VoxEU Column COVID-19 Development Global economy

The IMF’s growth forecasts for poor countries do not match its COVID narrative

The IMF is forecasting a substantially more muted impact of the COVID crisis on GDP for developing countries compared to advanced economies. This column argues that the discrepancy cannot be explained by external vulnerabilities, which afflict developing countries more. Nor can it be explained by the domestic shock, because social distancing and lockdowns have been similar across both groups, while fiscal policy responses have been significantly weaker in developing countries. Relative optimism should not guide international policy responses.

Diagnosing the impact of the COVID crisis on developing countries and formulating appropriate policy responses are of utmost concern (Baldwin and Weder di Mauro 2020 discuss many relevant issues; Bolton et al. 2020 make a compelling case for a debt standstill for developing countries). Recently, the IMF sounded appropriately dire about the economic impact of the raging COVID-19 pandemic. In the latest World Economic Outlook, IMF Chief Economist Gita Gopinath suggested that the economic impact of this crisis would be significantly worse than the Global Crisis, and that developing countries were especially vulnerable (IMF 2020, Gopinath 2020). The World Economic Outlook forecasts global growth to be -3% in 2020, compared with -0.2% in 2009 in the aftermath of the Global Crisis.

However, there is dissonance between this narrative and its GDP numbers. The IMF is actually forecasting a significantly more muted impact of the COVID-19 pandemic on emerging markets and developing economies (EMDEs) than advanced economies. Figure 1 plots the forecast decline between 2020 and 2019. Not only is there significant variation across countries, from a projected decline of over 12% for Ireland to just over 2% for India, but there is a pattern. GDP growth is forecast to decline by 8.4 percentage points for the advanced economies and by only 5.3 percentage points for emerging markets and developing economies, a full 3.1 percentage points lower. Indeed, the World Economic Outlook forecasts imply that the COVID crisis will be less severe in its toll on economic activity than the Global Crisis for most of the large emerging market countries.

Figure 1 Optimistic forecasts?

Divining the IMF’s methodology is probably a fool’s errand, but the text of the IMF report describes the main factors driving the forecast, which can be broadly divided into (i) a domestic shock and (ii) external vulnerabilities.

Rather than explaining the Fund’s relative optimism about emerging markets and developing economies, external vulnerabilities should make the growth outlook more not less grim for developing countries. On a range of indicators suggested by the IMF itself in the World Economic Outlook as drivers of the COVID crisis – such as vulnerability to commodity price shocks, reliance on tourism or remittances, or a history of debt crisis – emerging markets and developing economies fare equally bad or worse and should therefore be more severely impacted by the COVID crisis (Sandefur and Subramanian 2020). 

This implies that the relative optimism of the IMF forecasts must be due to the domestic impact of the health shock. Here the striking facts are these: emerging markets and developing economies’ policy response to the virus are very similar to that in advanced economies whether measured in terms of de jure policies or de facto outcomes. The former is proxied by the Oxford policy stringency index and the latter by the google mobility index. As Figure 2 shows, since the start of the crisis actions by both sets of countries have tracked each other closely. 

Furthermore, the fiscal policy response has been significantly greater by advanced economies. On average, the announced policy responses by advanced economies amount to about 7.8% of GDP while that by EMDEs has amounted to 1.5% of GDP. 

It is true that to date, COVID cases and deaths have been markedly higher in advanced economies. But as callous as it may seem, the economic impact of a pandemic is likely to have less to do with death and disease per se than the responses by economic agents and policymakers, namely social distancing and lockdown policies, and offsetting economic policy actions via fiscal and monetary stimulus.

In short, the combination of similar virus-combating policies but significantly weaker offsetting action by developing countries should imply a greater toll on economic activity and hence a more pessimistic forecast for their GDP growth.

Nor can the relative optimism for developing countries be reconciled by recourse to the health dimension of the shock. The health shock is relevant only to the extent that it signals likely private and public policy responses. But epidemiological forecasts for future deaths and cases at the time of the World Economic Outlook were, if anything, greater for developing countries. It is only current deaths that are significantly greater for advanced economies. Only if this was the proxy for policy responses can the relative optimism for developing countries be explained. But it begs the question why current deaths as opposed to actual policy actions is the better proxy for, well, policy actions.

Figure 2 Parallel lockdowns

Could there be other factors that explain this discrepancy? One possibility is that although the IMF’s World Economic Outlook was released on 14 April, the forecasts were prepared a few weeks earlier by the area departments when the data on the COVID pandemic and the policy responses presented here may not have been available. But as Figure 2 shows, the major policy choices made by many of the developing countries became public information no later than the last week of March (India announced its lockdown on 24 March and in fact, the stringent policy actions in China and Korea were taken well before March). In any event, since the actions were so dramatic and potentially impactful (as the China data available in March revealed) IMF forecasts should have reflected the actions. 

Another explanation is political. If lower growth requires higher levels of financial resources to offset the shock in developing countries, and the IMF (and World Bank whose forecasts for developing countries are 1.5 percentage points more optimistic than the IMF) feels unable and/or unwilling to galvanise those resources, it might lean toward raising the forecast in order to lower the financing requirement. We cannot speak directly to these issues but the evidence we have presented is not inconsistent with that view. 

Yet another explanation is bureaucratic. Growth forecasts are made by area departments, and even though the World Economic Outlook finally is the product of the research department, the latter’s say is limited. Area departments deal directly with client country government and have a greater incentive to internalise the latter’s desires/preferences. The question, of course, is why this should lead to differential optimism between advanced and developing economies. 

If the main finding of this analysis is that optimism on developing country growth is unwarranted, it also has a first-order implication for the global growth forecast. Suppose as an illustrative, ball-park exercise, we say that developing countries as a whole should be impacted no less than advanced economies. Suppose too that the IMF’s forecast decline for advanced economies is reasonable. Then, the developing country growth decline and the global growth decline should be close to 8 percentage points below the 2019 rate. In turn this implies that global growth for 2020 should be closer to -4.5% rather than the current forecast of -3% because developing countries account for about 55% of global GDP in purchasing-power parity terms. In other words, correcting just the relative optimism for developing countries yields a global growth forecast that should be about 50% more pessimistic than the current one. 

Going forward, we wonder whether the IMF will change its narrative or its numbers. We fervently hope that the current numbers do not lead to complacency on the part of individual countries, eliciting a policy response that is inadequate to the COVID pandemic. We also fervently hope that these numbers do not legitimise an ungenerous, conditionality-addled international response to helping developing countries in the face of an unprecedented calamity. 


Baldwin, R and B Weder di Mauro (2020), Economics in the Time of COVID-19, a eBook, CEPR Press.

Bolton, P, L Buchheit, P-O Gourinchas, M Gulati, C-T Hsieh, U Panizza and B Weder di Mauro (2020), “Born Out of Necessity: A Debt Standstill for COVID-19”, CEPR Policy Insight no 103.

Gopinath, G (2020), “The great lockdown: Worst economic downturn since the great depression”, IMF Blog, 14 April. 

Hale, T, S Webster, A Petherick, T Phillips and B Kira (2020), “Oxford COVID-19 government response tracker”, Oxford University, Blavatnik School of Government.

International Monetary Fund (2020), World Economic Outlook, April 2020.

Sandefur, J and A Subramanian (2020), “The IMF’s growth forecasts for poor countries don’t match its COVID narrative”, Center for Global Development, Working Paper No. 533.

210 Reads