Social Insurance in Market Contexts: Implications of the Structure of Workers’ Compensation for Job Safety and Wages

Abstract

Social insurance programs whose costs are tied to particular behaviors represent more than simple income transfers between members of the economy. In the case of job safety insurance programs such as workers’ compensation, in which the costs of the program are tied to the firm’s safety records, the market incentives for safety that are created by the insurance program can be quite strong. At the same time, when those programs benefit injured workers, they simultaneously provide disincentives for safety and incentives for workers to extend periods of recovery and to file more claims.

We analyze these issues using a large data set on worker wages and characteristics, coupled with information on fatality risks and workers’ compensation benefits. We find that workers’ compensation insurance provides incentives for safety to firms that outweigh the moral hazard effects. We further find that the insurance provided to workers on unsafe jobs reduces the net compensation paid to these workers. In particular, workers’ compensation benefits create a negative compensating differential that offsets the positive compensating differential that must be paid for exposure to job risks. This negative wage differential leads to a reduction in the wage bill that compensates employers for the cost of workers’ compensation premiums. Reductions in fatality risks, and the resultant reduction in the wage, supplement the direct wage-benefit effect.

An early version of this paper was presented at Northwestern University and at Duke University. Comments provided by seminar participants, and also the comments provided by Georges Dionne, John Ruser, Gary Zarkin, and two anonymous referees are gratefully acknowledged. Viscusi’s research was supported by the endowment of the George G. Allen chair, and Moore’s research was supported in part by the Business Associates Fund at the Fuqua School of Business, Duke University.