Does gender matter in banking? If gender is associated with differences in screening and monitoring abilities, perhaps loan officers’ incentives should reflect gender. Some analysts have suggested that bankers’ gender played a role in the global crisis (Sibert 2009). More broadly, researchers have recently begun to document gender differences in a wide range of fields – everything from performance in high-pressure sports (Paserman 2007) to labour force participation by recent immigrants (Blau et al. 2008).
This column reports on recent research examining the relationship between borrowers’ and loan officers’ gender and the riskiness of loans, providing direct evidence for the first time that gender matters in banking.
How gender might matter
Theory and intuition provide ambiguous predictions of how the gender of the loan officer might be related to the likelihood that their borrowers incur arrears.
Bharat et al. (2009) argue that male executives have better access to private information, which in the case of loan officers could result in a superior performance of male loan officers. Further, consider that in patriarchal societies male loan officers might have a stronger standing vis-à-vis (male or female) borrowers in terms of monitoring and disciplining them, thus ensuring loan repayment. Hence, we would observe a lower riskiness of loans screened and monitored by male loan officers.
Several studies, however, have shown that women are more risk-averse than men (e.g. Charness and Gneezy 2007). Thus, it may be that loans handled by female loan officers are less likely to become problem loans because women grant loans more restrictively. Other studies (e.g. Barber and Odean 2001) argue that men are more overconfident than women. With respect to loan officer performance, this could result in male loan officers screening and monitoring more loans than would be optimal, eventually leading to a superior performance of female loan officers. Another hypothesis in favour of a superior performance of female loan officers is that female loan officers have typically fewer outside options in the labour market and therefore have stronger incentives to excel in building their loan portfolio.
Yet another hypothesis is that loan officers might have an easier time monitoring and disciplining borrowers of their own gender, in which case we would expect to find a lower arrear probability of female borrowers if the loan is approved and monitored by a female rather than by a male loan officer and vice versa.
Given the ambiguity of theoretical predictions, it is an empirical question whether men or women are better loan officers and whether this effect varies with the gender of the borrower. In recent work (Beck, Behr and Güttler 2009), we use a unique loan-level data set for an Albanian bank over the period 1996 to 2006 to test whether male or female loan officers experience a lower arrear probability on their loans.
Data from Albanian lending
The data set contains both rejected and accepted loan applicants from a bank serving individuals and small- and medium-sized enterprises in Albania. We have information on over 43,000 loan applications and 31,000 loans given by the bank from January 1996 to December 2006. In addition, our data set contains information on 203 loan officers and covers five branches of the bank in the Albanian capital, Tirana. In the sample period, the lender granted mainly loans for business, housing improvement, and consumption purposes. While it clearly focuses on the low-income and small-enterprise segment, the bank’s primary goal is financial sustainability and therefore profitability.
We define a loan as being in arrears if at least one of a borrower’s payments was in arrears for more than 30 days at any point over the whole lifetime of the loan.1 We also control for an array of borrower-specific, loan-officer specific, and loan-specific information to rigorously assess the relationship between loan officers’ gender and loan performance.
For our baseline regression analysis, we restrict and cut the data in several ways. First, we restrict our attention to actual borrowers and their arrear likelihood and thus drop unsuccessful loan applicants. Second, we focus on first-time borrowers so that our comparison is consistent, as all loan officers have the same limited information about the respective borrower at the time of the application.2 Loan officers already have historic information about repeat borrowers that they can take into consideration when granting and monitoring the loan. Focusing on the first loan by each successful loan applicant thus allows us to cleanly study gender-specific loan officer performance differences.
The main findings
Our main findings can be summarised as follows:
- Loans by female borrowers and borrowers screened and monitored by female loan officers are less likely to turn problematic.
- Controlling for the interaction between borrower and loan officer gender, we find that female loan officers experience a lower share of problem loans both in the case of male and of female borrowers, with the effect being economically stronger and more robust for female than for male borrowers.
