Are longer hours reducing productivity in manufacturing?

This paper provides statistical evidence of the effects of overtime hours on worker productivity using aggregate panel data for 18 manufacturing industries within the US economy. An economic production function model is specified and estimated using data for the years 1956‐1991 provided by the US Department of Labor, Bureau of Labor Statistics, the US Department of Commerce, and the Federal Reserve Board. Standard approaches are applied to specify and estimate a factor‐augmented production function model, with possible effects of overtime on productivity incorporated through the specification of factor effort functions. The empirical results suggest that use of overtime hours lowers average productivity, measured as output per worker hour, for almost all of the industries included in the sample. These results hold up under several alternative specifications and estimation techniques, including controls or corrections for autocorrelation, heteroskedasticity, rates of capacity utilization, and possible endogeneity of the constructed variable representing use of overtime hours.