Regardless of their stage of development, most economies exhibit significant gender gaps in wages and other labour market outcomes. A growing number of studies explore whether or not such gaps might be due to gender differences in attitudes to competition or to risk. However, economists have not yet reached consensus about whether observed differences in economic preferences are innate, or instead differ across the environments in which men and women find themselves. If they are not innate, then policy may play a role in reducing gender gaps in labour market outcomes.
Experimental studies have found the competitive choices made by men and women differ according to whether they are competing against men or women (e.g. Gneezy et al. 2003, Gneezy and Rustichini 2004, Niederle and Vesterlund 2011). Moreover, observed performance also varies with the gender of competitors, as in Gneezy et al. 2003. Further studies have explored the role of nature or nurture in explaining these differences, and found that these competitive preferences can be modified through nurturing (Gneezy et al. 2009, Booth and Nolen 2012a).
Why do women perform worse when competing against men?
To investigate this, in a recent paper we analyse unique performance data from a real-world activity – speedboat racing in Japan – that is by its very nature competitive, and where the potential payoffs from winning are high (Booth and Yamamura 2016). The novel features of these data are that participants are randomly allocated to either single-sex or mixed-gender groups for the competition, and that the equipment (boat and engine) used by each participant on race-day is also randomly assigned. Moreover, women and men compete under exactly the same conditions. Detailed information about the institutional features of this sport can be found in our paper (Booth and Yamamura 2016).
Speedboat races in Japan
There are 24 speedboat racing stadiums throughout Japan and boat races are randomly held about four days per week in each stadium. Racers go to many different stadiums to compete. In each racing fixture, there are 12 races, and six racers compete in any given race. The circuit is a large artificial pond or sectioned-off body of water 600 metres in length. Competitors race around it three times, leading to a total race-distance of 1800 metres.
Women represent approximately 13% of all speedboat racers. The rules of the races are strictly monitored and any breaching of the rules results in disqualification. The potential payoffs are very high and are sourced from betting. Average annual earnings range from 5 million Japanese yen for racers in the bottom grade up to 33 million yen for racers in the top grade.1
Races are tightly monitored, and severe sanctions on disqualified racers mean they cannot participate, resulting in a fall in annual revenue. Consequently, racers have strong incentives to follow the rules in order to win the race. But they also face trade-offs because, in order to win, they may have to engage in risky lane-changing to improve their position.
Figure 1 Racing around corner at the stadium at Suminoe in Osaka
Using data from seven stadiums with complete information on all racers’ records over 18 months, we explore how female and male performance and strategies in the mixed-gender races differ from the single-gender races. Our data are in panel form, where we have information for each racer’s performance time and strategy across all the races in which they have competed. Thus we have a total of over 15,000 women-race observations and over 127,000 men-race observations – a far larger dataset than used by other studies to date. Note that because of the preponderance of men, the majority of participants in the mixed-sex races are male (this mimics the position of women in the STEM disciplines of science, technology, engineering, and mathematics.)
Our conjectures about behaviour
Hypothesis 1: Women racers are faster in the single-sex races than in the mixed-sex races, while men racers are faster in the mixed-sex races than in the single-sex.
What underlies this conjecture? A society's prescriptions about the appropriate modes of behaviour for each gender might result in individuals experiencing a loss of identity should they deviate from the relevant code (e.g. Akerlof and Kranton 2000). If women are in general expected to behave less aggressively than men, they are more likely to follow this behavioral code when they are in a mixed-gender race than an all-female race. This is because the mixed-gender race – to which they are randomly assigned – triggers in them an awareness of their gender identity.
Hypothesis 2:Women racers follow a less aggressive strategy than men, and this will be more pronounced in the mixed-sex races.
This conjecture follows from findings from an earlier study that shows gender differences in risk-aversion vary across single-sex and co-ed environments (Booth and Nolen 2012b). Racing involves skill not only in maneuvering the boat but also in jockeying for a desirable position, since the inner lanes confer an advantage. While lane-changing can bring benefits, it can also bring costs (strict rules, leading to tough penalties). If men are more confident and less risk-averse, they will be more likely to adopt an aggressive strategy. Within our dataset, ‘aggression’ is proxied by lane-changing.
