Abstract:
The phenomenon that married men earn a higher wage on average than unmarried men, the so-called marriage premium, is rather well established. However, the robustness of the marriage premium across the wage distribution and the underlying cause of the marriage premium are not well known. Focusing on the entire wage distribution and employing recently developed nonparametric tests for stochastic dominance, our findings question the current conception of the marriage premium, calling instead for the introduction of a broader concept incorporating wage dispersion. This broader notion arises from evidence suggesting that the marriage premium is primarily confined to the lower tail of the wage distribution; the premium is negligible at best in the upper tail. Finally, the majority of the premium is explained by selection, but there is a small role for ‘causal’ explanations.Length: 54 pages