Abstract
Occupational segregation and pay gaps by gender remain large, while many of the constraints traditionally believed to be responsible for these gaps seem to have weakened over time. We explore the possibility that women and men have different tastes for the content of the work that they do. We relate job satisfaction and job mobility to measures that proxy for the content of the work in an occupation, which we label ‘people’, ‘brains’ and ‘brawn’. The results suggest that women value jobs high on ‘people’ content and low on ‘brawn’. Men care about job content in a similar fashion, but have much weaker preferences. High school students show similar preferences in a discrete choice experiment and indicate that they make their choices based mainly on preferences for the work itself. We argue that the more pronounced preferences of women can account for occupational sorting, which often leads them into careers with large pay penalties for interruptions due to childbearing.
Highlights
There is a strong negative relationship between female satisfaction and the share of males. This relationship is fairly stable across different specifications and contexts, and the magnitude of the association is not attenuated by personal characteristics or other occupation averages
The effect is muted for women but largely unchanged for men when we include three measures that proxy the content and context of the work in an occupation, which we label ‘people,’ ‘brains,’ and ‘brawn.’ These results suggest that women may care more about job content, and this is a possible factor preventing them from entering some male dominated professions
Where Yijt is either job satisfaction or a binary variable which indicates whether a person stayed in the same occupation in the period for individual i in occupation j and year t, share of males (SOM) js is the proportion of males in a particular occupation, X js is a vector of other occupational averages, Xijt is a vector of individual-level control variables, t are wave effects, and a are region effects
Summary
Including these measures reduce the coefficient on the share of men in the occupation by a third or more for women, while it does little to the coefficient in the male job satisfaction regressions. While we argue that these results point to differences in tastes for job attributes, the share of males may proxy for differences in the work environment, which are perceived differentially by men and women.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