- The effects are much stronger and more robust for first loans than for subsequent loans to the same borrower. We interpret this as evidence in favour of a gender effect, as the effect should fade away with subsequent loans as loan officers learn about repeat borrowers.
- The effect is economically significant. Loans by female borrowers screened and monitored by female loan officers have a probability of turning problematic that is 4.6 percentage points lower than if handled by male loan officers, with the corresponding difference for male borrowers being 4.5 percentage points. The overall probability is 13.5%.
The gender performance gap – exploring the mechanisms
We explore different mechanisms and channels that could explain the gender advantage that female loan officers experience vis-à-vis their male colleagues in terms of loan performance.
- We do not find that female loan officers get ex ante less risky borrowers. We establish this by estimating the likelihood that loan applicants are screened by a female loan officer. We then compare the characteristics of loan applicants screened by female loan officers with borrower characteristics that make their loans more likely to become problematic, according to our main regression analysis.
- Female loan officers are not more restrictive in the screening of loan applicants than male loan officers, at least not based on observable applicant characteristics. Estimating the likelihood that a loan applicant is accepted does not yield a significant coefficient on the female loan officer dummy, neither for male nor for female borrowers. This suggests that women are not more risk-averse based on observable borrower characteristics.
- Female loan officers are not better because they are more experienced. Female loan officers are, on average, roughly two years younger than their male counterparts, which can be explained by the compulsory military service that male Albanians have to do. In addition, the performance advantage of female loan officers holds for different levels of experience, not just high levels of experience.
- Female loan officers do not have a lower workload than their male colleagues, suggesting that our findings cannot be explained by overconfident male loan officers taking on too many borrowers.
These four tests leave us with the tentative conclusion that the better performance of female loan officers is explained with better screening capacities on unobserved borrower characteristics (smell test) and/or better monitoring capacities. One plausible explanation for this finding may be that female loan officers typically have fewer outside options in the labour market and are less mobile, giving them stronger (intrinsic) incentives to build loan portfolios with fewer problem loans. Our finding is also consistent with female loan officers being more risk-averse with repercussions for their screening.
Gender indeed seems to matter in banking. Loans handled by female loan officers have a lower probability of turning problematic than loans handled by their male colleagues. It also contributes to the behavioural banking literature, underlining the importance of gender when designing proper incentive schemes for loan officers.
Agarwal, S., Wang, F.H., 2008, “Motivating Loan Officers: An Analysis of Salaries and Piece Rates Compensation”, Working Paper, Federal Reserve Bank of Chicago.
Barber, B.M, Odean, T., 2001, “Boys Will Be Boys: Gender, Overconfidence, and Common Stock Investment”, Quarterly Journal of Economics 116, 261-292.
Beck, T., Behr, P., Güttler, A. 2009. "Gender and Banking: Are Women Better Loan Officers?” CEPR Discussion Paper 7409.
Bharath, S.T., Narayanan, M.P., Seyhun, H.N., 2009, “Are Women Executives Disadvantaged?”, Working Paper.
Blau, F., Kahn, L., Papps, K., 2008, “Source Country Characteristics and the Labour Market Assimilation of Immigrant Women”, VoxEU.org, 20 December 2008.
Charness, G., Gneezy, U., 2007, “Strong Evidence for Gender Differences in Investment”, Working Paper.
Hertzberg, A., Liberti, J.M., Paravisini, D., Forthcoming, “Information and Incentives Inside the Firm: Evidence from Loan Officer Rotation”, Journal of Finance.
Paserman, D., 2007, “Gender-linked Performance Differences in Competitive Environments: Evidence From Pro Tennis”, VoxEU.org, 26 June 2007.
Sibert, A., 2009, “Why Did the Bankers Behave So Badly?” VoxEU.org, 18 May 2009.
1 For robustness we also use time periods of 15, 60, and 90 days in the empirical analyses.
2 This rests on the reasonable assumption that loan applicants and loan officers did not know each other before the loan application was forwarded.