WASHINGTON - When regulators release their report on the millions of mortgage applications processed last year, the public is likely to focus almost exclusively on one figure: the disparity in denial rates for blacks and for whites.
But experts say this statistic's value in explaining trends in minority lending is significantly overstated. Focusing exclusively on that number masks much of the real value of the Home Mortgage Disclosure Act data, they say.
Instead, anyone wanting to truly understand whether the banking industry's emphasis on minority lending is making a difference should look at several other variables, including application flows and market share analyses.
"The rejection rate information has got to be looked at very carefully," said Allen Fishbein, general counsel for the Washington-based Center for Community Change. "Questions like the volume of applications are equally as important."
"Denial. rates, market share, and applications - those are three good barometers." he added.
Regulators are expected to release national data on 1992 mortgage lending today. Preliminary information suggests that while the total number of applications increased last year - in part because of heavy refinancings - the disparity in denial rates between whites and blacks narrowed only slightly.
The HMDA data - and denial rates in particular - have become the standard by which progress in minority lending is judged. In the last few years, the industry has come under increasing pressure from regulators, the public, and law-enforcement officials to step up lending to minority consumers.
Interpreting the Numbers
HMDA data consistently show that mortgage applications from blacks are more than twice as likely to be denied as those from whites. Most agree that a huge disparity in the rate at which whites and minorities are denied mortgage applications is egregious. But economic data suggest that some of that difference can be attributed to income and credit history.
A study by the Boston Fed last year found that when many financial variables were controlled for, disparities in denial rates were lessened: Blacks were only 60% as likely to be denied mortgages as whites with similar economic status.
But even that study has not settled the debate about what disparity would be appropriate if no discrimination existed.
We still don't know what the norms should be," said Jeffrey Graham, vice president at Bank of Boston. "The Boston Fed study suggested there should be some disparity, but even the Fed hasn't been very clear about saying exactly what it should be."
Increasingly, lenders are relying on data about application flows to gauge their progress. As long as absolute numbers of applications from minorities are increasing, they say, they feel confident their outreach and marketing efforts are working.
"There's probably going to be increased interest in applicant-flow issues" instead of denial rate, Mr. Graham said.
A study earlier this year by the Cleveland Fed confirmed the importance of application flows. More than any other factor, the study found, strong fair-lending programs were correlated with large numbers of minority applications.
But increasing application flows could have the perverse effect of worsening differences in denial rates, some bankers say. Widening the pool of applicants could mean more are less qualified - and more likely to be denied.
Another good figure to look at, experts say, is market share. Lenders should ask whether their market share is significantly higher in a predominately white neighborhood than in a nearby minority area.
This question gets back to the original intent of the HMDA numbers: detecting whether lenders redline, or systematically ignore minority neighborhoods.
This dimension of HMDA was overshadowed by denial rates, though, when Congress amended the law to include applicant information, including race, income, and gender.
But a study released this summer by Ralph Nader's group Essential Information reminded the industry that this is an important issue. That study used market share analysis in 16 cities to identify 49 lenders that it said "substantially excluded or underserved minority neighborhoods."
Even more carefully examining a broader range of variables - including application flows and market share - may not be enough to understand changes in minority lending through the HMDA data.
Many bankers and activists say the long-held lending practices and attitudes of the banking industry - as well as consumers - are just too complex to be boiled down to simple statistics.
"It's just not that simple," Mr. Graham said. "There's still an awful lot that we don't understand."