Occupations, workplaces or jobs? An exploration of stratification contexts using administrative data

Occupations have long been held by sociologists, from the older status attainment tradition to the more recent micro-class tradition, to be at the center of stratification writ large. Occupations are specifically argued to be central to shaping wages. Indeed, this has been understood as the comparative advantage of sociology relative to economics in understanding wage setting. However, an undercurrent has for decades existed in sociology that suggests other contexts, mainly workplaces and jobs, may be as important if not more important stratification contexts. Until recently data with the capacity to simultaneously assess all three contexts has been virtually non-existent. In this paper we use administrative data from five countries (Denmark, Finland, Germany, Japan, and South Korea) to assess the relative contributions of occupations, establishments, and jobs to wages. Our core finding is that there is no universal link between occupations and wages, with occupations explaining between 30 and 56 % of wage variance across country-years. As well, in all countries except Finland establishments explain more of the variance in wages than do occupations. Jobs and establishment figure prominently in the social organization of wages, and must be included in theoretical models and whenever possible in empirical analyses of social stratification.