In popular discussions on the moral dimensions of the financial crisis, there is sometimes reference to a gender dimension. A striking fact, for example, is that those responsible for the bankruptcy of the
In order to address just one dimension of this, I am in the process of finalizing public good experiments with male and female students in computer labs in The Netherlands. We try to find out not whether women behave at higher moral standards than men, but whether men and women believe that women behave more cooperatively than men. The literature on gender differences in experimental economics has already widely confirmed that women do behave more cooperatively than men and that women are more risk averse than men. These two results would suggest that more women in the financial sector may have led to less risky security packaging and less risky sales of mortgages and more win-win outcomes instead of predatory outcomes led by self-interested bankers and financial traders. Our question is whether the stronger cooperative attitude of women is recognized by both sexes and also leads to more cooperative outcomes of behaviour. Our preliminary findings, which are just based on one set of experiments and are currently being followed up by two more sets, are the following. First, we find that female subjects still behave more cooperatively than men, even when corrected for differences in risk aversion. Hence, their higher cooperativeness can not entirely be attributed to their lower risk taking than men. For the financial sector, this implies that when female bankers and traders just take as many risks as male traders they still go for more win-win results as compared to their male colleagues. Second, we find that women who are given the information that the other players are also female cooperate more than women and men who do not receive information about the sex of their partners. At the same time, we found that male players do not cooperate more when being informed that their partners are female. This may imply that when a sufficiently large number of women (at least a majority) dominates the financial world, behaviour may becomes less predatory and more cooperative, next to, on average, less risky. In a recent overview of gender differences in experimental economics, Rachel Croson and Uri Gneezy in the Journal of Economic Literature argue that women’s decisions are more context-specific than men’s. For example, they conclude from studies on the ultimatum game and the prisoner’s dilemma game that women’s decisions vary more with the gender of their partner than men’s decisions, indicating that also gender context is more influential for women as compared to men. But they offer no explanation why this would be the case.
Our explanation of the gender differences that we found in our public goods experiment is based on the literature in sociology, gender studies, and social psychology on gender beliefs. Gender beliefs can be defined as different interpretations and expectations about the personality traits (how women and men are) and behaviour (how women and men behave) of men and women. Moreover, gender beliefs are not only descriptive, but also prescriptive, stating how women and men should be and should behave. The differences in gender beliefs about men and women can be summarized around achievement-oriented traits for men – agentic traits – and service-oriented traits for women – communal traits. Both descriptive and prescriptive dimensions of gender beliefs contribute to individual self-definitions as masculine or feminine, and operate at the interpersonal level. Gender beliefs contribute to individuals’ definitions of their self-schemas, social identities, and self-evaluations and they operate in the interpersonal domain, defining the behaviours that are appropriate to various social contexts. The empirical literature on gender beliefs has widely demonstrated that both men and women hold gender beliefs. In particular, they both believe that men are more agentic and should behave more agentic than women, and that women are more communal and should behave more communal than men. Men tend to have st6ronger gender beliefs than women. Social dominance theory and expectation states theory state that because men tend to have on average a higher socio-economic status than women and they want to preserve that advantaged position, it is in their interest to hold on more strongly to traditional stereotypes about gender roles and traits compared to women. Status beliefs may even interact with gender beliefs so that even when individuals do not endorse dominant status beliefs, their recognition that these beliefs are widely shared will lead them to assume that others will treat them according to those beliefs, which will in turn affect their own behaviour in a stereotype way.
Gender beliefs are rather essentialist of character, in the sense that they change only very slowly and are often reproduced in new contexts, for example becoming attached to newly emerging jobs in the labour market. This resilience of gender beliefs may signal an evolutionary origin. The dominant evolutionary explanation is through sex selection theory, which holds that men are by nature more competitive because, having many sperm, they want to maximize their number of offspring, whereas women, having only a few eggs, are more cooperative so they seek to maximize the quality of their offspring. This would explain why males exhibit more dominance behaviour than females, both among animals and among humans, and they compete with each other for access to females. Moral gender differences, hence, seem to have an evolutionary origin. Recently, however, sex selection theory has received serious criticisms, within biology, psychology and the social sciences. Biologist Joan Roughgarden has shown internal inconsistencies in
The experiments that I and my co-authors are running do not allow us to distinguish between nature and nurture as explanations for gender beliefs. But the literature and our findings suggest that social context rather than biology seems to be a more convincing explanation. As indicated by Ridgeway, the interaction between status beliefs and gender beliefs through expectation status theory is especially likely in cooperative, goals-oriented contexts in which group status beliefs become salient. A public goods game centres around cooperation, a typical feminine trait according to general gender beliefs. Hence, the task in the experiment was not gender-neutral but positively linked to a feminine gender belief, which allowed for the expression of behaviour in relation to the intrapersonal level of gender beliefs: one’s own behaviour independently of others. The other context provided in the experiment was information about the sex of one’s partners, which allowed for the expression of behaviour in relation to the interpersonal level of gender beliefs: one’s behaviour in relation to one’s expectation of the behaviour of others. So, in our experiment we had a cooperative context (task), which is in line with a feminine gender belief, and a gender context provided as treatment variable (information that the other players are female or male). This experimental setting allows us to explain the results in terms of interpersonal gender beliefs and intrapersonal gender beliefs. On the first finding, that the female subjects behaved more cooperatively than men when corrected for differences in risk aversion, this can be explained by intra-personal gender beliefs. Female players believe that they are more cooperative than men, and hence they act more cooperatively than male players who believe that they are less cooperative than women. On the second finding, that women who are given the information that the other players are also female cooperate more than women and men who do not receive information about the sex of their partners, this can be explained by inter-personal gender beliefs. Female players will cooperate more when they are given the information that their partners are also female, because they are socialised into cooperation with other women. At the same time, we found that male players do not cooperate more when being informed that their partners are female. This can be explained on the one hand through socialisation: men have less personal experience with women’s stronger cooperativeness, and on the other hand through the interaction of gender beliefs and status beliefs, which lead men to reassert their higher status by not adopting stereotype feminine behaviour in interactions with women.
In conclusion, our experiments on gender beliefs in a public good game do suggest that with a significant majority of women in the financial sector, some of the damaging behaviour that helps to explain the current financial crisis would likely to have occurred less than with the current male dominance in the financial sector. But this finding seems to be more a result of a social origin of gender differences in morality than of biological differences in the morality of men and women. The good news about this is that men can learn to behave more cooperatively, to take less excessive risks, and to attach less value to a competitively-defined type of masculinity. This may save us some financial crisis in future, but may also help to achieve more win-win outcomes in other sectors of our economy and society…