Pipeline: Lenders Biased? How Do You Know?

Expert studies have proven that the mortgage lending process is riddled with racial bias.

True or false?

Ask David Andrew Price, a Washington lawyer and writer, and he will answer with a resounding "false." What is more, he believes that journalists have been taken in by poor-quality or misrepresented research, and in some cases have done shoddy research of their own.

Mr. Price works for the Washington Legal Foundation, a conservative public-interest law firm. Writing in the current issue of Forbes Media Critic, a quarterly journal, he says, "The journalism on lending bias makes for a virtual case study on how not to handle statistics."

He is no easier on researchers, accusing them of painting their results to maximize press coverage.

The Boston Fed study, for example, has been held up by many as the definitive study of racial bias in lending. He points out that among the study's findings were that there was no evidence of redlining or sex discrimination. But the press release did not mention these findings, and the study itself relegated them to a footnote, he says.

He continues: "In interpreting statistics, an unusually large gap seems to exist between what reporters know and what they think they know. Much reporting on statistics suggests that the writers put too much stock in their ability to eyeball a study and decide whether it makes sense."

He says the problem is magnified when newspapers undertake their own statistical studies. He quotes author Stephen Hess, who wrote that "good journalists may become bad social scientists."

While journalists may seek views of third parties people when they write about studies, he added, they will seldom expose their own research to independent evaluation.

Mr. Price takes pains to point out that the problems journalists may have with statistics don't suggest that lending bias doesn't exist. He points to extensive legwork in some cases that produced useful anecdotal evidence of lending bias. But he adds that the statistical studies were billboarded by the newspapers rather than the results of their own legwork.

***

The impending sale of Prudential Home Mortgage Corp. could have an impact on Lomas Financial Corp., which has itself been looking for a buyer for more than a year.

The Pru unit bought an information systems unit from Lomas late last year, paying $2.5 million in cash, a $8 million note due in five years, and contingency payments based on Pru Home's revenues.

In its latest quarterly filing with the Securities and Exchange Commission, Lomas notes that it would have the right to accelerate payment on the note in the event of a sale of the information systems unit by Pru, either separately or as part of its mortgage company.

The company also estimates it should receive contingency payments from Pru that it presently values at $40 million after applying a discount of 20% a year.

"The company's contingent earn-out interest in the future revenues of the information systems business could be adversely affected by the future performance of the purchaser," the Lomas 10Q said.

For reprint and licensing requests for this article, click here.
MORE FROM AMERICAN BANKER