This article examines the labor market impacts of establishment shutdowns implemented in response to coronavirus disease 2019 (COVID-19). The analysis focuses on sectors directly exposed to the shutdowns and uses data from the Quarterly Census of Employment and Wages and Occupational Employment Statistics programs of the U.S. Bureau of Labor Statistics.
Notice: On April 24, 2020, this article was updated to include data on shares of employment and wages, by metropolitan and nonmetropolitan statistical areas, in the sectors most exposed to COVID-19 shutdowns. See appendix tables for more information.
To reduce the spread of coronavirus disease 2019 (COVID-19), nearly all states have issued stay-at-home orders and shut down establishments deemed nonessential. Answering the following questions is crucial to assessing the potential labor market impacts of the shutdown policy: How many jobs are in the industries that are shut down? Where are these jobs located? What wages do they pay?
We provide answers to these questions by using data from the U.S. Bureau of Labor Statistics (BLS) Quarterly Census of Employment and Wages (QCEW) and Occupational Employment Statistics (OES) programs.1 The QCEW program produces quarterly counts of employment and wages reported by employers, covering more than 95 percent of all U.S. jobs. These data are available, by industry, at the county, metropolitan statistical area, state, and national levels. The OES program produces annual employment and wage estimates for nearly 800 occupations.
Our analysis begins by identifying industries directly affected by the shutdown orders. To this end, we borrow from a recent analysis by Joseph S. Vavra, who notes that “six of the most directly exposed sectors include: Restaurants and Bars, Travel and Transportation, Entertainment (e.g., casinos and amusement parks), Personal Services (e.g., dentists, daycare providers, barbers), other sensitive Retail (e.g., department stores and car dealers), and sensitive Manufacturing (e.g., aircraft and car manufacturing).”2
To estimate the shares of total U.S. employment in these sectors, Vavra relies on data published by the BLS Current Employment Statistics program.3 Using QCEW and OES data and focusing on Vavra’s sectors, we provide finer sector breakdowns by firm size class, geography, and wage rates. We focus on data from the second quarter of 2019. This period precedes the current shutdown by roughly a year, so possible seasonality in the data is not a concern.
The QCEW employment counts represent the number of covered workers who worked during, or received pay for, the pay period including the 12th of the month.4 QCEW wage data represent the total compensation paid during the calendar quarter, regardless of when the services being paid were performed. Under most state laws and regulations, wages include bonuses, stock options, severance pay, the cash value of meals and lodging, and tips and other gratuities. We supplement this information with OES data on wages in each sector.
Table 1 presents information on employment in June 2019 and total compensation in the second quarter of 2019, focusing on the most exposed sectors identified by Vavra. Using this information, figure 1 shows the shares of employment and wages in those sectors.5 The most notable result revealed by the figure is that restaurants and bars accounted for a little more than 8 percent of total employment in June 2019. Employment in the other exposed sectors is substantial as well. In all, 20.4 percent of all workers are employed in industries most immediately affected by the COVID-19 shutdowns.
|Firm size (number of employees)||Total||All other||Most exposed sectors|
|Restaurants and bars||Travel and transportation||Entertainment||Personal services||Other sensitive retail||Sensitive manufacturing||Most exposed sectors combined|
Employment levels in June 2019 (thousands)
10 or less
11 to 50
51 to 100
101 to 500
More than 500
Total wages paid in second quarter 2019 (billions of dollars)
10 or less
11 to 50
51 to 100
101 to 500
More than 500
Note: Firms are identified by Employer Identification Number.
Source: Authors’ calculations based on U.S. Bureau of Labor Statistics Quarterly Census of Employment and Wages data for June and second quarter 2019. The North American Industry Classification System codes used to define the most exposed sectors can be found in Joseph S. Vavra, “Shutdown sectors represent large share of all U.S. employment” (Chicago, IL: Becker Friedman Institute for Economics at the University of Chicago, March 31, 2020), https://bfi.uchicago.edu/insight/blog/key-economic-facts-about-covid-19/.
Figure 1 also shows the share of total wage earnings accruing to workers in the most at-risk sectors in the second quarter of 2019. In all of these sectors, except sensitive manufacturing, the wage share is lower than the employment share, indicating that, on average, workers in the most highly exposed sectors earn less than workers employed elsewhere. On average, wage earnings in exposed sectors make up 12.2 percent of total wage earnings, which is substantially less than the employment share (20 percent).
Figure 2 shows the size distribution of firms employing workers in the most affected sectors. About 11 percent of workers in those sectors are employed by firms with 10 or fewer workers, 24 percent are employed by firms with 11–50 workers, 10 percent are employed by firms with 51–100 workers, 15 percent are employed by firms with 101–500 workers, and 41 percent are employed by firms with more than 500 workers. In the other sectors of the economy, these figures are 9 percent, 13 percent, 6 percent, 17 percent, and 55 percent, respectively.
These results show that, on average, firms in the most exposed sectors tend to be smaller than firms in all other sectors. However, a substantial number of workers in the sectors directly affected by the shutdowns are employed by large firms. While not insubstantial, wage payments to employees at small firms in the exposed sectors do not account for a very large share of total wage payments in the economy. Table 1 reveals that roughly 12 percent of total wages are in the most exposed sectors, and figure 2 shows that 54 percent of these wages are paid by small firms (firms with fewer than 500 workers). This means that about 6.6 percent of total second-quarter wage payments are made by small firms in the most exposed sectors. For wage levels, table 1 shows that $139.3 billion of the $2,119.4 billion in total second-quarter wage payments are payments made by small firms in the most exposed sectors.
