We address long-standing concerns in the literature on compensating wage differentials: the econometric properties of the estimated value of statistical life (VSL) and the wide range of such estimates. We confront prominent econometric issues using panel data, a more accurate fatality risk measure, and systematic application of panel data estimators. Controlling for measurement error, endogeneity, latent individual heterogeneity possibly correlated with regressors, state dependence, and sample composition yields VSL estimates of $4 million to $10 million. The comparatively narrow range clarifies the cost-effectiveness of regulatory decisions. Most important econometrically is controlling for latent heterogeneity; less important is how one does it.

To resolve the theoretical ambiguity in the effect of age on the value of statistical life (VSL), this article uses a novel, age-dependent fatal risk measure to estimate age-specific hedonic wage regressions. VSL exhibits an inverted-U-shaped relationship with age. In the year 2000 cross section, workers' VSL rises from $3.7 million (ages 18–24) to $9.7 million (35–44), and declines to $3.4 million (55–62). Controlling for birth-year cohort effects in a minimum distance estimator yields a peak VSL of $7.8 million at age 46, and flattens the age-VSL relationship. The value of statistical life-year also follows an inverted-U shape with age.

Using a large data set, the authors find that smokers select riskier jobs, but receive lower total wage compensation for risk than do nonsmokers. This finding is inconsistent with conventional models of compensating differentials. The authors develop a model in which worker risk preferences and job safety performance lead to smokers facing a flatter market offer curve than nonsmokers. The empirical results support the theoretical model. Smokers are injured more often controlling for their job's objective risk and are paid less for these risks of injury. Smokers and nonsmokers, in effect, are segmented labor market groups with different preferences and different market offer curves.

This paper incorporates a Bayesian learning model into a hedonic framework to estimate the value that residents place on avoiding cancer risks from hazardous-waste sites. We show that residents are willing to pay to avoid cancer risks from Superfund sites before the U.S. Environmental Protection Agency (EPA) releases its assessment (known as the Remedial Investigation) of the site. Residents' willingness to pay to avoid risks actually decreases after the release of the Remedial Investigation, suggesting that the information lowers the perceived levels of risk. This estimated willingness to pay implies a statistical value of cancer similar to the value-of-life estimates in labor market studies.

Using survey data on consumer product purchases, this paper introduces an approach to estimate jointly individual utility functions and risk perceptions implied by their decisions. The behavioral risk beliefs reflected in consumers' risky decisions differ from the stated probabilities given to them in the survey. These results are not consistent with a Bayesian learning model in which the information respondents utilize is restricted to what the survey presents. The results are, however, potentially consistent with models in which prior risk information is influential or models in which people do not act in a fully rational manner.