On Friday, June 26, 2020, Erik Hurst joined the Princeton Bendheim Center for Finance to discuss a new paper on labor market conditions during COVID-19.
Hurst is the Frank P. And Marianne R. Diassi Distinguished Service Professor of Economics at the Booth School of Business.
By using ADP data, the authors are able to study employer/employee links. The researchers use ADP data to look at a range of labor market trends. One benefit of the ADP data is that it includes linked employee/employer data that helps researchers understand where workers go and if they come back. Another is scale. The data is comparable to Bureau of Labor Statistics (BLS) data and covers 1/6 of the U.S. workforce.
The U.S. lost about 20% of employment through mid-April. This mirrors analysis using BLS data. There were more job losses among small businesses. 37% of workers in the bottom fifth of the wage distribution lost their jobs through late April, compared with 9% of workers in the top fifth. See chart of active and paid employment since February.
Employment declines were bigger among women than men, which is different than in typical recessions. Employer or industry characteristics explain very little of this. See chart of employment declines by sex.
About two-thirds of employment losses were from “continuing firms” that continued paying some workers. Hurst notes there was a lot of heterogeneity in job gains and losses among continuing firms, with some firms gaining employees. See charts of employment in continuing firms and distribution of employment gains.
The average firm that re-opened is at 40% of its February size. Most of the workers that came back are recalled workers previously employed with the firm. The average continuing firm, however, is at 95% of its February size. Employment growth within continuing firms is also largely recalled workers. See chart of firm size or re-opening firms.There’s been a lot of talk of rising wages, but the authors find this is entirely due to selection effects. The bottom of the distribution left the labor market. When the authors adjust for that, there is no increase in wages. Overall, employers are cutting wages of about 11% of workers.