The effect of unionization on labor productivity: Some time-series evidence

This paper presents evidence concerning the effect of unionization on the average productivity of labor using time-series data from the private, domestic sector of the U.S. economy over the 1948–73 period. Aggregate technology is specified by a constant-returns-to-scale, Cobb-Douglas production function which incorporates union and nonunion labor and proxies for both embodied and disembodied technical change. Maximum likelihood estimates of the model indicate that union membership significantly decreased average labor productivity, holding constant the quality and mix of capital and labor and controlling for cyclical effects. I thank John Addison, Maxim Engers, Barry Hirsch, William Johnson, Duane Leigh, Roger Sherman, and Jonathan Skinner for helpful comments on an earlier version of this paper. Of course, responsibility for remaining errors is mine.