by Aniket Bhushan and Lance Hadley
Published: April 28, 2020
This analysis is the third in a series on risks to development outcomes stemming from donor response to the COVID-19 crisis. We hypothesize that risks to development are longer term are independent of the significant temporal uncertainty (both health and economic), because the crisis is already long and deep enough to set off chains of perverse feedback loops.
The purpose of this analysis is to characterize these feedback loops.
Donor positioning relative to risks is lagging behind the curve. While the probability of irreversible effects is greater than that of recovery to pre-crisis development programming, most donors are publicly positioning to ‘stay the course’. They may need to update thinking rapidly to salvage programming that can be preserved and preserve elements of programming that may be sub-optimal or even unviable but necessary, a process that is inevitably messy.
Ultimately, this analysis highlights the probability that the COVID-19 crisis will cause a fragmentation of donor programming portfolios and that this fragmentation typically favors multilateral institutions – to the detriment of NGOs and CSOs.
Risks of portfolio fragmentation
Portfolio fragmentation stems from geographic, temporal and economic structure.
Firstly, lockdowns in developing countries can have highly disproportionate effects. Daily wage earners for instance are thrown into possibly irreversible stress.
- A key factor is the degree of informality. In countries or sectors where the informal sector represents a high share of the population, workforce, or supply chains, traditional instruments (tax deferrals, wage subsidies) are not options to reach affected firms and households.
Secondly, many developing regions (e.g. Latin America) already lacked fiscal space going into this crisis. The risk is now a full-blown macroeconomic and financial crisis. These macroeconomic risks primarily stem from the following sources:
- A dual shock from the oil and commodity price collapse. Obvious impacts in terms of export and terms of trade shock are already rapidly transitioning into currency, budgetary and fiscal crises in resource reliant countries.
- A sudden stop and reversal in FDI, portfolio and other capital flows, especially remittances.
- Collapse in the tourism industry, which could be irreversible not only in the short but also medium to long term.
Thirdly, risks are compounded by temporal uncertainty. E.g. while the first death from COVID-19 in the G7 was in January, the first in L. America was only in March, and some countries (e.g. Ecuador) face serious outbreaks. Several developing countries are facing each of these trends simultaneously, increasing social unrest.
- Donor responses to temporal uncertainty can compound effects.g. it would be rational for donors to pull funding away from other sectors to support health infrastructure, especially if doing so is viewed as in their self-interest (prevention of re-emergence by way of imported cases).
- These temporal uncertainty risks do not even include weak health systems (an understatement given some like Central African Republic had three ventilators for a population of 5 million going into the crisis).
Development portfolio at risk
In the immediate term, donor spending may increase to support COVID-19 needs in developing countries. However, it is hard to see donors match demands. Africa’s COVID-19 emergency funding request to the G20 at $150bn equals the total volume of 2019 ODA to all regions.
Portfolios fragment because not all projects are alike in terms of their relative importance/necessity, or, viability, even pre-crisis, let alone in a constrained COVID/post-COVID economy.
- Projects with high frontline footprint, travel, local engagement, either are or will rapidly become unviable.
- Some, given their urgent nature made worse by the crisis (e.g. gender-based violence prevention), may yet need to be preserved. This requires serious risk reassessment.
Given the macro nature of the crisis, all aid spending sectors will be hit. But the nature of the impacts may be very different across sectors.
- Economically-linked sectors are likely to be hit faster and harder (e.g. microfinance, SME finance) and in some cases irrecoverably. While other areas require urgent increases even if ultimate impacts are uncertain (e.g. health systems infrastructure, urgent budgetary and macroeconomic support).
Donors should condition expectations likewise. It makes sense for private sector instruments, blended finance funds, hybrid risk instruments, DFI portfolios to take a hit.
Similarly, donors will need to urgently estimate what may be salvageable and what may not.
Fragmentation favors multilateral channels, especially for ‘free-riders’
At the onset of crises donors lean heavily on multilateral institutions, and with good reason given their size, scale and specialization.
