This paper adds to the empirical literature of health as a potential endogenous explanatory variable in a wage equation. In particular, we use the estimation method proposed by Semykina and Wooldridge (2010) to deal with unobserved heterogeneity, nonrandom sample selection and measurement error (in the health variable) in one comprehensive framework. We use individual-level panel data from the European Union Statistics on Income and Living Conditions (EU-SILC) to explore the differences in the impact of health on wages across European country groups, for both men and women. The health variable is, in general, found to suffer from measurement error, and a number of tests provide evidence of selectivity bias in wages. Our preliminary results suggest that only men from Nordic and Transitional countries who are in relatively better health (measured by a one-unit increase in a health index) have, on average, a 13 and 8 percent higher hourly wage rate. For men from Continental and Mediterranean countries, and for women, the positive impact of health on wages largely disappears once we account for unobserved heterogeneity and sample selection bias. So far, and in line with the literature, our findings confirm that the relationship between wages and health is ambiguous.

Keywords: Health, wages, EU-SILC.