What would be the counterfactual wage of civil servants if they were employed in the private sector? Using the French European Household panel, we present a new approach to the wage differential between the public and the private sectors. We estimate a model, which controls both for selection into employment, and for self-selection into the public sector. We also introduce unobserved heterogeneity in the propensity to be employed in either job sector, and in the sector-specific productivity. Evidence based on the counterfactual distributions suggests a large public-private wage premium for low public wages. This conclusion also holds for women but may be explained by a weaker discrimination in the public sector. Unlike women, most male civil servants would earn more in the private sector.