How will the COVID-19 pandemic impact the US economy? Several researchers have provided estimates of how workers are likely to be adversely affected by considering the extent to which people can work from home (Dingel and Neiman 2020, Hicks 2020, Koren and Peto 2020, Barrot et al 2020). But for workers in essential industries, this distinction is irrelevant: despite being unable to work remotely, nurses and ambulance-drivers are still required to perform their job. And working in an essential industry or being able to work from home still may not mean that one’s job is secure. Take public transportation. Even though it is an essential industry, many bus drivers are no longer required as consumers refrain from travelling. Hence, not just supply but also demand-side factors are important when assessing the impact of COVID-19. As people aim to avoid the risk of infection, changing patterns of behaviour and consumption are impacting industries and the people they employ in different ways.
In del Rio-Chanona et al. (2020), we aim to provide analytical clarity about the first-order supply-shocks caused by public health measures (e.g. social distancing) and first-order demand shocks (e.g. preference changes as people aim to avoid infection) by industry and occupation. By ‘first-order’ we mean ‘direct’ or ‘immediate’ effects hitting the US economy in the near term. Such effects are not to be confused with ‘overall’ impacts, which would include additional follow-on effects such as further demand reductions as workers’ incomes decline, further supply reductions as potential shortages propagate through supply chains, and cascades of firm defaults (Baldwin and Weder di Mauro 2020). Calculating these second-order effects would requires an economic model. Our focus here is to provide estimates of first-order shocks that such a model can use as inputs.
Predicting supply and demand shocks
To analyse the supply shocks associated with COVID-19, we consider two factors impacting workers’ ability to perform their job: (i) their ability to perform their activities at home, which we estimate by assessing specific work activities each occupation performs, and (ii) their likelihood of being employed in an essential industry1 (Figure 1).
Figure 1 Supply shock factors by occupation
Occupations on the lower left side of the figure (such as dishwashers and rock splitters) are less able to perform their job at home and are also less likely to be employed in an essential industry. Workers in these occupations are consequently particularly vulnerable to supply shocks associated with social distancing measures. Based on employment data from the Bureau of Labour Statistics, we estimate that around 21% of those currently employed in the US economy are likely to be in this category. In contrast, occupations on the upper right side of the figure (such as credit analysts and editors) are more likely to be employed in an essential industry and are also more able to perform their job at home. Those occupations are less vulnerable to supply-side shocks. However, they could still face employment risks due to demand-side effects.
Our demand-shock analysis is based on estimates from the US Congressional Budget Office (CBO 2006), which attempted to predict the potential impact of an influenza pandemic. While these estimates are admittedly rough and based on Hong Kong’s SARS experience, the associated predictions of demand-side reductions in sectors such as transportation, restaurants, accommodation, arts and recreation are highly relevant to the present COVID-19 situation.
Comparing demand and supply shocks across industries and occupations
Different sectors may be differentially hit by supply and demand shocks (Figure 2). To determine which shock is likely to be more relevant for the change in output, consider the following thought experiment: Suppose that due to social-distancing measures, an industry is capable of producing only 70% of its pre-crisis output. But if consumers on top reduce their demand by 90%, the industry will produce only what will be bought, that is, 10%. If instead consumers reduce their demand by 20% only, the industry will not be able to satisfy demand but will produce everything it can, that is, 70%.
As shown in Figure 2, industries such as transport, telecommunications, government, food and healthcare are considered essential and are consequently predicted to experience no supply shock. Yet for the transport sector, dramatically declining demand is likely to be the more relevant factor. And indeed, while most trains and buses are currently running and ticket prices remain mostly unchanged, they are almost empty. In contrast, for non-essential manufacturing sectors such as computer equipment manufacturing, supply shocks are more likely to dominate, although there is much uncertainty on whether purchases of durable goods will simply be delayed or cancelled altogether. Non-essential industries such as entertainment, restaurants and hotels are likely to experience adverse shocks on both the supply and demand side. Many of these industries involve activities that can’t be performed at home, and even if they could, consumers seeking to avoid infection are unlikely to demand much of these services in any case.
