What have we learned after six months of tracking covid-19 funding? We don’t just need more money, we need a different approach.
by Ruth Hill, Dillan Patel and Michèle Plichta [1]
Covid-19 has demonstrated, more than ever before, the weakness of our current begging bowl approach to funding disasters, where the money is found after disaster strikes and is given in a discretionary manner. Large amounts of money have been given at great speed, but it is not enough. And when everyone is in need at the same time, the lack of a plan results in an inequitable allocation of funds.
It has been three months since our previous blog about international funding for the covid-19 response in low- and middle-income countries. With the Annual Meetings of the International Monetary Fund and World Bank taking place next week, now is a good moment to take stock again. Substantial new funds have been committed since our last post—US$87 billion has now been committed by the multilateral organisations that we are tracking—and much of it has been disbursed (69%)[2]. However, this is not enough given the extent of the need, and the pace of new commitments has slowed (Figure 1).
Figure 1: Amount committed and disbursed by multilateral organisations for covid-19 response
Note: Funding from International Monetary Fund (IMF), World Bank Group, Asian Development Bank, African Development Bank, Inter-American Development Bank, Islamic Development Bank, and the United Nations (UN) COVID-19 Global Humanitarian Response Plan (GHRP). We can track disbursements for the IMF, the World Bank, the Asian Development Bank and the UN COVID-19 GHRP. For all repurposed loans by the World Bank, 15 April 2020 is being used as the commitment date, and disbursements are being tracked from April onwards. For UN COVID-19 GHRP flows “paid contributions” count as disbursements and “commitments” count as commitments that have not disbursed. The approval date is counted as the commitment date for paid contributions (or the first day of the same month as the flow month if no approval date is recorded).
Concerningly, the allocation of funds remains inequitable. Funding per capita continues to be less likely to go to poor countries and continues to be less likely to go to countries where poverty is going to increase the most as a result of the covid-19 crisis (Figure 2). It is becoming increasingly clear that we are losing a decade of progress against extreme poverty because of covid-19, yet funds are not being allocated to address it. Countries where it is expected that poverty will increase the least as a result of covid-19 have received US$93 per capita compared to US$26 per capita in countries where poverty is expected to increase the most. (This difference would be even higher if we were also counting loans that these institutions have made to high income countries.)
This suggests that we don’t just need more funding we need funding to be done differently going forward.
Figure 2: Funding per capita has not gone to places with the biggest poverty needs
Note: Pre-covid-19 poverty data are the 2020 pre-covid-19 $1.90 poverty projections referenced in Mahler et al., 8 June 2020. Countries are classified as having a small increase in poverty if the share of the population living on less than $1.90 per day is estimated to increase by less than 0.5 percentage points by Mahler et al. A moderate increase is an increase of between 0.5 and 1 percentage points; a moderately large increase is an increase of between 1 and 2 percentage points; and a large increase is an increase of more than 2 percentage points.
Focusing on poverty—a call for debt relief and need-based funding rules
This inequity is driven by the nature of funding, which is mostly given as loans, not grants. Poorer countries are much more likely to be in debt distress, so cannot borrow and need to rely on grants to meet needs. Aid money that is given as grants is quite well targeted to poorer countries. This is also true for aid money that is used to make the cost of borrowing cheaper (the grant element of concessional lending). But Figure 3 shows that there is room to improve the poverty allocation of some funding. For example, only 46% of the UN COVID-19 GHRP is going to countries where poverty is expected to increase the most. And these hardest hit countries are only getting 29% of all grant elements of IMF concessional loans. Every institution has room to improve the targeting of its grant resources.
Every institution has room to improve the targeting of its grant resources.
The biggest problem is that grant funds are not enough for the poorest countries. Countries that are seeing large increases in poverty, need to be able to borrow to meet the urgent needs they face today. Failing to meet those needs today will have long-run development costs. There is a strong rationale for borrowing to meet those needs today. This points to the urgent need for debt relief for countries in debt distress.
Figure 3: Institutions can improve the poverty-targeting of grant resources
Note: Countries are classified as having a small increase in poverty if the share of the population living on less than $1.90 per day is estimated to increase by less than 0.5 percentage points by Mahler et al., 8 June 2020. A moderate increase is an increase of between 0.5 and 1 percentage points; a moderately large increase is an increase of between 1 and 2 percentage points; and a large increase is an increase of more than 2 percentage points.
Timeliness—being fast requires being prepared
There has also been an important lesson for the future on speed. Financing was fast, much faster than the speed of crisis financing documented in our review of the World Bank’s Crisis Response Window in which the average number of days from shock to disbursement was 183 days for a health emergency. In this crisis, half of funds committed so far had been disbursed 100 days after the pandemic was declared. It is hard for financing arranged post-disaster to get much faster.
But households needed immediate support in April. There were immediate and large impacts of covid-19 lockdowns on incomes of poor households in urban areas. In April, household incomes had declined by an average of 75% among survey respondents in urban Bangladesh, and 80% of survey respondents in Nairobi reported partial or total income losses. This resulted in some of these households reducing their consumption in April with long-run health and cognitive development implications for children in those households. The only instrument that was able to disburse to governments in April were catastrophe-contingent loans that had been put in place in advance of the crisis, such as the World Bank’s Cat-DDOs (Figure 4). Some governments have some flexibility in budgets and may not need immediate funds to provide immediate assistance, but this is not always the case.
Figure 4: Speed of disbursement by instrument (share of total commitment disbursed over time)
Note: UN here refers to paid contributions by a given date as a share of the total paid and committed in the UN COVID-19 GHRP by 2 September 2020
More funds are needed for covid-19 response, but this closer look at the data shows that it is not just more money that is needed. Debt relief and better targeting of grant funding are essential. We also need to move away from the begging bowl of finding money for disasters only after they happen and prepare for future crises with appropriate financing products and equitable allocation rules to be used when crises occur.
[1] Author order has been randomised using the AEA Author Randomization Tool.
[2] We can track disbursements for the IMF, the World Bank, the Asian Development Bank and the UN COVID-19 GHRP. The disbursement share figure refers to the share of those commitments disbursed.