Adam Smith was the first to suggest that persons in dangerous or unpleasant jobs should be paid a high wage. His example was extreme: public executioner. Smith reasoned that despite the black hood that conceals the executioner’s identity, people would not want that job unless they were paid additional compensation above the average wage for the community. Economists have since used the term compensating wages to refer to this additional compensation. The compensating wage hypothesis holds that compensating wages are generated in competitive labor markets to equalize the net benefit (wage minus risk) workers derive from safe and dangerous jobs. If true, the hypothesis and corresponding empirical estimates of the amount of compensating wages for different levels of risk have implications for occupational safety and health policy, workers’ compensation, and statistical estimates of the value of life.
Evidence for compensating wages is undeniable in high-profile and dangerous jobs such as iron workers constructing tall buildings, airline pilots, bounty hunters, ocean fishers, and coal miners. But these jobs comprise only a fraction of all jobs, and the associated risks are obvious. It is not clear that compensating wages are generated across all dangerous jobs throughout the economy. (See Mason 1995 for a critique of compensating wage differentials.)
The hypothesis requires that workers be mobile, informed about job hazards, rational, and risk averse. Most debate about the hypothesis has to do with mobility and information. Job attachment increases with age, marriage, and children in the family, but jobs can become more or less dangerous over time, and attached workers may not change jobs. In addition, some workers, especially poor ones, may not have much mobility if the only choice is between a dangerous job or none at all. Information is also problematic. Vehicle crashes and assaults are frequently overlooked by the public as likely job hazards, yet together they are responsible for 40 percent of all fatal injuries. The frequency of vehicle crashes helps explain why relatively low-wage occupations such as gardeners, construction laborers, traveling sales workers, pizza delivery drivers, garbage collectors, and farm workers face excessively high death rates. The rate of assaults helps explain why clerks in convenience stores and fast food restaurants, as well as gas-station attendants, also face high death rates. In addition, even the strongest advocates for the hypothesis acknowledge that due to lack of information, compensating wages are unlikely to be paid for occupational diseases, yet job-related diseases cause roughly 60,000 deaths per year in the United States, compared to 5,000 injury deaths.
The empirical evidence with large data sets is mixed. The best evidence supporting the hypothesis derives from U.S. Bureau of Labor Statistics data on fatality rates across industries. But these industry data conflate blue-collar with white-collar jobs, and death rates across industries are correlated with historical interindustry wage differentials. For more than 100 years and within many developed countries, wages for blue-collar workers have been high in transportation, construction, and manufacturing, and low in services, wholesale, and retail trade, independent of job hazards. Studies suggest that when interindustry wage differentials are accounted for, evidence supporting the hypothesis evaporates.
Despite the controversy, virtually all economists agree on one point: More information regarding job hazards would help “the market” generate compensating wages. One idea would require firms to include hazard information and fatality rates on job application forms the same way food manufacturers list fat content on packaged food.
Leigh, J. Paul. 1995. Compensating Wages, Value of a Statistical Life, and Inter-Industry Differentials. Journal of Environmental Economics and Management 28: 83–97.
Mason, Patrick L. 1995. Race, Competition, and Differential Wages. Cambridge Journal of Economics 19 (4): 545–568.
Viscusi, W. Kip, and Joseph E. Aldy. 2003. The Value of a Statistical Life: A Critical Review of Market Estimates throughout the World. Journal of Risk and Uncertainty 27 (1): 5–76
J. Paul Leigh