Mitigating the economic consequences of COVID-19 - give priority to a demand-side policy

  • -
  • M. Castanheira

By: Micael Castanheira (ECARES, SBS-EM, Université Libre de Bruxelles)

The COVID-19 crisis hit hard both healthcare and economic systems. The objective of this short column is to try and identify how badly the pandemic may damage our long-term productive capacity. While it is clear that it will bring about an extremely deep recession, we argue that the effects will mainly be determined by the demand-side of the economy. This calls for aggressive demand stabilization policies.

At the onset of the COVID-19 outbreak, a typical estimate of health risks was as follows: about 10-15% of the people getting infected would need heavy medical assistance (hospitalization or close monitoring at home). Those who could not access medical care would face substantial risk of death. Flattening the curve was a necessity to reduce that risk: with access to hospital care, the estimated risk of death would be reduced to about 0.7-1.5%.

These are huge numbers: 10-15% of Belgium would have represented 1.1-1.6 million patients in need of hospitalization. To save lives, a tight lockdown was the only safe option to buy time and organize a more informed policy. Opponents were arguing that 0.7-1.5% risk of death was not much higher than a bad flu strain: why panic? The fundamental difference is the fact that a flu never sends 1.1-1.6 million people to the hospital.

Besides, among survivors, many hospitalized people would end up being incapacitated for several months, if not up to a year. This could have meant about 10-15% of the working age population in Belgium! Our economy could have collapsed, adding further economic pain and suffering. The dangers of a prolonged recession due to the crippling of our productive capacity must not be underestimated.

We now have more hindsight and, thanks to the latest report by Sciensano (May 15, 2020), we can reassess these numbers. Today, 8’959 people are reported to have lost their life to COVID-19 in Belgium, of which 4’277 died in a hospital. In total, Sciensano reports that 14’301 other people had recovered after hospitalization, and 1’862 are still hospitalized (of which 380 in intensive care). In total more than twenty thousand people needed heavy medical assistance:

Recovered/left the hospital 14’301
+ still hospitalized 1’862
+ died at the hospital 4’277
Total 20’440

The horror of these numbers imply that 1 out of 5 hospitalized patients actually died from COVID-19. The “positive” element is that 5 people will end up incapacitated for each deceased, instead of 10 or 20 as initially anticipated.

As we all learned, however, focusing exclusively on hospitals was a mistake, since many people (4’588) died in a nursing home. The exhaustive testing of nursing homes reveals that 4% of the residents got infected (that is, 5’607 people), and out of these 25% (that is, 1’369 people) display symptoms. In other words, 75% display no symptoms. Among younger and healthier people, it is probably higher.

Besides the numbers in the Sciensano report, a study by the university of Antwerp estimated that 6% of the Belgian population had been exposed to the virus by April 20. The figure today is rumoured to be 9%. That is almost 1 million people. Looking at the number of hospitalized and deceased by May 14 thus provides orders of magnitude: 20’440/990’000 = 2.1% of those infected had to go to the hospital or died, and 8’959/990’000 = <1% died, either in hospital or in a nursing home. Thus, the risk of death was exactly as initially estimated, and is about 100 times higher than for the 2019 flu epidemic of in the US, and 10 times higher than in 2018. The risk of incapacitation is instead much lower than initial estimates, even more so for young people (30% of those hospitalized were between 20 and 60 years old, 10% of the deceased were less than 64).

In the absence of a vaccine, the most optimistic estimates suggest that we can reach population immunity with two thirds of the population having been exposed to the virus. That would mean another 60’000 deaths, and 136’000 hospitalizations followed by a (slow) recovery – just for COVID-19. If we cannot control the epidemic, 95% of the population could get infected, and these figures would increase to 95’000 deaths and 216’000 hospitalized.

Insofar as we can extrapolate past data, about a third – or 72’000 people – of those hospitalized would be of working age. This is 1% of the working age population, much less than the 1.1-1.6 million initially feared. It has been identified that epidemics, just by the sheer death toll, can have long-lasting economic consequences. The above shows that this component of the “supply-side shock” is likely to be more muted than initially feared. The priorities for economic policy must thus be to smoothen the demand shock, and assist the other adjustments that are needed on the supply side.

This implies that the first priority of our economic policy must be to focus on demand stabilization, and on the prevention of bankruptcies caused by a liquidity shock (i.e. activities affected by a temporary lack of demand but that are otherwise sustainable and will be needed after the crisis). If (but only if!) we manage to provide this insurance buffer, our productive capacity per se need not be damaged as much as initially feared.

The second priority must be to reinforce they aid to sectors that will be hit for a longer period. In particular, the sectors that rely on dense interactions among the public (theatre, restaurants, public transport…) will face structurally higher costs as long as social distancing is necessary. Like with the oil shocks of the 1970s and 80s, this risks costing jobs and/or hurting wages as long as contagion risks persist. Public support will be needed, either to maintain these jobs, or to assist workers in finding other jobs, especially for the younger generations.

Last, the amounts involved will be enormous, and cannot translate into another debt crisis, even more so in the euro area. The ECB intervention so far appears entirely proportionate to the challenge in that regard. Epidemiological data shows that, with prompt and focused action, the economic recovery can be faster than what information from 2 months ago would have suggested.1 But the economic data today demonstrate that, like for health measures in March, failing to act at the economic level is not an option today.

  1. A recent paper by Correia, Luck, and Verner also shows that protecting the population stimulated the recovery after the 1918 flu pandemic: We find that cities that intervened earlier and more aggressively do not perform worse and, if anything, grow faster after the pandemic is over. 

Learning from the curve

An open source research project on COVID19 and economics. A collaboration between academics to reach out to policy makers and the general public.

Website design by Alessandro Gallina