Testing the Computer-use Wage Premium Hypothesis: Evidence from CPS Surveys

This research investigates whether and to what extent the use of computers at work affects employee
wage premium. We construct a panel dataset from the 2000 and 2001 US Current Population
Surveys (CPS) and apply fixed effects models to estimate the computer use wage premium. While
cross-sectional analysis indicates a wage premium of about 14%, this premium drops to 3.4% after
controlling for employee and employer heterogeneity within the panel analysis. In addition, our results
show that the computer-use wage premium is highest for women employees who started to use
computers and the Internet in the second year of the panel. The lower wage premium found in the
panel analysis confirms the existence of omitted variable biases in cross-sectional analysis, but at the
same time provides support that the estimated computer-use wage premium identified in prior studies
cannot be purely attributed to unobserved employee heterogeneity.