A Note on the College-Premium Framework

Abstract

This paper demonstrates that the estimates produced by the canonical college-premium framework are nonnegligibly sensitive to the assumptions on the relative supply index, the key object in the analysis of the college premium. In particular, we argue that this framework does not offer a systematic way to account for the time-series evolution of the efficiency units of labor supply. This fragility is a source of empirical non-robustness and has important implications for the analysis of wage inequality. We show how the main predictions of the standard model change upon relaxing the assumptions on the relative supply index.