Nature of Work and Distribution of Risk: Evidence from Occupational Sorting, Skills, and Tasks

Abstract

How does the nature of work—teleworkability and contact intensity—shape the distribution of health, earnings, and unemployment risks, created by the COVID-19 pandemic? To answer this question, I consider two contexts. First, I show that the existing patterns of spousal occupational sorting in the United States matter for the distribution of these risks. In particular, I show that sorting into occupations with similar contact intensity in the workplace mitigates the risk of intra-household contagion relative to the situation where spouses match at random in terms of occupations (zero sorting). Furthermore, I show that sorting into occupations with similar teleworkability exacerbates the exposure to earnings and unemployment risks relative to the case of zero sorting. Second, I document that teleworkable occupations more likely require higher education and experience levels as well as greater cognitive, social, character, and computer skills, compared to non-teleworkable occupations. This difference in skill requirements affects earnings and unemployment risks by increasing the likelihood of skill mismatch for the newly unemployed. My results imply that the current economic downturn may have long-run effects on employment prospects and earnings of workers who had non-teleworkable or high-contact-intensity jobs at the onset of the COVID-19 outbreak. I discuss the relevant policy implications and associated policy constraints that follow from my findings.