Gender Differences in a Cooperation Game: Evidence from the Field in Matrilineal, Patriarchal and Gender-Neutral Societies

Female borrowers display better repayment behavior than males, irrespective of the roles they play in society and the nature of the loan.

FindingsMethodology

Women make up more than 70% of microcredit clients worldwide. One potential explanation for this gender imbalance is that women have traditionally displayed a better track record of loan repayment than male borrowers. For instance, a study that investigated repayment records for over 350 microfinance institutions around the world found that having a higher percentage of female clients is associated with a lower portfolio risk, fewer write-offs, and fewer provisions.

Gender differences in repayment behavior may be shaped by the distinct roles that men and women play in society and/or the opportunities and constraints they face. For example, if women have less access to sources of credit they may be more likely to repay their loans in order to continue receiving credit. On the other hand, there might be something intrinsic to women driving this difference. This study investigates whether gender roles affect the repayment behavior of men and women.

The researchers in this study compare repayment behavior of men and women who come from two distinct patrilineal and matrilineal societies that are culturally similar. These are the Khasi and the Patro communities from the Sylhet district of northeast Bangladesh. The Khasi have a matrilineal family structure and a matrilocal residence system. Women are the heads of Khasi households, and children are known by their mother’s family names. Only female children among the Khasis inherit ancestral property, and a Khasi husband has no authority over decision-making or distribution of power. The Khasis have lived side by side with members of another indigenous community, the Patro, for hundreds of years. The Patro community has a patriarchal family structure, in which males are considered to be household heads and the inheritors of property.

Researchers played experimental games with men and women from both groups to determine if societal roles influence men and women’s repayment rates.

Findings

Women are better credit risks in both patriarchal and matrilineal societies, irrespective of whether the loan contracts stipulated joint or individual liability.

  • Across the entire sample, the average repayment rate of women was 17 percentage points higher than that of men. Indeed, the authors found that women participants had better repayment rates in every society.
  • Even when researchers controlled for age, education, marital status and assets, females were more likely to repay their loans regardless of their social group.
  • In the matrilineal society, the average repayment rate of women was 12 percentage points higher than the average repayment of men, controlling for demographic information.
  • In the patriarchal society, the average repayment rate of women was 30 percentage points higher than that of men, controlling for demographic information.
  • In the control group, i.e. the gender-neutral society, women were 18 percentage points more likely to repay than men, once again controlling for demographic information.
  • The repayment rate of men appeared to depend on what kind of society they were raised in: the average repayment rate of the men in the matrilineal society was 17.5 percentage points higher than that of the men in the patriarchal society, indicating a greater cooperative spirit in the former group.

Female borrowers display better repayment behavior than males, irrespective of the roles they play in society and the nature of the loan. Men in matrilineal societies demonstrate much more cooperative behavior than do men in patriarchal societies, which results in higher repayment rates.

Methodology

This study was conducted among the Khasi and Patro communities in the Sylhet district of northeast Bangladesh. A total of 280 subjects participated in 28 experimental sessions. Of these, 160 were Khasi (70 male and 90 female), and 120 were Patro (60 male and 60 female). Each session had either 10 male or 10 female subjects belonging to the same community. Subjects were informed they would receive 50 Bangladesh Taka (BDT) for their participation and would earn more money depending on their choices. Subjects participated in two identical repayment games. The first game was an individual loan repayment game, and the second game was a joint-liability repayment game, in which loans were given to groups of five borrowers. In each round, each subject received a loan of 5 BDT, which had to be invested in a project that had a 5/6 chance of succeeding. If the project succeeded, the subject received 12 BDT. If the project failed, the subject received 0. If a project was successful, the subject had the choice to repay the loans with 20% interest or to default strategically. If the subject decided to repay, the game moved to the next round; otherwise, the game ended. In the joint liability loan groups, the group was responsible for repaying 30 BDT.

As a robustness check, the researchers conducted this same game on the Marma indigineous community in the southeastern district of Khagrachori, who are considered to have a gender-neutral society. Their results also held for that group. A total of 80 male and 70 females participated in that experiment and it had the same experimental procedure.