Estimating Workers' Marginal Willingness to Pay for Safety using Linked Employer–Employee Data

Abstract

In this study, Norwegian linked employer–employee panel data covering 1994–96 are used to estimate workers' marginal willingness to pay (MWP) for safety, thus providing unique comparisons between estimates of MWPs from hedonic wage, quit and job duration models. Hedonic wage regressions show that higher injury hazards imply higher wages. Quit and duration regressions show that higher injury hazards imply higher job exit probabilities, while higher wages reduce the job exit probabilities. Differences in the estimated MWP figures indicate that search frictions cause sizeable bias in MWP figures from hedonic wage models, and that MWP issues should be addressed from a dynamic perspective.