 2a); the changes in the current account before the crisis in 2019 with the latest projections for next year (2022) (Figure 2b); the changes in export projections before and after the crisis (Figure 2c); the per capita revenue losses between pre-crisis levels in 2019 and the end of the IDA20 cycle in 2024 (Figure 2d). The correlations of these variables with the IDA per capita allocation by country are weak or non-existent.

Figure 2: The correlation between past IDA allocation and four macroeconomic indicators

(To view the figures in the accompanying PDF, please see p. 6.)

Allocating resources through the Crisis Response Window

Beyond its long-term development objectives, IDA also plays a role to help countries affected by a crisis to return to their pre-shock levels of economic activity. In the most recent IDA cycles, the Crisis Response Window was introduced to address short-term economic and health crises and emergencies, and natural disasters (partly avoiding the risk of replicating the lending patterns during the 2008–2009 Global Financial Crisis – see Box 1). The amounts of the CRW have historically been small and restricted in scope.

Table 1: CRW – financial resources
Start End IDA replenishment
(US$ billion)
CRW (percentage of replenishment) CRW financing (~US$ billion) CRW commitments (US$ billion) Notes
IDA15 2009 2011 41.6 4% ~1.6 1.4 Piloted; included economic shocks only
IDA16 2012 2014 49.3 Capped at 5% ~2 0.9 Extended to include natural disasters
1DA17 2015 2017 52.1 3% 1.8 1.8 Extended to include public health emergencies as a response to Ebola
IDA18 2018 2020 75.0 4% 3
IDA19 2021 2023* 82.0 3% 2.5

Figure 3: CRW – sector allocation – commitments

Figure 3: CRW – sector allocation – commitments. Comparing IDA allocations 15 to 18 for public health emergencies, natural disasters and economic shocks.

The nature and scale of the shock from the Covid-19 pandemic requires an expansion of the CRW and also a revisiting of the criteria for accessing resources. The shock triggered by the Covid-19 pandemic has reached lower-income countries through multiple and mutually reinforcing channels: the slowdown in the global economy, lockdowns and travel restrictions, and the health emergency. Its consequences on economic and social development in lower-income countries and on their ability to tackle the still ongoing health emergency have been far-reaching and severe (Figure 4).

Figure 4: Transmission channels of the Covid-19 pandemic in lower-income countries

Figure 4: Transmission channels of the Covid-19 pandemic in lower-income countries. Showing the relationship between transmission channels, mediating factors and health, social and economic impacts.

Existing criteria may allow for the flexible handling of a dynamically evolving situation, but the criteria should be adapted and reflect the nature of the shocks and the response to the Covid-19 crisis in three areas:

  • Macroeconomic impact. While country diagnostics might highlight specific dimensions, these are difficult to run within a short timeframe and across many economies. Therefore, some common metrics measuring the impact of the crisis across economies should be considered for the allocation of resources in the CRW. We proposed some potential indicators in Figure 2, for example:
    – Allocating IDA funds to those countries that are estimated to have suffered the greatest and most persistent income losses can help close some of the gaps the crisis has caused, reflecting the expected impact on GDP per capita levels. This is particularly true if funds are directed towards expenditure that has a high multiplier attached to it. This includes the financing of infrastructure projects, extending emergency credit to hard-hit industries, and social expenditure to both stimulate domestic demand and mitigate the losses suffered by the most vulnerable.
    – Many countries felt the impact of the crisis most acutely through forgone foreign exchange (forex) inflows, undermining their ability to finance vital imports, and to maintain exchange rate and price stability, with worsening external balances. IDA is in a good position to help fill this foreign exchange gap. Possible proxies to quantify the impact include the year-on-year difference in the current account or changes in merchandise trade. As many countries paradoxically improved their trade balance because imports contracted even more than exports, it may be adequate to focus on the decline in gross exports to approximate the impact of the crisis on a country’s external position.
    – As a final example, governments’ capacity to finance an adequate response to Covid-19 varied greatly from the onset, and public revenues have further been squeezed by the downturn in economic activity. Allocation under the CRW should therefore partly reflect forgone revenues, in order to prop up fiscal space in those countries where it has been squeezed most severely.
  • Social impact. We know that the impact of the crisis has not been felt ‘evenly’ across countries. Among high-contact industries, the garment industry and tourism – two sectors that are the main drivers of economic growth in many countries – still suffer from low demand from advanced economies and travel restrictions. The social costs of a country with a tourism industry that is being battered by crisis are probably worse than the impacts of a country whose revenues have fallen as a result of commodity price fluctuations. The allocation criteria of the CRW should aim to reflect the source of economic growth affected by the crisis and the impact on jobs, for example.
  • Health impact. Flexibility to address public health emergencies is already included in the criteria for access to resources of the CRW – and introduced as a response to the Ebola crisis in West Africa in IDA17. Yet, the demand for this support might escalate. Lower-income countries whose health systems have been left relatively unscathed in earlier waves of the pandemic might experience severe second and third waves (as recently in Uganda).

