Explaining the competition gender gap: The role of sexism

Men occupy the overwhelming majority of positions of power within business and politics and tend to earn higher wages, both in the United States and around the world. Researchers hypothesize that a part of this gender gap can be attributed to women’s lower preference for competition. While a large literature documents this gender differential (Niederle and Vesterlund, 2007; Gupta, Poulsen and Villeval, 2013; Sutter and Rutzler, 2010), few papers have explored the reasons for these differences.  Some argue that men are innately more willing to compete than women (Lawrence, 2006).  In contrast, Gneezy, Leonard and List (2009) measure the willingness to compete in matrilineal and patrilineal settings and offer evidence that the differences may be cultural.  We hypothesize that living in societies where sexism against women is prevalent lowers women’s confidence and thus causes them to shy away from competition.

This paper uses a controlled experiment with priming to examine the role of sexism in explaining the gender differential in the willingness to compete.  Using experimental data collected online with the crowd-sourcing website MTurk, this paper tests whether exposure to sexist statements alters men’s and women’s willingness to compete.  In particular, we test how exposure to different primes (benevolent sexism, hostile sexism, and neutral) influence a subject’s willingness to compete.  We measure the willingness to compete by asking subjects to select between a piece-rate payment scheme (no competition) and a tournament payment scheme (competitive) for a hypothetical task where they were to add a series of two-digit numbers over a period of two minutes.  We found that exposure to sexism had no significant impact on the willingness of females to compete, but males were significantly more willing to compete when exposed to sexism primes.  This is an interesting and novel finding that we believe might shed light on disparate labor market outcomes of men and women.

In January, we plan to run additional experiments in order to test (a) whether the results are sensitive to the hypothetical nature of the task we asked our MTurk subjects to complete, and (b) the willingness to compete in other domains (i.e., solving logical or word problems rather than math problems).  The results should be fully analyzed in time for the IAES meeting in March.