The Old Boys' Club

A paper published in the American Economic Review sheds light on a new factor that makes it more difficult for women to climb the corporate ladder: Male managers support their male subordinates more in their career progression.

What are the results?

Cullen and Perez-Truglia (1) have shown that when male employees are assigned to a male manager, they are promoted faster than when they are assigned to a female manager. However, this effect is not observed for female employees: Women’s career progression is the same, no matter the gender of their manager.

What explains this result? The authors hint at “the old boy’s club”: Informal social interactions that take place between a male manager and his male subordinates, but not between a male manager and his female subordinates. Indeed, the authors show that the beneficial effect of a male manager on his male subordinates’ career is only observed when they work near each other (and therefore can have informal interactions), or when both manager and subordinate are smokers (and therefore take smoke breaks together).

How did they show that?

The “male-to-male advantage”

The main hypothesis of Cullen and Perez-Truglia is that being assigned to a male manager will increase the likelihood of promotion for male (but not female) employees.

However, to test this idea, they cannot simply compare the career progression of male and female employees who were assigned to a male vs. a female manager. Imagine for example that people who are more career-oriented insist to work under a male manager. In this case, we may observe an association between being assigned to a male manager and being promoted faster, but we shouldn’t conclude that the faster promotion was caused by the male manager: It might simply reflect the stronger career-orientation of the employees. This is what researchers call a “confounded relationship”.

To control for the many hidden factors that may affect the relationship researchers seek to study, experiments are the “gold standard”. To test if being assigned to a male manager has an actual (“causal”) relationship on promotion, the authors would need to run an experiment in which employees (men and women) would be randomly assigned to a manager (a man or a woman). They would then measure the effect of this random assignment on employees’ career progression in later years.

Unfortunately, this is typically the kind of experiment that is practically impossible to run. No organization would accept to randomly shuffle its employees and managers, and for good reasons: It would be way too costly and disruptive!

Cullen and Perez-Truglia therefore had to find a middle-ground: A test that would eliminate many possible alternative explanations, but that would be feasible to study. To do so, they used the natural rotation of managers in a large company: When managers move laterally from one team to another, independently of the employees. Even if this rotation of managers was not completely random, the data suggests that it was not influenced by the characteristics of the employee or those of the manager.

The researchers had access to four years of record (2015-2018) of a company, including almost 15 000 employees, among whom around 1300 were managers at some point. In this time frame, they identified 8670 “transition events” in which a team underwent a change in manager.

Studying these events revealed that, in the two and a half years following a manager transition, male employees who transitioned from a female manager to a male manager experienced a salary increase 15% greater than male employees who stayed with a female manager. In a mirror effect, when male employees transitioned from a male manager to a female manager, they experienced a salary increase that was 7.5% lower than employees who stayed with a male manager.

In contrast, female employees experienced similar salary increase, regardless of the manager’s gender they were assigned to after the transition.

According to Cullen and Perez-Truglia, without this male-to-male advantage, the gender pay gap between men and women would be reduced by 40%.

Are social interactions the process through which this “male-to-male advantage” occurs?

They then tested whether this male-to-male advantage could be explained by what they call “schmoozing”: Social interactions through which managers and employees get to know each other, and employees can gain the sympathy and favor of their manager.

They found evidence that this male-to-male advantage was large when male managers and male employees work in close physical proximity, but absent when they are physically distant (e.g., working on different floors). They also found that male employees were more likely to share breaks with their manager when he was male (rather than female), while female employees spent as much break with their manager regardless of their gender.

Finally, to confirm that this male-to-male advantage occurs through “schmoozing”, and not through other unobserved variables (for example that men would be more ambitious than women and would not hesitate to make bold asks when they think they deserve something), Cullen and Perez-Truglia used a quasi-experiment.

Again, in an ideal scenario, the researchers would run an experiment: Male employees would be randomly assigned to socializing (vs. not) with their male manager, then the researchers would test for differences in career advancement. However, this kind of experiment is once again not feasible.

Instead, Cullen and Perez-Truglia rely on a clever proxy: They looked at what happens to male employees who smoke, and get assigned to a male manager who smoke as well. The idea is that employees and managers who are both smokers would spend more time together, and that their shared smoking habit would give opportunities for “schmoozing”. The advantage of this proxy is that it is not associated with other unobserved variables that might drive the effect.

