Centralization of Bargaining and Wage Inequality: A Correction of Wallerstein

Wage inequality is a subject of intense political and substantive concern, and differencesin the extent of wage inequality between the Anglo-American economies and those of con-tinental Europe animate public and scholarly discussion. The wide earnings differentialsobserved in the United States and the United Kingdom are commonly justified as the properand necessary incentives for a mature, growth-oriented economy; the narrower income dif-ferentials encountered in continental Europe are rationalized as necessary for the protectionof the less-well paid and the integrity of the social fabric. The upward secular trend in wageinequality in the US since the 1970s, which is a stark reversal of the pattern observed duringthe preceding 40 years, is also a topic attracting wide public and scholarly attention.Despite the importance of the subject and the amount of attention that it has received,the extensive literature on the economic impact of wage bargaining institutions in advanceddemocratic societies contains only a few widely accepted findings. One of these is that morecentralized bargaining institutions strongly encourage wage equality. No less an authoritativeorgan than the OECD notes, for instance, that “there is a fairly robust relation betweencross-country differences in earnings inequality and bargaining structures” (OECD 1997,p. 64). Several well-designed country-specific studies support the claim that the system ofcollective bargaining is a major determinant of variations in the extent of wage inequality(Hibbs and Locking, 1996, 2000, Erickson and Ichino, 1995). Yet even the most careful andwell executed country level results may be affected by idiosyncratic country-specific factors.Evidence at the cross-national level is required. The most careful and convincing cross-national empirical documentation of the link between centralized bargaining institutionsand wage inequality appears in Wallerstein (1999), published in this journal