Worker and Employer Heterogeneity in Hours Determination

Alex Mas, Princeton University
Wednesday, April 14, 2021, 1 - 2:30pm

Joint with Marta Lachowska, Raffaele Saggio, and Steve Woodbury.

This paper uses newly available administrative data to answer the question: whose preferences are revealed in work hours? Hours (and earnings) exhibit a high degree of variability both within job spells and in the cross-section of workers. Using high-quality data we show that the labor supply responses to wage changes explain only a very small share of the variance of hours changes. Firm-level shocks explain a larger share of the variance of hour changes, but most of the variability is idiosyncratic. We apply a two-way (“AKM”) fixed-effects approach and a new random coefficients method to decompose the within-job variance in hours into a portable component of variability that workers carry with them between jobs and a risk component that is specific to employers, reflecting firm demands. We apply a similar approach to decomposing the level of hours and provide a number of auxiliary analyses to characterize the degree of hours constraints in worker hours.