Growing evidence suggests that women are more likely to repay collateral-free microloans than men. However, we know little about what explains such gender differences. We hypothesize that better repayment performance of women microcredit borrowers can largely be explained by gender differences in innate trustworthiness. We conduct a trust game and a microloan repayment game in rural Bangladesh. We find that women are more trustworthy than men and that they are more likely to repay their loans irrespective of any control mechanisms, such as joint liability or dynamic repayment incentives. The results of a mediation test suggest that the gender effect on loan repayment is significantly mediated by differences in innate trustworthiness. We conduct a sensitivity test to check the extent to which unobserved confounders might have influenced the mediation effect, and find no evidence of significant omitted variables bias.
Trust refers to the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will not behave opportunistically (Berg et al., 1995; Mayer et al., 1995; James, 2002). Trustworthiness, on the other hand, refers to the innate personal characteristics of an individual reflecting her or his preference to reciprocate to the act of trusting in the absence of any economic incentives. In a lending relationship, the lender shows trust by accepting the risk that the borrower may strategically default even if the project yields adequate return to repay (Jaffee and Russell, 1976). Under asymmetric information, the borrower shows trustworthiness by repaying a loan (Saparito et al., 2004; Becchetti and Conzo, 2011). The trust-repayment relationship in microcredit has been investigated by several authors. Karlan (2005) conducts a trust game with the borrowers of FINCA (Foundation for International Community Assistance) in Peru and finds that trustworthy individuals are more likely to repay their loans. Cassar et al. (2007) and Cassar and Wydick (2010) conduct laboratory experiments in Armenia, Guatemala, India, Kenya, South Africa, and the Philippines. They also find a positive association between trustworthiness and microloan repayment decision. There are, however, two important gaps in this literature.
First, the role of borrowers’ gender in the trust-repayment relationship has not yet been studied. Biological research suggests that trustworthiness is positively associated with the level of oxytocin release in a person (Kosfeld et al., 2005; Zak et al., 2005), and that the magnitude of oxytocin release is significantly higher among women than men (Carter, 2007). Evolutionary psychologists argue that gender-differentiated behaviors have roots in the evolutionary past (Wood and Eagly, 2002). Ancestral women competed with other women to attract long-term mates to protect the future of their offspring, which evolved dispositions such as risk-aversion and trustworthiness. It is therefore not surprising that women have shown more trustworthiness than men in most of the experimental trust games (Croson and Gneezy, 2009). We argue that higher trustworthiness translates into better repayment behavior by women borrowers. The study that is closest to ours is by Aggarwal et al. (2015), who report that MFIs are more likely to target female clients in countries with a low level of social trust. Social trust refers to the extent to which the members of a society can be trusted (Guiso et al., 2004). Thus, the targeting of female clients by MFIs is a substitute for low levels of trust in society. This conclusion is based on an untested premise that better repayment performance of female microcredit borrowers can largely be attributed to their innate trustworthiness. We formally test this proposition.