We propose a new method of estimating wage differences among individuals from different populations, which is based on looking at the distribution of fixed effects from a wage regression. We use this method to compare males’ and females’ wage distribution at different levels of education and at different ages. In most cases, unobserved, time-constant factors increase the wage of males, compared to that of females, by a constant share. But this is not true when looking at the tails of the distribution. We also discuss the advantages of our method.

JEL: J71, J31

中文翻译：

---

性别差距：看整个分布

我们提出了一种新的估算不同人群之间工资差异的方法，该方法基于工资回归的固定效应的分布。我们使用这种方法来比较不同教育水平和不同年龄的男性和女性的工资分布。在大多数情况下，与女性相比，未观察到的时间常数因素使男性的工资增加了一定的比例。但是，从分布的尾部来看，情况并非如此。我们还讨论了我们方法的优点。

JEL： J71，J31