Sex discrimination in labor markets may generate a wage gap between men and
women that exceeds any gap in marginal productivity. We test for this type
of discrimination using unique firm-level data on manufacturing firms in I
srael. There is a statistically significant negative association between wa
ges and the proportion of a firm's workforce that is female. However, there
is also a statistically significant negative association between marginal
productivity and the proportion of females. The difference between the wage
and productivity gaps is small relative to wage-regression estimates of wa
ge discrimination, and is not statistically significant, which is most cons
istent with no discrimination.