By Paul Walker 21/05/2016

New research suggests that women earn, on average, around 31% less that men in STEM (Science, Technology, Engineering or Mathematics) subjects within a year of completing a PhD.

An obvious question is why.

Perhaps surprisingly the answer to this questions can be reduced to just two factors: 1) field of study and 2) kids.

But after controlling for differences in academic field, the pay gap between males and females is reduced to around 11% in first-year earnings. There is a tendency for women to graduate in less-lucrative academic fields – such as biology and chemistry than comparatively industry-friendly fields, such as engineering and mathematics.

This 11% difference can be explained entirely by the finding that married women with children earned less than men. Note that an unmarried, childless woman earned, on average, the same annual salary after receiving her doctorate as a man with a PhD in the same field.

These results come from a paper, “STEM Training and Early Career Outcomes of Female and Male Graduate Students: Evidence from UMETRICS Data Linked to the 2010 Census” by Catherine Buffington, Benjamin Cerf, Christina Jones and Bruce A. Weinberg published in the American Economic Review: Papers & Proceedings 2016, 106(5): 333–338.

There are a few things to keep in mind with the study. Given that the data is on PhD holders these are specialised labour markets. There are 1,237 students – 867 male and 370 female who graduated between 2007 and 2010 from just 4 universities in the study. Also the labour market outcomes likely reflect some unobserved heterogeneity, including in hours worked, and potentially household decisions on housework and child care. There is also the question of what happens in later post-doc years.

That said, the results, in particular the effects of children on earnings, are consistent with other work I have seen on pay differences between men and women. A number of other studies have reported similar gender pay gaps and have identified similar contributing factors — but few have systematically broken down the relative contributions of the different variables.


Feature image: CC flickr U.S. Department of Agriculture