Publication Date



To what extent do workers work more hours than they are paid for? The relationship between hours worked and hours paid, and the conditions under which employers can demand more hours “off the clock,” is not well understood. The answer to this question impacts worker welfare, as well as wage and hour regulation. In addition, work off the clock has important implications for the measurement and cyclical movement of productivity and wages. In this paper, I construct a unique administrative dataset of hours paid by employers linked to a survey of workers on their reported hours worked to measure work off the clock. Using cross-sectional variation in local labor markets, I find only a small cyclical component to work off the clock. The results point to labor hoarding rather than efficiency wage theory, indicating work off the clock cannot explain the counter-cyclical movement of productivity. I find workers employed by small firms, and in industries with a high rate of wage and hour violations are associated with larger differences in hours worked than hours paid. These findings suggest the importance of tracking hours of work for enforcement of labor regulations. *


Any opinions and conclusions expressed herein are those of the author and do not necessarily represent the views of the U.S. Census Bureau. All results have been reviewed to ensure that no confidential information is disclosed. This paper has benefited greatly from comments and suggestions of John Abowd, Richard Mansfield, Karel Mertens, Lars Vilhuber, Kathryn Edwards, Hautahi Kingi, Mark Kutzbach, Sabrina Pabilonia (discussant), Jim Spletzer, as well as seminar participants of the Labor Dynamics Institute, Cornell Labor Work in Progress Seminar, Census CES Brown Bag Seminar, Census-BLS conference, and Cornell Labor Seminar. I acknowledge the use of the Cornell ECCO computing cluster. All errors are my own. E-mail: