VoxEU Column Gender Labour Markets

Addressing gender inequality via choice architecture

Many work environments require their employees to apply for promotions, a process that results in fewer women opting to compete. This column presents evidence to suggest that changing promotion schemes to a default where everyone is considered but has the option to ‘opt out’ could help close the gender gap in applications to compete for promotions. 

Only three women have received the Nobel Prize in Physics. The most recent of them, Donna Strickland, held the rank of Associate Professor when she was awarded the prize. Positive reactions to her win came with puzzlement over how such an eminent scientist had not yet been appointed a Full Professor. Responding to a reporter’s question about the matter, Strickland explained simply: “I never applied” (BBC World Service 2018). 

The gender gap in the propensity to apply and compete for promotions goes well beyond a single anecdote. Indeed, explanations for the gender promotion gap abound. Some scholars focus on employer-side bias and discrimination against women (Heilman et al. 2004, Milkman et al. 2015, Riach and Rich 2006). Others attribute the gap to gender differences in behaviours that facilitate career advancements, such as negotiating a higher salary (Bowles et al. 2007). Yet another body of research suggests that women do not advance to powerful positions because, as in our motivating example, they just do not apply, or ‘compete’, for competitive selection processes (e.g. rewards and promotions). Past research has found that women are less likely than men to participate in competition, in the absence of any performance differences (for a review, see Niederle and Vesterlund 2011). Women are less likely to participate in competitive selection processes that require self-nomination, especially when putting themselves forward risks provoking backlash or social rejection (Amanatullah and Morris 2010, Bosquet et al. 2018, Moss-Racusin and Rudman 2010). 

The gender gap in competition has important consequences for gender equality in the labour market and in organisations, because competitions are pervasive in both the pipeline to employment and within organisations. Competition is required for everything from university admissions, scholarship and award applications to extracurricular activities, job applications, and promotions.          

Applying a choice architecture lens to competitive selection processes

In a recent paper (He et al, 2019), we take a closer look at the architecture of competitive selection processes to examine whether the framing of the choice to participate in competitive selection processes affects the gender gap in participation in competition. Typically, competitive selection process (e.g. promotions) require participants to self-nominate. In other words, individuals must actively apply, or ‘opt in’, because the default option is to not participate. 

Drawing on the literature on choice architecture, we explore whether using an ‘opt-out’ frame (i.e. making competition the default) leads to greater participation in competition for women (Down et al. 2009, Johnso et al. 2002). Opt-out framing exploits the tendency to stick with a default condition to encourage enrolment into a desired choice by making that choice the default. Opt-out framing has successfully increased enrolments in retirement savings and organ donation programmes (Choi et al. 2002, Johnson and Goldstein 2003).

To test this research question, we recruited 482 undergraduate students to participate in our study. We told participants that they would complete multiple rounds of a maths task requiring them to add up double-digit numbers, and that they had 5 minutes each round to complete as many questions correctly as they could (this is a well-established experimental task; e.g. Niederle and Vesterlund 2007). They completed three rounds of this task, and the compensation changed for each round. In the first round, participants received $0.50 per correct answer; this was the non-competitive piece-rate compensation. In the second round, a participant’s number of correct answers was compared to three other participants in the room (competitive tournament compensation). If their score was the highest, they won $2 per correct answer; if not, they won nothing. Finally, in the third round, participants had the option to choose either compensation scheme. We introduced random assignment to treatments at this stage.

Specifically, participants were randomly assigned to an opt-in or opt-out frame of the choice to compete. In the opt-in condition, participants were told that they would be paid $0.50 per correct answer (the non-competitive piece rate), but they had the option to opt in to the competitive tournament. In other words, the default was to not compete, with the option to opt in to compete. In the opt-out condition, participants were automatically enrolled in the competitive tournament, but told that they had the option to opt out to the non-competitive piece-rate compensation. In this treatment, the default was to compete, with the option to opt out of competition.

