Mortgage Lending Study Raises Questions
He recently did his own mortgage lending analysis focusing strictly on the Grand Rapids market.
So Martin, a partner at Warner Norcross & Judd, wasn’t surprised by the results of the study conducted by Michigan State University Professor Joseph Darden and his consulting team on mortgage lending discrimination in the Grand Rapids MSA, released May 14.
The mortgage loan denial rate information provided in Darden’s report is consistent with denial rates around the country in that national studies show blacks are denied home loans twice as often as whites, he said.
Martin’s own analysis of the Grand Rapids market, conducted in the month before the study’s anticipated release, showed denial disparities.
“The thing that needs to be recognized is that denial disparities don’t answer the question of whether or not there’s discrimination, because the information available publicly is HMDA (Home Mortgage Disclosure Act) data,” Martin pointed out the day after the study was made public.
HMDA data includes information on race, income and the outcome of the application — but nothing else. It doesn’t include information on an applicant’s credit rating.
“So the value of reports like this is they show you where the disparities exist,” Martin said, “but unfortunately the reports can’t tell you whether or not the disparity is the result of discrimination.”
Martin said he works with a number of banks and mortgage companies affiliated with banks, all of which are examined by federal regulators who conduct denial disparity studies much like Darden did.
But when federal regulators find an area of high denial disparity rate, they’ll sample files and take about 35 credit characteristics out of each file and do a multiple regression analysis to determine what caused the disparity, he explained.
Martin said he has worked in a number of situations where banks or their mortgage affiliates have gone through that process.
“In my experience, what typically happens is that the multiple regression analysis will explain away the discrimination,” he said. “If it doesn’t, the federal regulators are obligated under law to refer the bank or bank-affiliated mortgage company to the Justice Department.”
Martin underscored that the Darden study is valuable for the community because the community needs to know disparity exists.
At the same time, it’s a valuable study for lenders as they have to be constantly concerned that they’re lending fairly and have to have second review programs in which an independent entity examines declined loans to determine whether or not they were proper declinations.
The denial rates in the Darden study weren’t out of the ordinary, Martin said, but he found some aspects of the study both interesting and questionable.
The study attempts to identify and rank lenders “that are relatively effective and ineffective in serving traditionally under-served communities” by the percentage of loans they generated in minority and low-income communities, rather than looking at the loans generated in the city of Grand Rapids by those institutions, Martin said.
“That has a real anomalous effect,” he added.
Based on the 1999 HMDA data used in the study, of the top 10 “best” lenders, eight of them did not make a loan that year to a Native American in Grand Rapids; three of the 10 did not make a single loan to an Asian person in Grand Rapids; three of the 10 did not make a loan to an African American in Grand Rapids; four of the 10 did not make a loan to a Hispanic in Grand Rapids.
And two of the 10 did not make any home loans that year to persons of color in the city of Grand Rapids.
The distinction is that Darden looked at the whole metropolitan statistical area rather than the city of Grand Rapids itself, Martin said.
“The top 10 lenders on his list are anomalous because he doesn’t look at the percentage of their involvement in the Grand Rapids community. Some of them actually have a significant involvement, but most of them do not.”
Another deficiency, as Martin sees it, is the study’s assessment of homeownership. Homeownership is vitally important, he observed, but the homeownership data used in the study dates to 1990, which was the most current data available. But 12-year-old data doesn’t accurately reflect the current market situation.
“A lot has happened in the last 10 years and I don’t think it would be prudent to draw conclusions based on 1990 data until we see 2000 data,” he said.
“I think we need to see how far the community has come in 10 years. In those 10 years there has been an enormous growth in the Hispanic community and 1990 data would tell you very little about what’s going on in the Hispanic community. Unfortunately, that was the timing of the report.”
According to Kent County census data, the Hispanic population increased by 45 percent and the Asian population by 63 percent between 1990 and 1998.
Martin also was disappointed with the study’s section on sub-prime lending. First, it didn’t put sub-prime lending in perspective, he said.
As the study points out, HMDA reports indicated that between 1993 and 1999 the number of sub-prime lenders in Grand Rapids grew from two to 262 and “their lending was targeted at minority and low-income borrowers.”
Sub-prime lending did grow dramatically during the 1990s. From 1993 to 1999, sub-prime lending grew by 880 percent, according to HUD.
In 1995 Michigan passed the credit reform act, which raised the permissible interest rate on consumer loans to 25 percent, Martin observed.
“What that did was that it allowed lenders to lend to customers that did not meet their usual standards of creditworthiness and, therefore, were more likely not to be able to repay the loan. So it was a more costly loan.
“It allowed them to charge more for the loan. They couldn’t have done that before 1995, so we wouldn’t expect to see a lot of sub-prime lenders before 1995 because it wasn’t permitted.”
The other reason for the sub-prime growth spurt was that a secondary market of mortgage-backed securities backed by sub-prime loans developed in the 1990s, which created the funding to enable the sub-prime market to grow, Martin pointed out.
“What that has done is increase the availability of credit to people who wouldn’t otherwise qualify for credit,” he said.
He said a recent report on sub-prime lending in all 331 MSAs in the country showed the Grand Rapids MSA in the bottom 20 percent for sub-prime lending.
Darden’s report notes that not all sub-prime lenders are predatory lenders, but that predatory lenders are more likely to fall within the sub-prime group of lenders.
What annoys Martin is that the report briefly addresses sub-prime lending, which is legal, and then goes on to devote a couple pages to predatory lending practices, which are illegal. The effect is a blurring of distinction between the two, and the report gives the impression the two are one and the same.
There could be predatory lenders in Grand Rapids, but the report does not document a single predatory lending practice here, Martin noted.
“The report doesn’t establish that predatory lending exists in Grand Rapids. It makes an enormous leap from a list of sub-prime lenders to the fact that the list is sometimes equated with predatory lenders. There’s nothing there.”
Kevin Kabat, president of Fifth Third Bank, Western Michigan, pointed out essentially the same flaws in the report in a letter to Mayor John Logie dated May 20.
Like Martin, Kabat acknowledged the study was valuable, but he cautioned that weaknesses in the methodology of the report “muddy its finding and lead to misleading conclusions” and stated that some of the report’s weaknesses are the result of limitations on the data.