Sectoral allocation and instruments for short-term relief support and long-term development objectives

The allocation of IDA resources should balance the need for short-term support for lower-income countries to avoid long-term scarring of their economies with financing for long-term development objectives. We can identify at least three trends in the allocation of IDA resources since 2015 (Figure 5).

  • First, the sharp rise in the share of resources towards social protection over time, especially in response to the Covid-19 crisis in 2020 and 2021.
  • Second, this relative increase comes at the expense of other sectors, especially productive sectors (such as agriculture, industry, trade and services) and infrastructure (energy and transport).

Finally, the proportion of IDA grants and loans to health and education rose in 2020 but fell in 2021. The fall in the relative share of assistance towards health is somewhat surprising given the emphasis on the health emergency from March 2020 onwards but these might be captured in multi-theme/multisector programmes. However, countries are apparently less willing to access IDA (and International Bank for Reconstruction and Development (IBRD)) funding for vaccine procurement, owing to a combination of factors, from stringent vaccine regulatory approvals to restrictions on procurement processes.

Figure 5: Sectoral allocation of IDA project approvals 2015–2021, share of total

Figure 5: Sectoral allocation of IDA project approvals 2015–2021, share of total

IDA has specific purposes and its resources are not limitless: it cannot address all social and economic challenges lower-income countries face. But IDA needs to be ambitious, given the scale of the threats ahead for many lower-income countries

  • As the crisis continues, IDA can support social protection to provide relief from impacts of further lockdowns.
  • It should also help countries in the vaccination rollout in the short term. IDA-eligible countries should be incentivised to expand vaccination programmes because of their benefits to the global economy, and the reduction in transmission rates and the spread of new variants. As vaccines become more freely available, IDA funds could also be usefully used to support health systems with the vaccination rollout.
  • Once the pandemic is brought under control, IDA should consider a greater share of support towards productive sectors and infrastructure to support recovery.

IDA resources should be deployed flexibly. In the immediate aftermath of the Covid-19 crisis the use of development policy finance (DPF) rose compared to 2019. However, more recently, the share of IDA projects approved as DPFs fell again (Figure 6).

Figure 6: IDA project approvals by type of financing (percentage (%) of total)

Figure 6: IDA project approvals by type of financing (percentage (%) of total). Showing the proportions for development policy financing, investment project financing and program-for-results financing.

The share of DPFs on total grants and loans should rise in IDA20 instead, once again to allow for the more flexible and rapid use of resources. In our previous proposal for policy priorities for the IDA replenishment, we argued for a greater share of IDA assistance delivered as development policy financing or budget support, with external financing directly channelled through governments. This provides a much needed balance-of-payments support. It is also estimated to have a greater local economic impact, and would therefore likely have higher short-term fiscal multipliers and be more effective for economic recovery. 'Program-for-results' operations can also be an effective way to provide relatively flexible financial support quickly.


The path of the Covid-19 crisis is still uncertain, especially as the global vaccination programme is not progressing fast enough in lower-income countries. Many countries are still dealing with the emergency of the Covid-19 pandemic rather than being able to set the basis for the long-term rebuilding of their economies. For these reasons, the IDA Crisis Response Window should be much larger than in previous replenishment rounds. Its resources should also be allocated to those lower-income countries most affected by the Covid-19 crisis to address its health, economic and social impacts. We argue for the more flexible deployment of resources and a greater use of the DPF modality (or budget support) than done in recent months across IDA grants and loans.

The IDA20 replenishment offers a landmark opportunity to help finance the emergency and the economic recovery from the Covid-19 crisis in lower-income countries. The allocation of IDA resources should reflect the scale and the nature of this crisis and learn from the lessons at the time of the Global Financial Crisis, aiming to reach the countries most affected and with flexible instruments.