Dualism in the German labor market? A nonparametric analysis with panel data

Orthodoxy in economics gives pride of place to the hypothesis of compensating differentials. Applied to job quality criteria, such as wage levels, job stability, and wage growth, the compensating differentials hypothesis implies that negative or positive job quality characteristics should-ceteris paribus-rarely coincide. Originating in the late 1960s from studies of American inner-city labor markets, dualist labor market theory has raised doubts about this assumption. At its core, dualist analysis proposes that a dualism exists between a primary labor market where "jobs possess several of the following traits: high wages, good working conditions, employment stability and job security, equity and due process in the administration of work, and chances for advancement" and a secondary market where jobs "tend to involve low wages, poor working conditions, considerable variability in employment, and little opportunity to advance" (Doeringer and Piore 1971). In the 1980s, there were several attempts to apply dual or segmented labor market theory to European economies, including the German labor market. The mixed findings of these analyses have cast doubt upon the transferability of dualist theory to these contexts. In the 1990s, as unemployment figures increased and labor market deregulation initiatives were launched, there has been renewed interest in the possibility of a segmentation of the German labor market. Whereas most studies have relied on qualitative evidence, this paper utilizes data from the first five waves of the German Socioeconomic Panel(1984 to 1989) in order to investigate whether "good" or "bad" job characteristics tend to cumulate or to compensate each ether. Using a nonparametric procedure that evaluates the simultaneous occurrence of case-characteristics, we show that evidence for compensating differentials in job quality characteristics for the aggregate labor market is weak. There are, however, some indications of a cumulation of "bad" and "good" job characteristics in the non-market sector (i.e., public, non-profit, and semi-public employees). We interpret the inconsistency of our results with the hypothesis of compensating differentials, as well as with dualist assumptions, to show the need for a differentiated assessment of job quality patterns in European labor markets.