Abstract

This article focuses on the issue of union wage determination in the U.S. For years labor economics was synonymous with the study of union wage determination but by the 1960s, the topic had been eclipsed by other components of labor economics, such as labor force participation, human capital, and employment and training policies. It has been the practice of labor economists to view the labor union as an interesting form of monopoly. In fact the union/nonunion wage differential has exhibited a tendency to widen over a relatively long period of time. The widening has not been evident in every period and alternative data sources produce slightly different results but it has been especially marked in recent years. Starting in the 1930s union membership grew rapidly due to a variety of economic and political circumstances. By the mid-1950s the proportion of union workers in the overall labor force peaked and began to decline. In the long run employers may be able to make certain adaptations to resist union wage pressures. They may substitute capital for labor, automate, subcontract, or move south. The union may detect these practices and press for specific contract language to prevent them. To the extent that employers can make slow adaptations to premium wage levels, the incentive for employer resistance to union wage demands is weakened. While employer resistance may serve as an adequate proxy for the wage-employment trade-off in the short run, it may be an inadequate proxy over the long haul. In the context of costly contracting, there is no law that says that the union/nonunion wage differential could not widen over many years. Such widening is consistent with the model of union wage behavior in which rational behavior is occurring in the short run, but in which the long-term consequences could be counter to union membership goals.