A new study finds that whites and blacks pay almost identical interest rates for home mortgages, although minorities pay substantially more when they refinance.

The study, conducted by the Federal Reserve Bank of Chicago and published in the July issue of the Chicago Fed Letter, appears to support industry claims that banks price first mortgages fairly, but it also provides fodder for critics who claim banks discriminate on second mortgages.

"The fact that there is a reduction in the overall statistical disparities is good news," said Deval L. Patrick, a partner in the Boston office of the Day, Berry & Howard law firm and a former assistant attorney general for civil rights. "But is it good enough news to declare victory and move on? No, that would be premature."

Paul F. Hancock, the acting deputy assistant attorney general for civil rights, said the refinancing data indicate that additional prosecutions may be warranted.

"The differences in rates charged to minorities compared to nonminorities is disturbing," Mr. Hancock said. "There may be a business justification for it, or it may be because of unlawful discrimination based on race."

The Justice Department has sued seven banks since 1993 for charging minorities more than whites.

The study found that minorities paid on average 8.1% for first mortgages in 1996, about 20 basis points more than whites. "We saw that difference as so small we didn't need to worry about it," said Paul Huck, a Chicago Fed economist who co-wrote the article with colleague Lewis M. Segal.

Mortgage rates were 8.4% for low-income households, about 50 basis points higher than for middle- and upper-income families. The economists attributed the difference to risk-based pricing.

The pricing differences were substantially greater for mortgage refinancings. Minorities paid 9%, 200 basis points more than whites.

Also, low-income households paid 8%, 100 basis points more than wealthier families.

The economists theorized that minorities and low-income consumers are more likely than wealthier whites to remove equity from their homes to pay off other debts when they refinance. This would result in higher loan-to- value ratios, which could cause lenders to raise rates. Future studies will explore this in more detail, Mr. Huck said.

The article was based on data from the 1996 Panel Study of Income Dynamics, which looked at 8,500 households. This is the first time the 28- year-old study has looked at lending data. Other findings include:

Forty percent of whites had previous relationships with their lenders, compared with 18% for minorities. Nearly a quarter of whites had a prior loan with the bank, compared with 4% for minorities.

Twice as many minorities as whites receive government-guaranteed mortgages, and 16% of minorities have variable rate loans, compared with 19% of whites.

Twenty-eight percent of minorities were referred to a lender by a real estate agent, compared with 20% for whites.

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.