Blacks and Hispanics were more likely than whites to take out higher-priced home loans in 2005, while Asians were slightly less likely to borrow money on such terms, according to a Federal Reserve report.
According to data lenders provide to the government, 54.7 percent of African-American borrowers took out “higher-priced loans,” compared with 17.2 percent of whites in 2005. Hispanics took out higher-priced loans 46.1 percent of the time, compared with 16.6 percent for Asians.
Minorities were also more likely to be turned down for loans then whites, the report found. There was little difference in loan pricing among borrowers by sex, with women buying a house on their own showing a lower incidence of higher-priced loans than sole male borrowers.
At 24.6 percent, higher-priced loans also made up a greater percentage of all home loans in 2005, compared to 11.5 percent in 2004 — a 13.1 percent increase, the report found.
Because the Fed defines “higher-priced loans” according to a threshold involving the spread between Treasury rates and the annual percentage rate (APR) lenders charge, there is not a direct correspondence between “higher priced” and “subprime” lending, the report’s authors warned.
The report, by Robert B. Avery, Kenneth P. Brevoort and Glenn B. Canner of the Federal Reserve’s Division of Research and Statistics, was based on 15.6 million loans granted by nearly 8,850 lenders — an estimated 80 percent of home lending nationwide.
The data collected under the Home Mortgage Disclosure Act is “rich, but … limited,” the report’s authors said. “The data reveal a great deal about what the lending patterns are but relatively little about what causes the patterns.”
The report was not able to pinpoint why lenders are more likely to lend money to minorities at higher interest rates than whites, but suggested some possibilities. Some companies may be engaging in aggressive “push” marketing that obscures the actual cost and terms of loans to borrowers who are less informed and who don’t shop around, the report surmised.
But lenders who do business in higher-risk markets may also face steeper costs because they pay more for funds, spend more on marketing, and have a higher “flow-through” rate in which fewer applications results in loan approvals, the report said.
The report also sheds light on other interesting trends, including a 74 percent increase in “piggyback” loans in 2005 compared to the previous year, and an increase in the share of non-owner-occupied loans often held by investors.
The Federal Reserve has been collecting data on home loans for three decades. In 1989, Congress required lenders to begin providing application and loan-level information on the income, sex, race and ethnicity of applicants.
But it was not until 2004 that lenders were required to disclose pricing information on loans above designated thresholds. The loans are referred to as “higher-priced” rather than subprime because not all of the loans that cross the threshold for reporting requirements meet the definition of higher-risk, subprime loans.
Any first-lien loan with an APR that exceeds 3 percent of the rate for Treasury securities of comparable maturity is considered “higher-priced.” The threshold for junior, or subordinate, loans is 5 percent.
The problem with that system is that as the yield curve between short- and long-term interest rates flattens, a greater proportion of adjustable-rate mortgages are defined as “higher-priced loans.” The report found that about 2 percent of the 13.1 percent increase in the incidence of higher-priced loans in 2005 was due to this effect. Other factors driving the increase might include an increased willingness by lenders to take on risk, and changes in consumers’ credit-risk profiles.
But other industry studies support the idea that subprime and near-prime lending is becoming more prevalent. Industry studies noted in the Fed’s report estimate that subprime lending accounted for 20 percent of the market in 2005, up from 18.5 percent in 2004 and 5 percent in 1994. Near-prime lending has also picked up market share, rising from 7 percent in 2004 to 13 percent in 2005, one study said.
Also, a “substantial growth” in piggyback lending “is consistent with financial stretching,” the Fed’s report concluded. Piggyback, or “80-10-10,” loans, in which borrowers with few assets take out simultaneous first- and junior-lien loans to satisfy down-payment and closing-cost requirements, are most common in California, Nevada and Colorado. They are most likely to be used by minorities, borrowers with middle or upper incomes, and those who purchased homes in lower-income census tracts.
In 2005, 22 percent of first-lien, owner-occupied home loans involved a junior-lien or piggyback loan by the same lender, up from 14 percent in 2004. All told, lenders reported a total of 1.37 million piggyback loans, up 74 percent from 2004.
While higher-cost and piggyback loans were becoming more prevalent, government-backed FHA loans continued to lose market share. FHA loans accounted for less than 3 percent of all loans, down from 16 percent in 2000, the report found.
Mortgage companies, which received more than 60 percent of the 30.2 million loan applications tracked in the report, accounted for 64 percent of government-backed loan originations, down from 83 percent in 2002.
Another interesting trend revealed in the report was the increase in non-owner-occupied loans of the type often used by investors.
Some of the strength of the housing market in recent years — and a potential weakness in a time of increased inventory — could be attributable to the growing number of sales to investors or people buying second homes.
Non-owner-occupied borrowing accounted for between 4.5 percent and 6 percent of loan volume in the mid-1990s, but has been rising in recent years. In 2004, non-owner-occupied loans accounted for 15 percent of loan volume and reached 17 percent last year. Because some investors pay cash or take out commercial loans, those numbers may even underestimate their role in the market.
Although the report illuminates many trends in the lending industry, none are as controversial as findings that banks and mortgage brokers are more likely to lend money to minorities at higher rates.
The report’s usefulness in looking at why this happens is limited because the data collected by the Fed does not include factors used by lenders to underwrite and price loans, including loan-to-value ratios, debt-to-income ratios, and credit history scores. Lenders will often use rate sheets offering different pricing depending on those factors.
“Discretionary, or flexible, pricing may be a legitimate business practice. Properly developed, monitored and administered, discretionary pricing programs may help to ensure that markets allocate resources in an efficient way,” the report notes. “However, when loan officers have latitude in deviating from rate sheets or in determining which rate sheet applies to each borrower, the lender runs the risk that differential treatment on a basis prohibited by law may arise.”
In an attempt to account for such factors, the report also compares the prevalence of higher-cost loans among different racial and ethnic groups after adjusting for borrower-related factors such as income and loan amount. The 37.5 percent gap between the rate at which blacks (54.7 percent) and whites (17.2 percent) took out higher-cost loans in 2005 is reduced by nearly 8 percentage points when adjusted for borrower-related factors, and the 28.9 percent gap between Hispanics and whites is reduced by nearly 12 percentage points.