As expected, the researchers found a smoker-to-smoker advantage, equivalent in magnitude to the male-to-male advantage: Not only male smoking employees who transitioned from a non-smoking male manager to a smoking male manager spent more breaks with their new managers, but they were also promoted faster, while no such effect was observed for non-smoking employees.

Why is it happening?

This “male-to-male advantage” might stem from a basic principle of social interactions called “homophily”: People have a natural tendency to interact with people who look like them. As such, it is often easier for men to interact with other men, and for women to interact with other women. This preference for similar others might explain why male managers and male subordinates spend time together, and therefore why male managers favor their male subordinates.

However, the present research also suggests that female managers do not favor their female subordinates as male managers do with their male subordinates. Why not? Research gives us some clues (2). When a group is stereotyped and has a hard time climbing the corporate ladder, its successful members can be reluctant to help other members, and on the contrary try to distance themselves from the rest of the group. As such, women who see their gender as a liability for their career (for example because in many organizations people keep seeing women as unfit for leadership positions) may be tempted to reduce the association with other less successful women within the organization to improve their personal outcome.

What are the implications of these results?

This tendency of male managers to favor their male subordinates can be particularly detrimental to women’s career in a world in which men keep occupying most positions of power within organizations (3,4): While their male counterparts can find support at each hierarchical level to get promoted at the upper level, women have less support, and end up being unable to compete with men for the best positions.

This finding also mirrors other findings about the way women are mentored compared with men (5). Women typically have difficulty finding senior executive mentors, and are more likely than men to be mentored by junior members of the organization. In addition, women tend to be over-mentored (they receive a lot of feedback and advice), but under-sponsored (their mentors are less likely to use their influence to advocate for their promotion to upper levels).

Finally, recent studies have suggested that men, particularly if they occupy a position of power or are more senior, might be reluctant to establish relationship with female colleagues, particularly if they are their subordinates or more junior (6–10). Indeed, following the #MeToo movement, many men have started avoiding one-to-one inetractions with female organizational members for fear that their behaviors might be misinterpreted.

How can organizations avoid this tendency?

A first step would be to encourage managers to engage in thoughtful consideration and evaluation of all their subordinates when distributing resources (e.g., pay raise, promotion, budget increase, assignment to developmental projects). Being aware of the favorable bias they might have towards their male subordinates might help them overcome it.

Then, if women in position of leadership within the organization felt comfortable enough in this position, they would also feel more comfortable helping their junior female colleagues. To do so, organizations need more women in positions of power: Not only to break the stereotype that women are less suited to leadership positions, but also to allow women at all hierarchical levels to find other women willing and able to sponsor them at the upper level.

Maybe the most important is that powerful members of organizations understand that they can become allies of women and change the organizational landscape: They can use their power to support, mentor, and sponsor members of traditionally stereotyped and marginalized categories; and use their voice to report and confront discrimination when it happens.


  1. Cullen Z, Perez-Truglia R. The Old Boys’ Club: Schmoozing and the Gender Gap. Am Econ Rev. 2023 Jul;113(7):1703–40.
  2. Derks B, Van Laar C, Ellemers N. The queen bee phenomenon: Why women leaders distance themselves from junior women. Leadersh Q. 2016;27(3):456–69.
  3. Joshi S. Measuring Diversity and Inclusion using Organizational Network Analysis. TrustSphere. 2018.
  4. Women in the workplace. Lean In, McKinsey & Company; 2019, 2020, 2021.
  5. Ibarra H, Carter NM, Silva C. Why Men Still Get More Promotions Than Women. Harvard Business Review. 2010.
  6. Atwater LE, Tringale AM, Sturm RE, Taylor SN, Braddy PW. Looking ahead: How what we know about sexual harassment now informs us of the future. Organ Dyn. 2019;48(4):100677.
  7. Working Relationships in the #MeToo Era, US sample. Lean In; 2018.
  8. Working Relationships in the #MeToo Era, UK sample. Lean In; 2019.
  9. Hewlett SA. Threatened by Scandal, Women Need Support. Harvard Business Review. 2010.
  10. Ziani Z. When Gender Stereotypes Prevent Women from Networking Efficiently.

Found this post insightful? Get email alerts for new posts by subscribing:

Zoé Ziani
Zoé Ziani

PhD in Organizational Behavior