We found that men and women performed equally on the maths task. However, in the opt-in condition, only 47% of women chose to opt in to the competitive tournament, compared to 72% of men. In other words, we replicate findings from previous literature and show that in the absence of performance differences, women are much less likely to compete. 

Figure 1 Choice of compensation in Stage 3 of Study 1

Notes: ‘Opt in’ indicates that participants were in the condition with a piece-rate default compensation. ‘Opt out’ indicates the condition that assigned participants to competing by default in a winner-take-all tournament within each group of four, but allowed them to opt-out and instead be compensated on a piece-rate basis. The vertical lines represent 95% confidence interval for the mean percentages.

When we look to the opt-out condition, the results are again striking: 76% of men chose to compete; but this time, 75% of women also chose to compete. In other words, the gender gap we observed in the opt-in condition (consistent with past research), was eliminated in the opt-out condition. When competition was the default, women were just as likely as men to compete.

Although these results are encouraging, a few concerns remained. By nudging women to compete more, are we truly encouraging women to behave in ways that benefit them, or will we see unintended negative consequences for performance or broader wellbeing? Fortunately, our results suggest that the answer to this question is the former. Women who were nudged to compete did not perform worse and did not experience more anxiety. Instead, we found that women were actually more likely to make a payoff-maximising choice in the opt-out condition than the opt-in condition. 

Figure 2 Correct responses and anxiety levels by condition, choice and gender

Notes: A higher value indicates a higher anxiety score. The vertical lines indicate 95% confidence interval for the mean percentages.

Beyond these initial results, we obtained nearly identical findings in a replication that also explored why women are more likely to choose competition under an opt-out frame. The results of this study suggest that participants in the opt-out frame expected more participants (especially those of their own gender) to choose the competitive tournament compensation compared to participants in the opt-in frame condition. Thus, opt-out framing might encourage women to compete more by changing a ‘descriptive’ social norm. Further, in a second experiment, we examined whether using an opt-out versus opt-in choice architecture for competition would affect the likelihood that evaluators would choose to promote a woman, and found no such difference.

Implications and conclusion

Previous understanding of the gender gap in competition suggested that women participate in competition less often than men because they have a different ‘taste’ for these behaviours, implying that women are less likely than men to engage in these behaviours across most, if not all, contexts. Our results show that women’s lower propensity to compete is not absolute, and that under certain contexts, women are just as likely as men to compete. In particular, our results suggest that women may compete less not because they are simply less competitive, but because they are responding differently to situational influences (e.g. defaults).

These implications have relevance for organisations aiming to close the gender gap in career advancement. To date, popular interventions to reduce the gender gap in hiring and promotion have primarily focused on diversity training (often in the form of unconscious bias training), or coaching women to be more active and visible at work – to “lean in” as Sheryl Sandberg puts it (2013). Both of these approaches have met with mixed success despite the millions of dollars invested. Unconscious bias training has mixed effects on attitudes (Kalev et al. 2006, Leslie 2018, Macrae et al. 1994, Sanchez and Medkik 2004), and little to no effect on behaviour (Chang et al. 2019). Meanwhile, coaching women to negotiate better and act more assertively puts them at risk of facing backlash for adopting these more stereotypically ‘masculine’ behaviours (Moss-Racusin and Rudman 2010, Rudman 1998, Rudman and Glick 2001), and also runs the risk of attributing responsibility for both causing and fixing the problem to women themselves (Fitzsimons et al. 2018, Kim et al. 2018). 

Our research suggests a new approach: instead of attempting to change people’s minds and biases or trying to ‘fix the women’, we focus instead on structural changes meant to de-bias organisations. Biases and systematic barriers are often built into the system itself. For instance, an opt-in promotion frame advantages those who are more comfortable and likely to self-promote, compete, and self-nominate. Our results suggest that framing this decision as an opt-out choice could make the process fairer for everyone. 

Although we need more research to directly apply opt-out framing for competitive selection processes in organisations (i.e. promotions), our results question the long-held assumption that women are simply less competitive than men, and challenge organisations to reflect on their own policies and processes to see if bias may be hidden in their choice architectures.


