WASHINGTON - Banks may have lost a defense against redlining charges.

Two economists with the Federal National Mortgage Association said earlier studies erred in supporting a reason often given for lender avoidance of low-income neighborhoods.

The idea that home values in such neighborhoods tend to fall is mistaken, the researchers found. In fact, they said, home values in poorer areas have actually gone up, while those in affluent areas have slipped.

Their study indicates that bankers, in avoiding low-income neighborhoods, lack an economic justification they could cite before to counter fair-lending charges. Now they must focus on the credit histories and incomes of particular applicants - rather than the area in which they live. That is much harder to do.

The economists, Eric Rosenblatt and Ying Li, in a paper presented Tuesday to the Society of Government Economists, said the flaw in earlier studies was a reliance on demographic information that wound up having no correlation to housing values.

The earlier studies looked at the number of welfare recipients and the number of vacant homes in an area to explain why bankers did not lend in certain communities.

Many economists have questioned the validity of those studies, arguing that bankers consider only the borrower and the house when making loans. Community characteristics such as the number of welfare recipients don't matter, they said.

The Fannie Mae study is the first research project to formally rebut the prior findings.

The researchers based their conclusion on data collected by Fannie Mae and the Federal Home Mortgage Corp. that covered housing purchases in California - including Los Angeles, Anaheim, and Oakland. They limited their review to homes that changed hands at least twice between 1986 and 1994.

Mr. Rosenblatt said the earlier studies assumed that communities where most residents attended college or had higher incomes would appreciate in value, and that areas with many welfare recipients would fall in value.

However, the researchers, comparing the California data to Census Bureau information, concluded that the demographic variables had the opposite effect.

"Higher-income areas did worse," Mr. Rosenblatt said. "The more welfare in those areas, the better those houses did."

With home values destined to appreciate in lower-income communities, a bank's collateral - in this case the house - thus became more valuable in areas with higher welfare rates and more boarded-up buildings.

And in high-income neighborhoods, where the opposite occurred, housing values fell by up to 15%, causing some borrowers to owe more on the house than it was worth. That left bankers in a dangerous position because they would not collect the total value of the mortgage if they had to foreclose.

The new study may not be the last word on the use of neighborhood characteristics. One bank economist who saw the Fannie Mae study said it contained an error that could reignite the debate.

The source, who asked to remain anonymous, said the Fannie Mae economists found that the ratio of rent to a property's value varied greatly. The economists said this shows crime and other factors do affect a house's value. The ratio should be stable if demographic factors are irrelevant, the economist said.

The economist also questioned the need for further research.

"These particular studies that the Fannie Mae economists are shooting at should not have been considered terribly important in the first place, because they suggest that banks make lending decisions based more on geography than they really do," the economist said.

"Bank lending decisions are driven by characteristics of individual applications," the economist asserted.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.