Owing to the rules of speedboat races, racers who behave aggressively are more likely to break the rules and be disqualified. If Hypothesis 2 is found to hold, we have a further hypothesis:
Hypothesis 3: Women racers are less likely to be penalised than men in the mixed-gender race.
The racing start is based on the premature start system in which lane-changing occurs. Prior to this, each race participant runs a short solo exhibition run, during which the participant obtains information about the boat they have been randomly allocated and reveals their prowess to the betting fraternity. Clearly strategy will play a big part in the actual race, but it will play a much smaller role in the solo exhibition run, where participants do not compete directly with the other racers. Thus, jockeying for position is not relevant.
We investigate how male-dominated circumstances affect women’s and men’s racing performance. We control for individual fixed-effects plus a host of other factors affecting performance. Our estimates reveal that women’s times are slower in mixed-sex races than in all-women races, whereas men’s times are faster in mixed-gender races than men-only races. We also find the same results when we use as the dependent variable ‘place in the race’. Thus, our evidence supports Hypothesis 1.
We also find that in mixed-sex races, male racers tend to be more aggressive – as proxied by lane-changing – in spite of the risk of being penalised for contravening the rules, whereas women follow less aggressive strategies. This supports Hypothesis 2. However, we find no difference in disqualification rates between genders, and thus no support for Hypothesis 3.
We suggest that gender differences in risk attitudes and confidence may result in different responses to the competitive environment and penalties for rule-breaking, and that gender identity is also likely to play a role.
The first finding above is of particular interest. It shows that female competitive performance – even among women who have chosen a competitive career and who are very good at it – is enhanced by being in a single-sex environment rather than in a mixed-sex environment in which they are a minority. The other findings listed above are also of great interest, since they follow from our investigation of the mechanisms through which our first finding operates. In particular, we suggest in the paper that male racers are aggressive but not imprudent by taking into account competitors’ condition as well as the risk of disqualification when jockeying for position.
The gender proportion in the mixed-sex speedboat races is skewed towards men. Women racers assigned by lot to a mixed-sex race will typically face five male competitors, and infrequently four. We argue that this gender imbalance is likely to trigger awareness of gender identity for both men and women, and that this might go some way to explaining our observed differences in behaviour across the mixed-sex and single-sex groups. For example, a man’s gender identity may lead him to view being defeated by women as more dishonourable than by men, and he will try to avoid it.
Our findings may well have implications for other activities in which men and women compete with one another and where the gender balance is skewed in favour of men. One example is in the STEM disciplines, where being in a minority may well affect the performance of the women. We hope that future research will investigate this further.
Akerlof, G A and R E Kranton (2000), “Economics and identity”, Quarterly Journal of Economics 115, 715–53.
Booth, A L and P Nolen (2012a), “Choosing to compete: How different are girls and boys?”, Journal of Economic Behavior & Organization 81(2), 542-555.
Booth, A L and P Nolen (2012b), “Gender differences in risk behaviour: does nurture matter?”, Economic Journal 122(558), F56-F78.
Booth A L and E Yamamura (2016), “Performance in Mixed-sex and Single-sex Tournaments: What We Can Learn from Speedboat Races in Japan”, CEPR Discussion Paper No. 11685.
Gneezy, U, M Niederle and A Rustichini (2003), “Performance in competitive environments: Gender differences”, Quarterly Journal of Economics 118, 1049-1074.
Gneezy, U, and A Rustichini (2004), “Gender and Competition at a Young Age”, The American Economic Review: Papers and Proceedings 94(2), 377-381.
Gneezy, U, K Leonard and J List (2009), “Gender differences in competition: evidence from a matrilineal and a patriarchal society”, Econometrica 77, 1637-1664.
Niederle, M and L Vesterlund (2007), “Do Women Shy Away from Competition? Do Men Compete Too Much?”, Quarterly Journal of Economics 122(3), 1067-1101.
Niederle, M and L Vesterlund (2011), “Gender and Competition”, Annual Review of Economics, Annual Reviews, 3(1), 601-630.
 One US dollar is roughly equivalent to 100 Japanese yen.