Figure 2 Supply and demand shocks for industries
We can also compare supply and demand shocks hitting occupations (Figure 3). While occupations in the top left corner (such as lodging managers and hotel desk clerks) face relatively mild supply shocks, their exposure to demand-side shocks driven by people refraining from travel and tourism activities is likely to be much larger – and consequently the more relevant factor in determining job vulnerability. Supply shocks, on the other hand, are likely to be the more relevant factor in determining job vulnerability for occupations in the bottom right corner (such as rock splitters and stonemasons). Even if there was sufficient demand, difficulties in performing these tasks at home is likely to constrain the output of those workers.
Figure 3 Supply and demand shocks for occupations
Compared to the pre-COVID-19 period, we predict that supply and demand shocks will cause a reduction of around 22% of US GDP and a total wage income loss of 17%. We further estimate that 24% of jobs are likely to be vulnerable. Until now, the US has lost about 30 million workers (Rushe 2020), corresponding to a loss of about 21% (note that at the time we made those predictions, job loss was only about 6 million).
Consistent with other studies (e.g. Adams-Prassl et al. 2020) our analysis suggests that these shocks are likely to hit the poorest workers the hardest (see Figure 4). Strikingly, there are no high-wage occupations (except perhaps airline pilots) that are projected to suffer adverse effects arising from the pandemic. Breaking down shocks by wage quartile indicates that the bottom 25% of earners could face employment reductions of 42%, while the top 25% are only expected to face a 7% decrease. Moreover, low-wage occupations that are unlikely to be economically vulnerable (such as janitors, cleaners and personal care aids) appear to be more physically vulnerable by becoming infected with the virus.
Figure 4 Labor shocks vs median wage for different occupations
For policymakers there are three key implications from this analysis. First, the magnitude of the shock is very large, with around a quarter of the economy not functioning. Bearing in mind the difference between first-order shocks versus overall impacts, and the key role that the duration of the lockdown will play, the potential direct effects are projected to be a multiple of what was experienced during the Global Crisis (where employment dropped by 3.28 percentage points) and comparable only to the Great Depression (where employment fell by 21.7%).
Second, the largest shocks are on the supply side. Strategies which enable people to return to work as quickly as possible without endangering public health must therefore be a priority. Virus mitigation and containment are essential first steps, but strategies such as widespread antibody testing to identify people who are safe to return to work, and rapid testing, tracing, and isolation to minimize future lockdowns, will also be vital until a vaccine is available. Furthermore, aggressive fiscal and monetary policies to minimize these first-order shocks cascading into second-order shocks are essential, in particular policies to keep workers in employment and maintain incomes (e.g. ‘paycheck protection’ schemes as announced by several countries), as well as policies to preserve business and financial solvency.
Third, and finally, the inequalities that we (and others) have highlighted will require targeted policy responses. In order to ensure that burdens from the crisis are shared as fairly as possible, assistance should be targeted at those most effected, while taxes to support such programs should be drawn primarily from those least effected.
Adams-Prassl, A, T Boneva, M Golin and C Rauh (2020), “The large and unequal impact of covid-19 on work”, VoxEU.org, 8 April.
Baldwin, R and B Weder di Mauro (2020), “Introduction” in Economics in the Time of COVID-19, CEPR, chapter 1 pp 1–30.
Barrot, J-N, B Grassi and J Sauvagnat (2020), “Sectoral effects of social distancing”, Covid Economics, 3: 85-102.
Congressional Budget Office (2006), “Potential influenza pandemic: Possible macroeconomic effects and policy issues”, 27 July.
del Rio-Chanona, R M, P Mealy, A Pichler, F Lafond and J D Farmer (2020), “Supply and demand shocks in the COVID-19 pandemic: An industry and occupation perspective”, Covid Economics, 6: 65-104.
Dingel, J and B Neiman (2020), “How many jobs can be done at home?”, Covid Economics, 1: 16-25.
Koren, M and R Peto (2020), “Business disruptions from social distancing”, Covid Economics, 2: 13-32.
Hicks, M J, D Faulk, S Devaraj (2020), “Occupational exposure to social distancing: A preliminary analysis using O’NET data”, Becker Friedman Institute white paper, 13 March.
Rushe, D (2020), “Another 3.8 million Americans lose jobs as US unemployment continues to grow”, The Guardian, April 30.
1 Our analysis of essential industries is based on the list published by the Italian government.