This study investigates the relationship between wages and the risk of work-related death for male blue-collar workers. The size of the premium for risk and its statistical significance depend heavily on the inclusion of industry dummy variables into the regression. Irrespective of the type of risk variable used, controlling for industry at a finer breakdown lowers the price of risk and its statistical significance. Estimates of the value of a statistical life (VSL) proved to be more robust when an aggregated risk measure at three-digit occupational level was used. In this case, the VSL varied from 0.79 million USD (for the model with industry dummy variables at the three-digit level) to 2.41 million USD (for the model without industry dummy variables). To the best of the author’s knowledge, this is the first study estimating VSL using the compensating wage differential approach for countries in Central-Eastern Europe.
Compensating wage differentialJob risksValue of statistical life