Association for Public Policy Analysis & Management

Although the Equal Credit Opportunity Act passed in 1974, allowing women the right to apply for credit, little research has been devoted to how men and women interact with credit markets. Using individual consumer credit bureau data for a subset of U.S homeowners, we study how the experiences between men and women differ when interacting in the bankcard and auto borrowing markets. Our results show significant differences between male and female credit outcomes with respect to bankcard and auto loan balances, the number of bankcards and auto loan trades, bankcard limits, and frequency of bankcard limits adjustments. We also observe that the gap between male and female borrowers have grown since 2009.

The existing literature on gender differences in borrowing has primarily focused on mortgages, with research showing that women are more likely to be subprime mortgage borrowers (Fishbein and Woodhall, 2006) and to pay more for all types of mortgages (Cheng et al., 2011). However, due to data limitations, there has been relatively little research done on gender differences for other types of borrowing. We are able to overcome this limitation by using linked Home Mortgage Disclosure Act data, McDash Analytics, LLC a wholly owned subsidiary of Black Knight Inc. data, and Equifax Credit Risk Insight Servicing data. The merged data set contains demographic information, including the borrower’s gender, along with traditional credit bureau data attributes. We restrict the data to include only male and female mortgage holders without a co-applicant from the years 2009 to 2016.

Preliminary results indicate that while female homeowners owned more bankcards than male homeowners (2.997 to 2.747), male homeowners’ average bankcard credit limits are significantly larger than females’ limits ($9,777 to $8,764). In addition to the magnitude of the limit, we observe differences in how often bankcard credit limits were modified, with females experiencing more changes than males (8.072 to 7.225). Prior observed differences in bankcard prices and usage within more heterogeneous populations have been attributed to different financial literacy levels between males and females (Mottola, 2013). However, because homeowners are assumed to be a group with greater financial literacy than the general public, that we find gender disparities in two presumably competitive credit markets is noteworthy.

To uncover the factors and composition of the gender disparity, we implement Oaxaca-Blinder decomposition and matching methods to explain these disparities. Our preliminary findings show that disparities in borrowing exist across risk score thresholds, which is consistent with the existing mortgage literature. These empirical results contribute to the gender and credit use literature by further identifying the factors that drive credit disparities among a more financial literate population.