Amanatullah, E T and M W Morris (2010), "Negotiating Gender Roles: Gender Differences in Assertive Negotiating Are Mediated by Women’s Fear of Backlash and Attenuated When Negotiating on Behalf of Others", Journal of Personality and Social Psycholog.

BBC World Service (2018), News Hour, Donna Strickland: “Nobody expects the Nobel prize.” 

Bosquet, C, P P Combes and C García-Peñalosa (2018), "Gender and Promotions: Evidence from Academic Economists in France", The Scandinavian Journal of Economics.

Bowles, H R, L Babcock, and L Lai (2007), "Social incentives for gender diVerences in the propensity to initiate negotiations: Sometimes it does hurt to ask", Organizational Behavior and Human Decision Processes 103: 84–10.

Chang, E H, K L Milkman, D M Gromet, R W Rebele, C Massey, A L Duckworth and A M Grant (2019), "The mixed effects of online diversity training", Proceedings of the National Academy of Sciences 116(16): 7778–7783.

Choi, J J, D Laibson, B C Madrian and A Metrick (2002), "Defined Contribution Pensions: Plan Rules, Participant Choices, and the Path of Least Resistance", 67–114.

Downs, J S, G Loewenstein and J Wisdom (2009), "Strategies for Promoting Healthier Food Choices", American Economic Review 99(2): 159–16.

Fitzsimons, G M, A C Kay and J Y Kim (2018), “ 'Lean In' Messages and the Illusion of Control", Harvard Business Review

He, J, S Kang and N Lacetera (2019), "Leaning In or Not Leaning Out? Opt-Out Choice Framing Attenuates Gender Differences in the Decision to Compete", NBER Working Paper No. 26484.

Heilman, M E, A S Wallen, D Fuchs and M M Tamkins (2004), "Penalties for Success: Reactions to Women Who Succeed at Male Gender-Typed Tasks", Journal of Applied Psychology 89(3): 416–42.

Johnson, E J, S Bellman and G L Lohse (2002), "Defaults, Framing and Privacy: Why Opting In-Opting Out", Marketing Letters 13(1): 5–15.

Johnson, E J and D Goldstein (2003), "Do defaults save lives?", Science 302(5649): 1338–1339.

Kim, J Y, G M Fitzsimons and A C Kay (2018), "Lean in messages increase attributions of women’s responsibility for gender inequality", Journal of Personality and Social Psychology 115(6): 974–1001.

Macrae, C N, G V Bodenhausen, A B Milne and J Jetten (1994), "Out of mind but back in sight: Stereotypes on the rebound", Journal of Personality and Social Psychology 67(5), 808–817.

Milkman, K L, M Akinola and D Chugh (2015), "What happens before? A field experiment exploring how pay and representation differentially shape bias on the pathway into organisations", Journal of Applied Psychology 100(6): 1678–1712.

Moss-Racusin, C A and L A Rudman (2010), "Disruptions in Women’s Self-Promotion: The Backlash Avoidance Model", Psychology of Women Quarterly 34(2): 186–202.

Niederle, M and L Vesterlund (2007), "Do Women Shy Away From Competition? Do Men Compete Too Much?", The Quarterly Journal of Economics 122(3): 1067–1101.

Niederle, M and L Vesterlund (2011), "Gender and Competition", Annual Review of Economics 3(1): 601–630.

Riach, P A and J Rich (2006), "An Experimental Investigation of Sexual Discrimination in Hiring in the English Labor Market An Experimental Investigation of Sexual Discrimination in Hiring in the English Labor Market", Advances in Economic Analysis & Policy 6(2)

Rudman, L A (1998), "Self-promotion as a risk factor for women: The costs and benefits of counterstereotypical impression management", Journal of Personality and Social Psychology 74(3): 629–645.

Rudman, L A and P Glick (2001), "Prescriptive Gender Stereotypes and Backlash Toward Agentic Women", Journal of Social Issues 57(4): 743–762.

Sandberg, S (2013), Lean in: Women, work, and the will to lead, Knopf Doubleday Publishing Group.

735 Reads