Can Wage Increases Pay For Themselves? Tests with a Production Function

Efficiency wage theories predict that companies that pay high wages will have higher productivity from high work effort, low turnover, and other efficiency-enhancing effects. This paper uses the PIMS line-of-business data set to test whether high wage businesses are more productive. Relative wages were measured by managers' assessments of their relative compensation, holding worker quality constant. A positive relation was found between changes in relative wages and changes in total factor productivity. The elasticity of output with respect to wages was of the magnitude predicted by efficiency wage theories. Also consistent with efficiency wage theories, the relationship between changes in wages and changes in productivity was weaker at businesses with high unionization. Results from two additional data sets imply that high wages are not merely measuring high human capital.