This paper investigates a new way to estimate the gender wage gap with the introduction of individual risk attitudes using representative Korean data. We estimate the wage gap with correction for the selection bias, which latter results in the overestimation of this wage gap. Female workers are more risk averse. They hence prefer working in the public sector, where wages are generally lower than in the private sector. It goes on to explain the reduced gender wage gap by developing an appropriate sample-selection model, with wage decompositions corrected for selectivity. Self-selection based on risk attitudes therefore explains, in part, what is popularly perceived as gender discrimination.
