The perception that women have higher turnover rates than men is widespread. A recent study has argued, with a touch of sarcasm, that "[p]opular stereotypes, which economists refer to as 'stylized facts, I portray women as relatively poor bets as workers because they have ... higher quit rates than males. 1I Waite and Berryman [1985: 61]. Indeed, in a recent article on occupational segregation, Goldin takes this IIfact ll as the premise for her model, although in support of this position she cites only a 1920 study. Goldin [1985]. While this may well be adequate for Goldin's historical analysis, a study conducted at a time when the labor market experience of women was completely different than it is today does not illuminate the question of the cur',:"ent relative turnover rates of men and women. The isslle of differences -- or perceived differences -- in expected job tenure between male and female workers may have an important bearing on the size of the male-female wage differential. If there are large personnel investment costs associated with job turnover, an employer attempting to earn a normal rate of return on a fixed personnel investment would not be willing to pay women the same wage as equally productive men who are expected to remain on their jobs longer.