Gendered Network Structures and Careers

Topic & Objectives

Many observers of the initial financial crisis – as triggered by excessive risk taking in the real estate market – have argued for boards to have stronger female participation. There is substantial experimental evidence that women tend to be more risk-averse, less over-confident and less prone to sort themselves into competitive payment schemes that are likely to produce short-termism. Women also tend to take more seriously the tasks of non-executives on boards, engaging in more monitoring activities. However, women are still rarely found in these positions. There are many explanations for this evidence – discrimination, social norms, stereotypes, lack of support for families, or statistical discrimination – and this project investigates an understudied explanation: the role of social networks.

In a first paper, we investigate the impact of social networks on earnings using a dataset of over 20,000 senior executives of European and US firms. The objective of this paper is to understand whether women obtain lower returns in terms of earnings from their social networks, compared to men. Because the social network information in this dataset comes from participation in joint activities, we can only study the effect of networking opportunities on earnings without being able to pin down the effect of actual investment in social interactions.

In a second paper, we therefore investigate whether men and women build their social networks differently. A first study confronts students with an experiment in which links are created by investment in simple trust games. The hypothesis is that women tend to invest less in relationships with people they do not know and, upon building successful relationships, are less likely to experiment with other potential partners. In a second study, we monitor the “real” social network emerging among students with the hypothesis that women have smaller but deeper networks.

Key Findings

SAFE Working Paper No. 123

• We show that the statistical association of networks with women's remuneration is significantly weaker than for men. Those firms that are relatively supportive of the most talented women tend to give them access to substantially smaller networks than are available to both male and female employees of firms with more traditional recruitment policies.
• We also find evidence of "window-dressing" by firms that appoint women to non-executive directorships without doing anything to support talented women in other ways. This suggests that policies designed to increase female board representation may do little by themselves to increase the representation of women in positions of executive power.
• These findings suggest the persistence of subtle forms of discrimination at the highest level in executive labor markets. However, whether the causes of such diminished visibility within corporate networks consist solely of such discriminatory behavior on the part of employers, or also of differential risk aversion on the part of some women, is difficult to judge on the evidence available here. It may be that the preferences and behaviors of women interact with those of men in complex ways.

SAFE Working Paper No. 168

• Our experiment shows that women are more selective in forming networks and only half as reactive as men to information about the economic benefits of interacting with different individuals.
• Subsequent formation of real social connections is consistent with the experimental results: being assigned to the same introductory group had a much larger positive effect on women's likelihood to report a subsequent friendship.
• If men are more opportunistic about the formation of networks, they may benefit in two distinct ways: they may invest more in sustaining weak links in their social networks, and they may be more likely than women to call in favors from their casual acquaintances when looking for new employment. These behaviors are likely to make them more visible in their professional network – and in particular in high-skill professional networks – which might in turn generate career benefits.