Women are less likely to occupy the top paying jobs in developed economies, in part because they are less competitive than men. A whole series of laboratory experiments has detailed the gap in competitiveness between the average woman and the average man, even when women are just as good, if not better than men. Is this result due to the fact that women are biologically female, or the fact that they are socialized as female? Although we often alternate between gender and sex in describing males and females, they are not strictly the same. OED defines gender as “the state of being male or female (typically used with reference to social and cultural differences rather than biological ones).” Thus, gender is a cultural definition that is linked to—but not synonymous with—the biological definition.
Recent work of mine (with Jeffery Flory and John List) illustrates this distinction and its importance in development economics. We show what appear to be strong effects of biological events (first menarche and menopause) on female behavior, but the effect in one culture is almost the opposite of the effect in another culture. Specifically, in traditional partriarchical cultures in Malawi, women increasingly avoid competitive situations as they move from childhood to puberty, but are as competitive as men after menopause. In contrast, in matrilineal cultures in Malawi (where women inherit property and households form around daughters, not sons) women become more competitive in puberty and continue to seek competitive situations for the rest of their lives.
Our work starts with the hypothesis that women are, in at least some respects, psychologically different from men; a biological distinction. One apparent manifestation of this psychological difference is the reluctance of women in most places in the world to enter competitive settings, even when they are good at what they do. We see this in economic laboratory experiments by presenting men and women with an unfamiliar task and then offering to compensate them according to their ability in this task either in isolation, or in comparison to other people. If they choose to compare their performance to others and they are better than three randomly selected other people they earn four times as much per success than if they choose to work on their own. Significantly more men than women choose to compete and this result stands up when we control for ability and risk aversion. It seems that men like to compete and women do not, even when they are pretty sure they will win.
This result has been used to explain part of the differences in labor market outcomes in the US and Europe, in particular, why there are so few women in the most competitive jobs. The idea is that women are innately less likely to enjoy working in competitive situations and therefore avoid them.
Our work took a different tact, looking to see what parts of the psychological differences could be ascribed to biology and what parts could be assigned to the cultural construct of gender. Our study took place in Malawi, which offered us two advantages over other work on competitiveness and gender. First, we studied men and women between the ages of 12 and 90, whereas most studies looked only at college students (which makes sense for a study of career choice, but not for a study of biology and gender). Second, within Malawi there is significant variation in cultural attitudes towards gender. In particular, many villages in Malawi practice matrilineal inheritance and matrilocal marriage, in which property follows the female lineage and men leave their natal villages to join their wives’ households. In contrast, in patriarchal cultures men inherit property and stay in their natal villages.
Thus, we have what should be significant variation in biology (across age) and what might be significant variation of the social construct of gender. Consistent with the biological story we find important behavioral changes for women at the age of 15 (average age of menarche in Malawi) and at 50 (average age of menopause). In patriarchal societies, we find that women are less likely to compete as they move through puberty, but are more likely to compete after menopause. This could easily be interpreted as a biological response; hormone levels condition the female brain and change behavior. Indeed Wozniak et al (2010) find that women’s preferences for competition change over the course of the normal monthly hormone cycle.
In matrilocal societies, we find that menarche and menopause are also important, but in almost the opposite direction. Girls become more competitive as they go through puberty and only slightly more competitive as they move through menopause. The same hormones are present in women from both cultures but the behavior they encourage or inhibit changes with culture. Becoming a “girl” or “woman” is equally important in each culture, but what it means to be a “girl” or “woman” is different.
I showed these results to John List and he asked, “How do we randomize the assignment of gender?” It isn’t a silly question or even a hope for medical advances. Sex is assigned by fate and hormone levels follow a well-documented deterministic progression, but gender is assigned by culture. What we need to know is how cultures are doing it and then we can test the impact of assigning gender differently to otherwise similar groups of individuals.
Why we might want to do this is a bit more complicated. It is relatively easy to describe competitiveness (after controlling for ability and risk aversion) as a type of self-confidence, over-confidence or comfort with ambiguity (the opposite of ambiguity aversion). These traits, in turn, have been described as important to success in entrepreneurial activities; it seems that being over-confident in the short term improves the chances of long run success in competitive environments. Individual entrepreneurship is one of the many qualities that can propel development. Thus, there is a trail of evidence linking competitiveness and entrepreneurial ability and if culture can make women want to be competitive it should be able to make them good entrepreneurs.
Thus, we need to design experiments in which some characteristic of gender is randomly assigned to people, and the impact of this assignment can be measured. In the short term we might want to measure it using laboratory experiments like those used in Malawi but in the long run we want to know if this assignment makes them better entrepreneurs. There is good evidence from these cultures in Malawi that initiation ceremonies are an important part of assigning cultural roles. What we do not know is whether the initiation ceremonies are teaching girls about the reality of the economic future they face (they own the land, for example) or whether they are teaching them specific behaviors (they should take advantage of their ownership rights, for example). In the first case, the economic reality is the true driver of behavior (and much harder to change) whereas in the second, initiation itself can teach behavior.
Ken Leonard is an associate professor in the Department of Agricultural and Resource Economics at the University of Maryland. His webpage is here.