Is There a Gender Gap in Getting Home Equity Loans?

In a country with a demonstrated gender pay gap, the ability to buy a home, build home equity, and access credit such as a home equity loan is also fraught with discrimination and difficulty. Whether this is attributable to an inequitable economic landscape or to discriminatory behavior and assumptions on the part of lenders is difficult to pin down. 

More likely, it’s a complex interplay between the institutional and the individual. Exploring the history of anti-discrimination legislation in mortgage lending as well as the current gap in home lending data can perhaps elucidate further.

A Brief History of Anti-Discriminatory Mortgage Lending Legislation

The practice of redlining, which generally refers to discriminatory lending behavior based on race and ethnicity, has been well-documented throughout recent history. In 2016, the term “pinklining” was coined to describe the financial sector’s systematic exploitation of women. But even before it had a name, gendered discrimination in the lending landscape was rampant and structural.

As recently as the 1970s, women were not allowed to sign a loan in their own name and had to either find a co-signer or sign under their husband’s name if they were married. It was also common not to count a working woman’s portion of a family’s income if she applied for a mortgage with her husband. Single women also faced many difficulties in getting loans approved.

Several pieces of anti-discrimination legislation came into play in the late twentieth century. The Fair Housing Act of 1968 came about as part of the Civil Rights Act and outlawed discrimination in housing-related purchasing, renting, and lending activities.

In the 1970s, the Congressional finding that lenders were often denying credit applications on the basis of some sort of discrimination (be it gender, race, marital status, religion, or age, to name just a few) fueled the creation of the 1974 Equal Credit Opportunity Act (ECOA).

The Community Reinvestment Act (CRA) followed in 1977 and was meant to address redlining and other anti-consumer and anti-business policies and behaviors. By systematically refusing to lend to minorities or invest in low-income areas, financial institutions were denying basic credit needs that could have contributed to the growth and success of their people and neighborhoods. Therefore, the CRA supposedly “requires the Federal Reserve and other federal banking regulators to encourage financial institutions to help meet the credit needs of the communities in which they do business, including low- and moderate-income (LMI) neighborhoods.”

Unfortunately, there has been much indication that these laws are imperfect and insufficient in preventing discrimination. For example, a comprehensive 1996 legal paper argues that the “responsibility for enforcing fair-lending and anti-insurance discrimination laws” should not “fall on the shoulders of federal and state agencies” as “courts, themselves, allow race, gender and other impermissible variables to influence both procedural and substantive rulings”.

Meanwhile, a 2009 report by the National Council of Negro Women and the National Community Reinvestment Coalition found that although many cases of discrimination had been reported to the Federal Reserve Board under the CRA, enforcement and follow up has been lacking from the governing body that is meant to carry out further investigation.

The Gender Wage Gap and Women’s Access to Credit

Of course, the gender wage gap also impacts women’s access to wealth and credit, going beyond a mere difference in earnings. Due to predatory lending practices (such as steering women towards subprime or predatory credit) and barriers in accessing fair credit, women, and women of color, in particular, face disparities in overall net worth, not just income. This inevitably negatively impacts their ability to build credit.

Since a home equity loan is a type of secured installment loan in which the borrower uses their home equity as collateral, the amount of home equity a borrower has will partially determine the potential loan amount, as well as the interest rate. Usually, the maximum amount one can borrow is equal to about 80% of their home equity.

Unfortunately, when it comes to home equity, single women build less over time than their male counterparts (calculated as 92 cents to the dollar as compared to single men). A 2017 study by the real estate company Redfin found that women tend to pay a smaller down payment on a home than men, which in turn affects their ability to build home equity.

Women are also more likely to overpay on a mortgage despite having a better track record of paying them off or being denied a mortgage in the first place. These facts alone may arguably contribute to a gender gap in the rate of women vs men who are able to secure a home equity loan (as well as the amount they are able to secure), as it certainly points to an uneven financial foundation underpinning the credit market.

