The Returns to Computer Use Revisited, Again

Abstract

Using North American data, we revisit the question first broached by Krueger (1993) and reexamined by DiNardo and Pischke (1997) of whether there exists a real wage differential associated with computer use. Employing a mixed effects model to correct for both worker and workplace unobserved heterogeneity using matched employer-employee panel data, we find that computer users enjoy an almost 4 percent wage premium over non-users. Failure to correct for the worker selection effect leads to a more than twofold overestimate of this premium, as does failure to correct for workplace unobserved heterogeneity.

Keywords: wage determination, computers, mixed models, linked employer-employee data

JEL Classification: J30, J31, O30