- Multilaterals are ‘scalable’ i.e. have stretch capacity. One of the first things the IMF did in response to the crisis is develop new marketing to promote its “$1trillion in firepower”.
However, different donors position differently with reference to the multilateral development finance system.
- Donors provide resources both in the form of core contributions to multilateral institutions, and additional contributions linked to ‘earmarked (so called ‘bilateralized’) funding channeled through multilaterals.
- For e.g. in 2018 donors provided approx. $43bn in core contributions to multilaterals but channeled a further $19.4bn through
Relatively speaking, some donors, are bigger users of the multilateral system than they are contributors to the system.
- Canada is a case in point. In 2018 Canada’s relative share of core contributions to multilaterals (ex-EU) was approx. 3.5%, however, its relative share of additional ‘earmarked’ aid channeled through multilaterals was approx. 6.3%.
- In this sense mid-sized donors like Canada are typically multilateral ‘free-riders’.
Relative ‘free-riders’ lean heavily on multilaterals during crises. Going into the 2008 crisis multilaterals accounted for approx. 50% of (former) CIDA funding by implementing partner type.
- As overall budgets were being cut, by 2011-12, their share had increased to over 65%. While the share of Canadian CSOs fell from 20% to approx. 15%.
This is significant in the long-term because multilateral funding (especially statutory assessed contribution) is near impossible to cut without significant consequences – while NGO funds can be cut overnight with no consequence.
Interactions between multilateral and other channels exacerbate risks
Multilaterals’ scalability pose additional risks should they need to further tap stretched donor budgets. If pushed too far in time, or if tail risks do not subside or increase, coupled with their by-design conservative balance sheets, multilaterals may hit capacity constraints more rapidly than is assumed.
- Only 44% of the IMF’s $1trillion firepower is in the form of normal resources, anything beyond that entails drawing on other funding, and ultimately tapping shareholders (like Canada).
In the context of overall budget constraints donors will prioritize multilateral needs.
Several developing countries had very high debt levels, which prompted the recent interest payment suspension by the G20. However, suspension may need to be turned into cancellation which feeds back to multilateral balance sheets and risk measures.
Unless aid budgets increase, donors will be forced to triage between partners, putting added pressure on those partners – NGOs, CSOs – that are typically most reliant on their home donors.
- A key source of funding for many of those partners involves raising money from the public in the form of charitable donations (data are dated however the broad pattern has not changed much). In a severe recession these sources easily dry up.
Given these stresses, several types of projects prove financially, logistically and practically unviable. Consequently, several counterparties, implementing partners and networks become rapidly unviable. Further necessitating portfolio shifts.
Arguably, relative to multilaterals, NGOs (donor-based, international and especially local in-country) are closest to the frontlines of response. They are critical to types of programming where local knowledge and networks are at a premium (e.g. sensitive gender projects, humanitarian, human rights, capacity development etc.). They face the greatest relative risks.
The pressure for donors to choose their programming channels is already in place. Several multi-donor initiatives, often incubated, housed or managed by multilateral institutions (like the World Bank Group) are in the process of crafting “COVID-19 plans”.
- These plans are of varying direct relevance and are at least partially motivated by frontloading replenishment timelines to forestall funding uncertainty. An entirely rational response to risks.
- Nevertheless, this entails a form of light moral hazard that donors, especially mid-sized multilateral free riders, are behind the curve on.
There is a high probability that the COVID-19 crisis will cause a fragmentation of donor programming portfolios. The first source of fragmentation stems from structural factors that are well beyond donor control but can make programming practically unviable. A second source of fragmentation stems from donor portfolio composition and differences in risks across sectors.
Fragmentation typically favors multilateral institutions. Prolonged crises exacerbate risks and create perverse feedbacks in the interaction between multilateral vs. other aid channels. This is often to the detriment of NGOs and CSOs.
This analysis presents a framework to assess portfolio risks. It can help guide collaborative scenario-building and shared risk assessments, especially across donors and NGOs, to forestall perverse feedback loops that may represent bigger and longer-term risks to development than the crisis itself.