Parsing the Data: The Home Mortgage Disclosure Act and Data Collection

Another piece of legislation meant to combat discrimination in the housing market is the Home Mortgage Disclosure Act (HMDA). The Act allows for the collection of data on loans that is made publicly available in the interest of transparency and accountability. While data collection does, in theory, make it possible to diagnose gendered and other discrimination within the housing industry, there are some limitations to the HMDA data.

Firstly, gender is a spectrum comprising a range of identities that can be fluid and dynamic. This is not captured by the HMDA data, as it presents gender in the binary categories of “male” and “female”. The other two data fields are “joint”, to designate co-applicants of mixed gender, and “sex not available.”

The HMDA data is also murky when it comes to home equity loans as it has historically excluded home equity loans that were taken out to consolidate credit card debt, or to pay for medical expenses.

As of 2018, after a large update to the HMDA data and an overhaul of the categories under which data was to be collected, the categories for “Loan Purpose” are as follows:

• Home Purchase
• Home Improvement
• Refinancing
• Cash Out Refinancing
• Other Purpose
• Not Applicable

In 2018, the category of “Other Purpose” contained around 502,000 loans that would not even have been reported before 2015.

Another important consideration in identifying and fighting discrimination is the intersectionality of race, class, gender and other aspects of identity. For example, researchers have identified “an elevated risk of vulnerability to high-cost lending among women of color” in the mortgage market. They have also identified historical inadequacies in the HMDA dataset, for example, missing information when it comes to race that implies a systematic bias that undermines fair lending.

In 2018, more ethnicity fields were added to the dataset, as well as racial subcategories that didn’t exist within the database previously. A new reporting category was even added to designate whether the race, ethnicity, or sex of an applicant was determined by visual observation or surname on the part of the data reporter. Although these changes are a step in the right direction for creating a more intricate and complete picture of the lending landscape and combating discrimination within it, the HMDA data is not perfect.

For starters, the data reporting process is highly nuanced and complex, and relies on the attention to detail, compliance, oversight, consistency, and objectivity on the part of the staff working at the applicable financial institutions. Unfortunately, this is not always a given. It’s also worth underscoring the Federal Reserve’s own words on HMDA: it is “a disclosure law that relies upon public scrutiny for its effectiveness. It does not prohibit any specific activity of lenders.”

Towards Better Data

In looking at the HMDA dataset, it would seem that more intricate and complete information gathering would be helpful in drawing conclusions as to whether a gender gap exists for specific types of loans, such as home equity loans. For example, not excluding certain loan purposes or lumping them into an “other” category could help researchers in painting a more complete picture.

Another arena in which data collection is central for preventing discrimination is AI and fintech. Historically, the approach to preventing discrimination has been to strive for neutrality – to remove the consideration of gender in a credit application, for example. This mentality carried forward from 20th century legislation such as the ECOA into the building of “gender-blind” machine learning algorithms and models.

However, researchers have come to the counterintuitive conclusion that factoring gender into algorithms actually improves their fairness. Otherwise, having been fed the input of historical lending data, AI is likely to reproduce the same discriminatory patterns that have been perpetuated historically – even if gender is not specified, as there are other correlating factors that create bias.

Given current anti-discrimination legislation such as the ECOA, creating models that differentiate based on gender is illegal. However, changing the law to accommodate for this type of AI, could open up the possibility of further discrimination.

All this is to say that while a useful tool in combating discrimination, data is not the be-all-and-end-all solution. In considering the HMDA data together with recent issues in the field of financial AI, it’s evident that although better data can help with preventing discrimination, it’s up to researchers to interpret it, intervene when necessary, and use it to advocate for the type of change we want to see. The same human element that creates patterns of discrimination can also prevent it. 

The Bottom Line

Historically, gendered discrimination has been a well-documented issue in the American credit industry, and legislation has been created in an attempt to prevent it. Although it’s obvious from related statistics that inequity remains in many aspects of home lending, unfortunately, it’s difficult to prove from the current data alone whether there is a gender gap in getting home equity loans. Better data collection could help in answering this question, but there is a responsibility to record and use this data responsibly in order to ensure better policy for the future.