Insider-outsider influences on industry wages

Abstract

The persistence of high rates of unemployment in Europe has focused renewed attention on the wage determination process. One aspect which has received particular attention recently is the extent to which wages are determined by insiders, with outsiders having little impact on the wage bargain. If outsiders are disenfranchised in the wage determination process, they may be unable to underbid themselves into employment. The insider-outsider model, and the associated implication of hysteresis, thus provides a possible explanation for persistent involuntary un-employment.

Most empirical work on the importance of the insider-outsider model and hysteresis has been done at the aggregate level, either by testing whether wage growth is related to the level or the change in unemployment, or by testing if employment or unemployment follows a random walk. This paper extends both of these lines of analysis to the industry level. Based on a specification which nests alternative models of the wage determination process, estimation results are presented for fifteen industries using annual data for fourteen industrialized countries, and for pooled time-series cross-country data. The paper also tests the hypothesis that only industry-specific variables are important to the determination of industry wages and considers whether differences in industry wage determination might help to understand cross-country differences in unemployment.

The results offer considerable support to the insider view of wage determination. Industries in which wages seem to be determined by a natural rate/Phillips curve model — the primary competing model considered in the paper — appear to be the exception rather than the rule. Moreover, at the industry level hysteresis does not appear to characterise wage formation in the high-unemployment European countries to a greater extent than in North America and Japan.