Resumen

This paper examines whether women leaders behave differently when they believe they are elected by a gender quota or by merit and if their behavior is affected by group gender composition. We conduct a laboratory experiment in which participants decide whether to invest or not in a project that presents coordination failures. Leaders can send a risky signal (their investment) to persuade the other members of their group to invest in the project. The group can attain the cooperative equilibrium if the leader is followed. We provide experimental evidence that when investing is costly, the behavior of women leaders is affected by both the way they are elected and the gender composition of their group. We find that gender quotas do not have a signifficant effect on cooperation but meritocracy triggers selfish attitudes of female leaders when they face all-male or mixed gender followers, undermining cooperation. We argue that group identity may be the underlying explanation of all our findings. When investing represents a social dilemma, leaders elected by merit don't identify themselves with their followers, they believe they are superior and reduce their investment. However, gender identity seems to offset this behavior and reconnect leaders with followers, as the negative effect of meritocracy does not hold when female leaders are facing female followers.