IZA DP No. 9036: Average Wage Gaps and Oaxaca–Blinder Decompositions
In this paper I develop a new version of the Oaxaca–Blinder decomposition whose unexplained component recovers a parameter which I refer to as the average wage gap. Under a particular conditional independence assumption, this estimand is equivalent to the average treatment effect (ATE). I also provide treatment-effects reinterpretations of the Reimers, Cotton, and Fortin decompositions as well as estimate average wage gaps, average wage gains for men, and average wage losses for women in the United Kingdom. Conditional wage gaps increase across the wage distribution and therefore, on average, male gains are larger than female losses.