Wages, Skills, and Skill-Biased Technical Change: The Canonical Model Revisited

The canonical supply-demand model of the wage returns to skill has been extremely influential; however, it faced several important challenges. Several studies show that standard approach sometimes produces theoretically wrong-signed elasticities substitution, yields counterintuitive paths for skill-biased technical change (SBTC), and does not account observed deviations in college premia younger vs. older workers. This paper shows these failings can be explained by mis-measurement relative prices supplies (based on demographic composition-adjustments) inadequate ad hoc functional form assumptions about path SBTC. Improved estimates variation across cohorts within narrowly defined groups help explain deviation premium workers, even with perfect substitutability age. Re-estimating a good fit better out-of-sample prediction robustly positive substitution between high low suggest greater more modest role We implement two new approaches modelling First, we study extent which recessions induce jumps or trend-adjustments bias find evidence both features are (but differ recessions). Second, link SBTC direct measures information technology investment expenditures evolution quite well. Together, previously employed literature too crude data well, leading incorrect conclusion slowed during early-1990s under-estimates elasticity