by James A. Bacon
Oh, brother, here we go again… Of the 14,700 mortgage applications submitted by black Virginians last year, 11.9% were turned down, reports the Virginia Mercury. By contrast, of the 70,400 applications from non-Hispanic whites, only 5% were rejected. The difference in acceptance rates cannot be attributed solely to differences in income, says the online publication. Racial disparities in loan denial rates exist at almost every income level.
The Mercury article does not state explicitly that the gap in rejection rates is attributable to racism, bias or discrimination, but it frames the issue as if it is. The article provides this assessment from Alex Guzmán, director of fair housing for the nonprofit group Housing Opportunities Made Equal of Virginia:
In light of the protests triggered by the death of George Floyd under the knee of a Minneapolis police officer, the latest [Home Mortgage Disclosure Act] numbers take on added meaning because “homeownership is probably the most powerful way to rectify the racial wealth gap.”
And this from Bruce Whitehurst, president of the Virginia Bankers Association:
Speaking for myself as a non-Hispanic White, we have to think a lot more about why the differences are there. I just think we’re at this place where we need to address the structural issues. And I think it’s fair for anyone in the Black community to say, ‘What took you so long to figure out what we’ve known for a long time?’”
As always with racial disparities = racial discrimination stories, there is a lot less to these numbers than meets the eye. Indeed, if there is a guilty party involved, it is the media, in this case the Virginia Mercury, which cherry picks data to push a left-wing Oppression Narrative that can serve only to inflame African-Americans’ sense of grievance and victimhood, and, thereby, discourage blacks from seeking mortgage financing!
First, there’s the old trick of presenting the cherry-picked data in a way that casts the U.S. banking industry in the least favorable light: 11.9% of blacks were rejected compared to only 5% of whites. By that reckoning, blacks were rejected at twice the rate of whites! A more optimistic way of framing the narrative would note that 89.1% of blacks’ mortgage loan applications were accepted, which was 94% of the rate (95%) at which whites’ loans were accepted.
But there’s a much bigger flaw in the article’s reasoning. Income is not the primary factor banks look at when approving mortgage loans. Here are the factors listed by SmartAsset.com:
- The size of your down payment. Industry standards say that homebuyers applying for conventional mortgages should put down at least 20% of the loan amount. A higher down payment reduces risks for the lender and increases the odds of acceptance.
- Credit history. Mortgage lenders often look at FICO credit scores as an indicator of how well borrowers manages their money and the likelihood that they will repay a loan.
- Work history. Mortgage lenders look at applicants’ employment histories. Demonstrating steady employment and sources of income will improve an applicant’s chances of being approved.
- Debt-to-income ratio. Lenders want to know how much student loan debt, credit card debt, and other debts a borrower has. Lenders tend to avoid lending money to applicants with DITs above 43%. “That’s because lenders want to ensure that borrowers can make all of their monthly payments without overextending themselves.”
- The type of loan. There are many kinds of loans. Some loans are bigger, some smaller. Some have 15-year amortizations, others have 30-years. Some are fixed, some are variable. Each type of loan has a different risk profile.
Comparing mortgage acceptance rates of different racial/ethnic groups without accounting for the size of loans, the size of down payments, credit history, debt-to-income ratios, and the type of loans is a meaningless exercise. Adjusting acceptance rates for income is likewise a meaningless exercise. Insinuating the existence of structural racism by citing differences in mortgage-rejection rates between whites and blacks without adjusting for those factors is reckless and irresponsible.
There is one more reason why this article fails abysmally to support the Oppression Narrative. The author posits no mechanism by which the implied discrimination takes place. An increasing percentage loan originations now are approved over the phone or online, not in face-to-face meetings in which an applicant’s race can be identified. There’s an old joke of a dog sitting behind a computer and talking to a canine friend, who says, “The best thing about the Internet is that nobody knows I’m a dog.” Perhaps it can also be said that on the Internet, nobody knows if an applicant is black, brown or white.
There is a narrative that says the algorithms replacing face-to-face meetings do discriminate — but not in the form of mortgage acceptance rates. According to a 2019 Berkeley study, algorithmic lenders charge minorities higher mortgage rates — 5.3 basis points more in interest for purchase mortgages and 2.0 basis points for refinance mortgages originated on FinTech platforms — but its findings “show no discrimination in rejection rates.”
Moreover, the discrimination in lending rates is inadvertent. “Lenders may be able to extract monopoly rents from minority borrowers because such borrowers might be prone to less shopping on average.”
As Samuel Clemens famously said, there are lies, damn lies and statistics. You can decide for yourself which label best applies to this particular.
P.S. Note to the Virginia Bankers Association. Find a spokesman who can do a better job defending your industry.
