Comment

Abstract

The paper by Guvenen, Kaplan, and Song provides many useful insights into the evolution of top earners’ earnings and the dynamics of the gender gap at the top of the earning distribution. The authors have access to a representative sample of 10% of individual earnings from the US Social Security Administration (SSA) between 1981 and 2012. The panel structure of the data allows them to follow individuals over time, shedding light on the persistence of top earners and addressing some of the shortcomings of the literature that uses cross-sectional data. Theyfind several interesting results. First, in 2012, the share ofwomen among the top 0.1% of earners is only 11% and among the top 1% only 18%. This gap was even larger earlier in the sample, suggesting a slow positive trend toward the reduction of the gender gap, with themajority of the gain concentrated in the 1980s and the 1990s. Second, the paper unveils a very large turnover over the life cycle among top earners, for both men andwomen, and highlights that gender improvements coincidewith a larger persistence over time ofwomen at the top of the income distribution. Although in the 1980swomenwere twice as likely asmen to drop out from the top earners, in recent years the probability of men and women of droppingout hasbecome similar. Third, industryandagevariationsprovide further insights on the dynamics of the gender gap. The authors do not find any obvious relationship between the evolution of income across industries and the mitigation of the gender gap over time, whereas the distribution of age among top earners shows significant changes, with new cohorts of women making inroads into the top 1% earlier in their lifetime than previous cohorts (this results does not hold for the top 0.1% earners).