WASHINGTON — The government's share of the refinance mortgage market is even larger than previously estimated, according to data released Thursday by the Federal Reserve Board.
Fifty-five percent of refinancings of owner-occupied homes originated in 2010 were sold to Fannie Mae and Freddie Mac, up significantly from an original 39.6% estimated by the central bank earlier this year.
"Our adjusted data shows a greater role for the GSEs than that implied by the raw Home Mortgage Disclosure Act data," the report said
The Fed said data collected under the act initially undercounted loans sold to the government-sponsored enterprises because mortgages guaranteed by Fannie and Freddie are identified that way only if they are sold directly to the GSEs or placed in a pool during the same year as origination. Thus, loans originated in the fourth quarter and later sold didn't count.
Lenders may also have opted to sell the loans to other financial institutions that form mortgage pools which receive GSE credit guarantees, rather than sell directly to Fannie and Freddie.
The new data on the GSEs was part of a slew of information released the Federal Financial Examination Council on transactions at more than 7,900 U.S. financial institutions, including commercial banks, savings associations, and mortgage companies.
The report showed that, as expected, the GSEs continued to increase their role in the mortgage market in 2010, especially when it came to refinancing. "Their share rose at the beginning of 2008, fell through August, and then rose again into 2009 and 2010," said the report. "Portfolio and other lending dropped precipitously from 2007 to 2009, before rising somewhat in 2010, particular in the refinance market."
The report also provided data on lending discrepancies to minorities. Like previous reports, regulators found that black and Hispanic-white borrowers were more likely to end up obtaining higher-cost loans while Asian borrowers were less likely compared to non-Hispanic white borrowers.
However, the data said other factors help explain the disparity among racial and ethnic groups.
"These differences were significantly reduced, but not completely eliminated, after controlling for lender and borrower characteristics," the report stated.
For example, the difference in the number of higher-priced loans obtained to buy a home between Hispanic whites and non-Hispanic whites of 3.7% fell to about 0.5 percentage points when other factors within HMDA data were accounted for.
The same was true for conventional and nonconventional lending when compared between black versus non-Hispanic white borrowers.
For conventional home purchase lending in 2010, the incidence of higher-priced lending was 6% for blacks, 7.1% for Hispanic whites, 1.0% for Asians compared with 3.3% for non-Hispanic white borrowers. The disparity was reduced to about 0.6 percentage points for both home-purchase and refinance loans when other factors were accounted for.
Loan denial rates remained about the same as in 2008. Like past years, denial rates varied across applicant groups by race or ethnicity, but the data did not include enough information to determine the extent to which these differences reflect illegal discrimination.
Blacks and Hispanic whites had notably higher gross denial rates than non-Hispanic whites, while the differences between Asians and non-Hispanic whites generally were fairly small by comparison, according to the report.
Denial rates for conventional home purchase loans in 2010 were 30.9% for blacks, 22.9% for Hispanic whites, 14.4% for Asians, and 12.3% for non-Hispanic whites.
HMDA data showed there was a similar pattern when it came to nonconventional home purchase lending; though the gap between blacks or Hispanic whites and non-Hispanic white was narrowder than for conventional home purchase loans. Additionally, the gap between Asian and non-Hispanic white was higher in this category.
"Conventional lending denial rate disparities between groups, both gross and controlling for other factors, have narrowed somewhat in the past several years," the report said. "This narrowing appears to stem more from changes in the composition of the applicant pool over time than changes in the way lenders act on specific applications."
The Fed said accounting for specific lenders used by the applicant reduces differences further. Even so, there are unexplained differences between non-Hispanic whites and other racial and ethnic groups.
"Both previous research and experience gained in the fair lending enforcement process show that unexplained differences in the incidence of higher-priced lending and in denial rates among racial or ethnic groups stem, at least in part, from credit-related factors not available in the HMDA data, such as measures of credit history (including credit scores) and LTV and differences in choice of loan products," the report stated. "Differential costs of loan origination and the competitive environment also may bear on the differences in pricing, as may differences across populations in credit-shopping activities."
Separately, research showed that mortgage originations fell to fewer than 8 million loans in 2010 from 2009, which was just shy of 9 million. That was led by a significant decline in refinance loans, despite historically low mortgage interest rates, and drops in home purchases.
The number of homes purchased dropped sharply in the middle of last year, right around the June closing deadline of the first-time homebuyers credit. The end of the program, which was extended to September, could help to explain a lower number of loans made to lower-income borrowers between the first and second halves of the year.
