PolitiFact report: Indeed, Clinton misspoke about mortgage discrimination

A reader recently asked us to fact-check something Hillary Clinton said about mortgage discrimination during a MSNBC-Telemundo town hall in Las Vegas on Feb. 18.

The reader noted that, in addition to reaching a nationally televised audience, Clinton's comment had been "going viral among real estate agents and lenders" after a YouTube commentator posted a video about it titled, "Hillary Lies about Mortgage Lending !!"

At the town hall, Clinton was responding to an audience member who asked how she planned to help Hispanics who want to buy or keep their house, particularly in Nevada, a state hit hard by the Great Recession.

Among Clinton's suggestions was to "provide more help so that more homeowners, Hispanic homeowners, African-American homeowners, those who want to have access to better credit, and better support. You know, credit has tightened up in ways that are just not fair. You are three times more likely to be able to get a mortgage if you're a white applicant than if you're black or Hispanic, even if you have the same credentials and you're presenting it to the people who are looking at it."

Clinton was essentially saying that mortgage lenders were engaging in significant racial and ethnic discrimination that can't be explained by other factors, such as income or credit history. That's a serious allegation, so we decided to take a closer look.

After checking with the Clinton campaign, we learned that she misspoke at the Las Vegas event. As evidence, they pointed to several other speeches in which Clinton had used a more accurate version of the statistic.

For instance, after winning the Nevada caucuses on Feb. 20, she said, "We see African-American families denied mortgages at nearly three times the rate of white families." She made similar comments at a Democratic dinner in Minnesota on Feb. 12 and interview with the BET network on Feb. 17.

So where did these numbers come from? Studies released by the real estate website Zillow in February 2015 and November 2015.

The Zillow study

The study looked at federal data on mortgages. The 2013 data showed a denial rate for conventional loans of 10.4 percent for whites, 21.9 percent for Hispanics and 27.6 percent for African-Americans. The 2014 data showed somewhat lower denial rates across the board — 9.4 percent for whites, 18.8 percent for Hispanics, and 23.5 percent for African-Americans.

These numbers pose a few problems with Clinton's statement.

For starters, these are denial rates, not acceptance rates, which was the metric Clinton used in her town hall comment. The acceptance rates for 2013 work out to 89.6 percent for whites, 78.1 percent for Hispanics and 72.4 percent for African-Americans, and using those numbers — or the ones for 2014 — her ratios don't work. If you do use rejection rates, it was 2.7 times higher for African-Americans in 2013 and 2.1 times higher for Hispanics — reasonably close to what Clinton said, though not a perfect fit.

A bigger problem for Clinton, however, is that she said "even if you have the same credentials," even though this study did not equalize for income, credit history or other factors before calculating those percentages.

Indeed, Zillow made a point of noting in the study of 2013 data that while the "playing field remains strikingly unequal" in mortgage loans by race and ethnicity, differences in loan denial rates stem at least in part from income and geographical disparities between the three groups. "The disparity in loan approvals is likely tied to a number of factors, including income," Zillow said.

Specifically, wrote Zillow chief economist Stan Humphries, "black and Hispanic applicants for conventional home loans make roughly $20,000 less per year than white applicants, resulting in much higher denial rates. Similarly, black and Hispanic communities are clustered in areas that saw huge run-ups in home values before the recession, and even larger drops during the crash."

Experts we contacted agreed with Zillow that differences by race and ethnicity can, and do, play a role in mortgage denials. Still, they said the scope of the racial and ethnic gap is hard to nail down, and is probably a lot lower than Clinton suggests.

Andrew Hanson, a Marquette University economist, cautioned that income and credit scores are just two of the factors that could be contributing to the racial gap. Other, harder-to-quantify factors such as wealth and market experience can magnify existing biases as well.

Hanson said he has conducted a smaller-scale study in which participants contact loan originators with basic inquiries. He then analyzed the loan originators' offerings to measure any differences by race. The result? The size of the discrimination effect was about 2 percent.

Hanson said there's another caveat when gauging racial and ethnic discrimination: Much loan underwriting today is done electronically.

"The deny-or-accept decision isn't left up to a person these days," he said. "Information is basically given to a computer and the computer decides if the loan should be made. That is not to say that humans do not make decisions along the way, but a human rarely has the final say on loan approval/denial."

Our ruling

Clinton said, "You are three times more likely to be able to get a mortgage if you're a white applicant than if you're black or Hispanic, even if you have the same credentials."

The Clinton campaign admits she misspoke. While there is evidence of disparities in mortgage acceptance rates by race, Clinton flubbed a talking point she has used correctly on other occasions. Experts say the gaps are not as drastic as Clinton says once you equalize for other key factors, such as income and credit history.

We rate her statement False.