The "Structure-Unionism-Wage" Paradigm in Labor Economics: Resolving the Stalemate
Zappala, Gianni, Journal of Economic Issues
The "structure-conduct-performance" paradigm in industrial organization finds its counterpart in labor economics in what can be termed the "structure-unionism-wage" paradigm, which suggests that unions organizing in industries in concentrated product markets can capture some of the monopoly rents in the form of higher wage levels.(1) Research in this paradigm has, however, reached an impasse. In the United States, research over the last decade has focussed on empirical investigation of the impact, if any, that unions in concentrated industries have on wages and firm profitability [Kwoka 1983; Freeman 1983; Salinger 1984; Karier 1985; 1988; Voos and Mishel 1986!. Yet, the impact of concentration is neither universal nor immutable [Dickens and Katz 1987; Hirsch and Connolly 1987!. In Britain, research in this tradition has also led to ambiguous and contradictory results [Blanchflower 1986; Stewart 1990; Machin 1991].
It is not the aim of this paper to review this transatlantic literature.(2) Rather, I argue that the inconclusiveness of this literature has become mired in the question of the empirical rigor and adequacy of the studies, thereby setting the context of the debate as a problem of econometric estimation or inadequate data.(3) I argue that much of the ambiguity may stem in part from the conceptualization of what constitutes a product market and from the omission of the role of management in the relationship between structure and wages.
My intent is to explore the implications for labor economics of the conclusions drawn from studies done from institutional and industrial relations perspectives on the product-labor market nexus. This view is ignored at the peril of misspecifying the product-labor market relationship, and it is unlikely that further empirical studies in the old mould will resolve the issue.(4) In other words, rather than attempt to discover the "true" empirical impact of an assumed relationship [Belman and Heywood 1990; Chappell et al. 1991!, it is now apposite to return to the original "union-capture" and "monopoly wage" thesis and to explore the relationship between market structure and unions, their interaction, and the wider industrial relations processes that may give rise to the specific wage and profitability outcomes.
The Structure-unionism-wage Thesis: The Context
The union capture thesis states that unions are better able to capture or share in the rents generated by firms when those firms operate in highly concentrated industries.(5) On the whole, studies of this thesis have remained at an empirical level.(6) Two factors may explain this trend.
First, from a neoclassical perspective, there is no a priori relationship between unionism and market structure. Where a relationship is said to exist is with respect to the well-known four Marshallian rules of labor demand. In brief, these state that a firm's demand for labor becomes increasingly inelastic, (1) the more inelastic is the demand for the product (that is, less product market competition), (2) the smaller is the elasticity of substitution with other factors of production, (3) the lower labor costs are as a proportion of total costs, and finally, (4) the more inelastic is the supply of labor.
In the neoclassical world of the profit maximizing firm, shifts in the product market (which determines the price of the product) cause shifts in a firm's demand for labor. The link between the product market and the labor market is therefore a very direct one, where shifts in the Marshallian conditions may alter a union's ability to capture any economic rents. The intervening mechanism through which a union's ability to capture any rents operates, however, is indirect and through the firm's labor demand curve. For example, decreases in product market competition may enhance a union's ability by altering the wage elasticity of the demand for labor.(7)
Apart from the Marshallian effects, there is little in neoclassical theory that suggests that a union's ability to capture rents is related to market structure per se [Chappell et al. …