WAGE DIFFERENTIALS, EMPLOYER SIZE AND UNEMPLOYMENT

Wage Differentials, Employer Size and Unemployment

Wage Differentials, Employer Size and Unemployment

Introduction

The purpose of this research paper is to analyze “Wage Differentials, Employer Size, and Unemployment” by Burdett and Mortensen, (1998). The presence of matching frictions in the form of lags in the arrival of information about the availability and terms of job offers unifies the labor market models analyzed in this study. Each model presented is related to a perfectly competitive counterpart in a natural way in the sense that its solution converges to the competitive equilibrium as frictions vanish. However, the characteristics of equilibrium when frictions are significant suggest novel theoretical insights and new empirical predictions:

First, important phenomena that cannot be explained by traditional equilibrium market analysis, such as unemployment and wage dispersion, naturally arise as characteristics of equilibrium in these models. Second, monotone cross-employer associations of the wage offered, the provision of non-wage attributes, and turnover with size of labor force are predicted. Third, the framework's implications for policy intervention, specifically the imposition of a minimum wage, are strikingly different from those of traditional competitive market analysis. (Simon 2005)

Empirical research has documented that both inter-industry and crossemployer wage differentials exist, are stable, and cannot be fully explained by observable differences in worker ability or job characteristics that might require wage. The current controversy over why workers of apparently equal ability are paid differently on similar jobs has two sides. On one, proponents argue that workers sort on non-observable ability in ways that explain the data without contradicting first principles of competitive market analysis, e.g. Murphy and Topel [1987]. On the other, adherents appeal to alternative wage determination theories, 'efficiency' and 'fair' wage arguments seem to be in particular vogue, e.g. Krueger and Summers [1987a].

In contrast, we argue that persistent wage differentials are consistent with atomistic wage formation and employment determination, at least in the nonunion sector, once the effects of matching frictions are appropriately recognized. Furthermore, the argument also provides a direct explanation for the mysterious fact that large positive wage differentials are associated with employer size, especially in the non-union sector. (See Brown and Medoff [1989] for a review of the evidence.)

Discussion

The framework used is based on Mortensen's [1988] generalization of Diamond's [1971] and Albrecht and Axell's [1984] equilibrium formulations of wage and employment determination as a non-cooperative wage-search and wageposting game played by self-interested workers and employers. Relative to these earlier papers, the model's principal innovation is that workers

obtain and respond to information about alternative job offers while employed. Although the equilibrium wage and employment levels limit to their perfectly competitive counterparts as frictions vanish in the sense that offer arrival rates become instantaneous, the model has numerous empirical implications that are very different from those of the standard competitive equilibrium when frictions are non-trivial. (Miller 2006)

We begin by analyzing a simple equilibrium labor market model with matching frictions where all workers are equally productive on all jobs and have the same opportunity cost of ...