Here are two papers from my project with Marie Lalanne.
Here is a major update of our earlier CEPR paper
We investigate gender differences in the impact of social networks on earnings using a dataset of career histories of over 22,000 senior executives and non-executive board members of European and US firms. There is a large positive impact on men’s salaries of the number of currently influential individuals they have previously worked with, while the impact on women’s salaries is significantly weaker, and statistically distinguishable from zero only in certain years. These findings hold also for non-salaried remuneration. Using a placebo measure of individuals who were employed in the same firm at different times we show that our network measures reflect genuine connections and not merely unobserved individual characteristics. We also demonstrate that individuals who are relatively central in the network thereby benefit, and that women benefit from having networks composed of other women. In contrast to executives, non-executive board members do not display systematic gender differences in the effectiveness with which they leverage their links into remuneration. We explore possible mechanisms, and note that the firms which do most to integrate women into positions of executive power appear to rely less on networks for recruitment
The second paper is joint with Guido Friebel, Marie Lalanne, Bernard Richter and Peter Schwardmann:
We test two hypotheses, based on sexual selection theory, about gender differences in individual choices with respect to social interactions requiring investment (of time or economic resources). The differential selectivity hypothesis predicts that women invest less than men in an interaction with a new partner, other things equal. The differential opportunism hypothesis predicts that women’s investment in a social interaction is less responsive to information about the likely economic payoff to that investment. Both hypotheses, if true, imply important differences in the formation of social networks by women and men. Two cohorts of a total of 363 students were matched randomly over two rounds with a partner to play a trust game. In the second round of the trust game they also had the chance to invite a new partner to play. We find evidence in favor of both hypotheses. In particular, women invest less in new partners in both rounds, and invest even less in a framing treatment that reminds them of the need to reflect on the decision. They also react less elastically to their a priori beliefs about the likely returns to their investment, and to information that is revealed at the beginning of the second round about the return to the amounts sent to their previous partner.