VoxEU Column Frontiers of economic research Gender Politics and economics

Gender and competition

At this week’s summit on the future of the euro, Angela Merkel will be one of few women in a room full of men. This column provides experimental evidence to suggest that women are often less driven by the desire to compete and have less belief in their abilities than men. The result is that even the highest ranks of power may be bereft of the most able of candidates.

Women continue to be underrepresented in the corridors of power. Despite recent gains, the numbers are striking. Fewer than 20% of national legislators are women. Just 3.2% of Fortune 500 companies currently have a female CEO. Such continuing gender disparities appear to fly in the face of the elimination of gender gaps in areas such as higher education, where we now see more women graduating from universities than men.

To help explain these continuing disparities, economists have been examining the role that individual preferences may play. The findings suggest that gender disparities in positions of power may be in part due to women experiencing less enjoyment from participating in these competitive fields and less confidence in their abilities to excel.

At the same time, gender disparities change dramatically when the cultural context is different or when the competition is conducted in a different way. Women choose to compete less frequently than men in most places, but not in matrilineal societies where traditional gender roles are reversed. Gender differences in competitive preferences also diminish substantially when people compete in teams rather than as individuals. These findings indicate the importance of cultural norms in driving gender disparities, as well as opening the door for policy interventions to address them.

The pioneering work of Muriel Niederle and Lise Vesterlund provided the initial insights into the underlying reasons why women choose to enter competitions less frequently than men (Niederle and Vesterlund 2007). They used a lab experiment to attempt to unpack these earlier results indicating a greater male propensity for entering competitions.

In their experiment, lab subjects completed a task – adding numbers – for which men and women generally perform equally well. Subjects first answered as many problems as they could and were paid a certain amount for each problem solved correctly. In the next round, subjects were told they would be competing in a tournament and would be paid a higher amount for each problem they answered correctly if they won the tournament, but zero if they failed to answer more problems than other competitors. After that, subjects participated in two rounds where they decided whether they wanted to be paid based on the piece rate or whether they wanted to enter the tournament.

They found that women were 38% less likely than men to choose to compete, even though men and women were equally good at the task. The different treatments additionally enabled them to disentangle the potential reasons why women choose to compete less often than men. They found that women, relative to men, have a lesser taste for competition and less confidence, even controlling for qualifications on the task. Alternative explanations such as risk aversion did not explain why women were less likely to choose to compete.

The large gender disparity in entering competitions builds on earlier evidence suggesting that competitive environments may sometimes increase performance for boys relative to girls, as well. For example, Gneezy and Rustichini considered the response of Israeli fourth-graders to competitive environments. They found that introducing a competitive aspect to running races caused the boys to perform better, but had no effect on the performance of girls.

Why do these gender differences in how people respond to preferences arise? How might these disparities be addressed? Recent research provides some insights. For example, Gneezy et al (2009) conducted experiments to determine the nature of the gender competition gap far from the university lab. They conducted experiments with two very different societies – the Maasai in Tanzania and the Khasi in India. Most importantly, the former of these societies is patriarchal, while the latter is matrilineal.

Their results are fascinating. For the Maasai sample, they found the standard result that men show a greater propensity than women in selecting into competitive environments. For the Khasi sample, however, the standard result flips altogether: women were actually more likely to compete.

These findings may not apply to all matrilineal societies, as suggested by some results from Malawi. However, the results do underscore the importance of cultural factors in driving gender differences in competition. There appear to be cultural contexts in which the gender competition gap reverses, indicating the importance of cultural norms in driving competitive preferences.

Another way to mitigate the gender competition gap is to change the terms under which competitions are conducted. In an experiment that I ran with Jennifer Pate, we attempted to ascertain the effect that team competition would have on gender disparities (see Healy and Pate 2011). We took Niederle and Vesterlund’s experiment and had people compete in two-person teams. Subjects were randomly assigned a partner at the beginning. Instead of being paid based on individual performance, they were paid based on the combined performance of their two-person team. They decided whether to enter competitions against other teams, with one person’s decision chosen at random to represent the group to avoid strategic considerations influencing decisions.

The results were striking. Compared to our individual-competition sessions, the gender competition gap in our team sessions was only one-third as large. Moreover, our findings indicated that the reason that teams decreased gender disparities was due almost entirely to changes in competitive preferences. Relative to men, women preferred to compete more in teams than as individuals.

These results are consistent with evidence from politics that team-based competitions can lead to markedly greater female participation. It appears to be the case that districts with party lists elect a much higher share of women than single-member districts. For example, in Germany and New Zealand, where some representatives are elected by each method, women are about three times more likely to be elected from the team-based competitions than the individual ones. While other motivations are certainly involved, our results suggest that greater female preference for team competition may be part of the explanation.

Altogether, the research on gender differences in competitive preferences is helping us to understand much better one of the reasons why women continue to be underrepresented in the halls of power. This phenomenon has substantial implications not only for equity, but also for efficiency. Most often, we see low-performing men choosing to compete and high-performing women opting out. The importance of cultural norms and the competition environment thus has substantial policy implications. Designing competitions to encourage greater female participation may lead not only to greater gender equity, but also to better outcomes.


Gneezy, Uri and Aldo Rustichini (2004), “Gender and Competition at a Young Age”, American Economic Review, 94(2):377-381.

Gneezy, Uri, Kenneth Leonard, and John List (2009), "Gender Differences in Competition: Evidence From a Matrilineal and a Patriarchal Society", Econometrica, 77(5):1637-1664.

Healy, Andrew and Jennifer Pate (2011), “Can Teams Help to Close the Gender Competition Gap?”, Economic Journal, 121(4):1192-1204.

Niederle, Muriel and Lise Vesterlund (2007), "Do Women Shy Away from Competition? Do Men Compete Too Much?", Quarterly Journal of Economics, 122(3):1067-1101.

420 Reads