As shown in figure 3, the share of employment in the most directly affected sectors varies greatly by state. Nevada (34.3 percent), Hawaii (31.1 percent), Florida (23.7 percent), and South Carolina (23.1 percent) all have more than 23 percent of their employment concentrated in the highly exposed sectors. This concentration is due to those states’ high share of employment in the travel and transportation sector, reflecting the importance of tourism to their economies. On the other hand, Midwestern agricultural states generally have less employment in the most exposed sectors. Nebraska (16.3 percent), Iowa (16.8 percent), Arkansas (17.4 percent), and Minnesota (17.4 percent) all have less than 18 percent of their employment in the highly exposed sectors. Additionally, because of the presence of the federal government, only 12.9 percent of employment in the District of Columbia is in the most exposed sectors.
We use May 2019 OES data to provide more detailed information on wages in the most exposed sectors. We obtain these estimates by first calculating the national average wage and total employment for each occupation and placing each occupation into a wage decile (so that 10 percent of total employment is in each decile).6 We then examine how occupational employment is distributed across the various affected sectors. Figure 4 shows the wage decile shares of employment in the most distressed sectors. If the wage distribution in those sectors perfectly matched the economywide wage distribution, 10 percent of employment would be in each wage decile. This is clearly not the case. Most notably, the wage distribution is considerably left skewed, with nearly 54 percent of jobs concentrated in the two lowest paid deciles and only about 13 percent of jobs located in the four highest paid deciles. This result confirms and extends the results from figure 1—while the most exposed sectors account for slightly more than 20 percent of total employment, they account for only about 12 percent of wage earnings.
Our analysis shows that a large number of jobs are in sectors directly affected by the shutdowns enacted to combat the COVID-19 pandemic. All told, about 20 percent of all employees work in the most exposed sectors identified by Vavra. The proportion of jobs in the shutdown sectors is particularly high in Nevada, Hawaii, Florida, and South Carolina, states that rely heavily on tourism. Besides affecting employment in these jobs, the shutdown orders have indirect impacts on sectors that provide inputs to the directly affected sectors.7 The longer the shutdown continues, the greater its cumulative direct and indirect effects.
We also find that occupations with lower wages are more common in the shutdown sectors than elsewhere in the economy and that higher paying jobs are less common in those sectors. Consequently, shutdown policies disproportionately affect workers in lower paying jobs.
ACKNOWLEDGMENTS: We are grateful for helpful comments from Jennifer Cassidy-Gilbert, Mike Dalton, Keenan Dworak-Fisher, Elizabeth Handwerker Weber, Julie Hatch, Joe Piacentini, Dave Piccone, Anne Polivka, Ken Robertson, Dave Talan, and Bill Wiatrowski.
Matthew Dey and Mark A. Loewenstein, "How many workers are employed in sectors directly affected by COVID-19 shutdowns, where do they work, and how much do they earn?," Monthly Labor Review, U.S. Bureau of Labor Statistics, April 2020, https://doi.org/10.21916/mlr.2020.6.
1 “Quarterly Census of Employment and Wages” (U.S. Bureau of Labor Statistics), ; and “Occupational Employment Statistics” (U.S. Bureau of Labor Statistics), .
2 Joseph S. Vavra, “Shutdown sectors represent large share of all U.S. employment” (Chicago, IL: Becker Friedman Institute for Economics at the University of Chicago, March 31, 2020), . In his communication with us, Vavra made it clear that his designation of certain industries as being highly exposed was subjective, but we found his listing quite reasonable. One could quibble about certain industries, but we are convinced that reasonable modifications to the list are likely to have relatively minor effects on our overall findings.
3 “Current Employment Statistics—CES (national)” (U.S. Bureau of Labor Statistics), .
4 Covered employees in the private sector and in state and local governments include most corporate officials, all executives, all supervisory personnel, all professionals, all clerical workers, many farmworkers, all wage earners, all piece workers, and all part-time workers. Workers on paid sick leave, paid holiday, paid vacation, and the like are also covered. Importantly, the QCEW excludes proprietors, the unincorporated self-employed, unpaid family members, and certain farm and domestic workers from having to report employment data; railroad workers covered by the railroad unemployment insurance system are also excluded. The QCEW is an establishment-based survey. We obtain firm estimates by grouping establishments with the same federal Employer Identification Number.
5 The employment shares in figure 1, which are obtained from the QCEW, are very similar to those Vavra obtains by using the CES. The QCEW is the sampling frame for the CES. Thus, as detailed in the previous note, independent contractors, freelancers, and other unincorporated self-employed workers are excluded from both sets of estimates.
6 Because occupational employment is lumpy, the employment of occupations that span the deciles is distributed so that exactly 10 percent of employment is in each decile. Our classification scheme ignores wage variation within occupations (which is much less important than variation across occupations).
7 Carlos Garriga, Juan Sanchez, and Ryan Mather, “How closing restaurants and hotels spills over to total employment,” On the Economy Blog (Federal Reserve Bank of St. Louis, April 2, 2020), .