The Journal on the Demographics of Subprime

The Journal has redeemed itself a bit with a page one story, “The United States of Subprime.,” in which it seeks to understand what type of people would up as subprime borrowers.

A caveat: I’m always leery of judging the quality of analytical work if I haven’t looked at the underlying methodology. One issue confounding all discussions of subprime lending is defining exactly what constitutes “subprime.” For example, the Mortgage Bankers Association classified lenders as either subprime or prime. And funny enough, they classified lenders like Wells Fargo and Countrywide, both of which have large subprime businesses, as prime lenders. According to IndyMac EO Michael Perry, as of May this year, the MBA’s classification scheme grossly understated delinquencies. It yielded a delinquency rate of 13%, when the real rate was in the 18-22% range.

Now the Journal appears to avoid this trap by focusing instead on “high yield loans.” That’s a much better standard, but they should have indicated where they drew the line. I also have a quibble with their interactive map, which is a great little feature (you scroll over various states and see the number and percentage of total mortgages in number and by value).

But coloring the map based on number of high interest loans, rather than by percentage of total mortgage value (in my mind, the most meaningful metric), makes the graphic look too much like an Electoral College map. Alaska, for instance, is colored in the lightest color because it has so few people, yet its proportion of high interest mortgages isn’t that different from Pennsylvania, which is two shades darker.

Nevertheless. the Journal did endeavor to get to the bottom of an important question: what sort of people wound up with subprimes? The authors found that subprimes aren’t a primarily urban or lower income phenomenon, but are more widely distributed than commonly believed across income levels, ethnic groups, and regions. That adds a new wrinkle to the problem.

However, one paragraph I object to characterizes subprime lending as the linear progression of a trend towards making credit available to lower income borrowers that began with the Community Reinvestment Act and continued with FHA loans. Huh? Those were government, not private sector, initiatives. FHA lending featured almost everything that subprimes lacked: sensible loan to value ratios, income verification and other assessment of the borrower, and realistic rates. In large measure, it was the favorable FHA record that led private sector lenders to take interest in lower to middle income borrowers.

It’s much more accurate to depict subprime lending as yet another instance of recent lax lending practices, which included cov-lite LBO loans, CDOs, historically low risk spreads for junk bonds, and overheated commercial real estate lending, that to depict it as the result of programs to promote affordable housing.

From the Journal:

The analysis of loan data by The Wall Street Journal indicates that from 2004 to 2006, when home prices peaked in many parts of the country, more than 2,500 banks, thrifts, credit unions and mortgage companies made a combined $1.5 trillion in high-interest-rate loans. Most subprime loans, which are extended to borrowers with sketchy credit or stretched finances, fall into this basket.

High-rate mortgages accounted for 29% of the total number of home loans originated last year, up from 16% in 2004. About 10.3 million high-rate loans were made in the past three years, out of a total of 43.6 million mortgages. High-rate lending jumped by an even larger percentage in 68 metropolitan areas, from Lewiston, Maine, to Ocala, Fla., to Tacoma, Wash.

I can’t contain myself. Since when is Lewiston, a metropolitan area? It has all of 38,000 inhabitants.

To examine the surge in subprime lending, the Journal analyzed more than 250 million records on mortgage applications and originations filed by lenders under the federal Home Mortgage Disclosure Act….the data contradict the conventional wisdom that subprime borrowers are overwhelmingly low-income residents of inner cities. Although the concentration of high-rate loans is higher in poorer communities, the numbers show that high-rate lending also rose sharply in middle-class and wealthier communities….

Yet last year’s data show that even as the housing market was weakening, some lenders still were eager to make riskier loans. Banks and thrifts grabbed 52% of the market for high-rate loans last year, up from 44% in 2005….

Higher-income home buyers began using such loans for larger purchases. Among borrowers characterized in the data as white with annual income of at least $300,000, the number of high-rate loans jumped 74% last year, the numbers show. The average high-rate loan grew 10% to $158,000 last year, compared with a 1% rise in the average size of all home loans. The 2006 data include records from 8,886 lenders nationwide, which generate an estimated 80% of U.S. home mortgages.

The high-rate loan data likely understate the potential peril posed by mortgages with low teaser rates. Under federal rules governing disclosure of high-rate loans, some subprime teaser loans do not have to be reported. Lenders weren’t required to report loan-pricing details until 2004.

The relaxation of credit standards by home lenders has been years in the making. The Community Reinvestment Act, a 1977 federal law, prodded banks to extend more credit in communities where they operated. That warmed many of them to lower-income and minority borrowers. The Federal Housing Administration, a New Deal-era mortgage insurer targeting buyers with little or poor credit, began losing market share to aggressive subprime lenders. These commercial lenders usually charged higher interest rates but promised less paperwork, faster approval and no-money-down loans that seemed more affordable to many borrowers.

Ambitious lenders such as Seattle-based Washington Mutual’s Long Beach Mortgage, which between 2004 and 2006 made $48 billion in high-rate loans, used armies of outside brokers to push subprime loans into the suburbs. (A company slogan: “The Power of Yes.”) The result was a mortgage bonanza that reached every racial and ethnic group, income level and geographic area.

By 2005, a list of subprime lending specialists compiled by the Department of Housing and Urban Development had grown to 210 lenders, from 141 in 1996. Their combined loan volume grew tenfold during the same period.

“Old industrial cities like Philadelphia have a poverty problem, and that’s why people had to use subprime loans,” says Kevin Gillen, a research fellow at the Wharton School of University of Pennsylvania. But in pricey areas such as Miami, where the high-rate market share jumped 25 percentage points from 2004 to 2006, subprime loans didn’t have a downscale reputation. They were seen as the answer to sky-high housing costs. “They are different groups, but subprime served both of them,” Mr. Gillen says.

It used to be that high-rate borrowers weren’t allowed to stretch as much as conventional borrowers on loan amounts, a reflection of their higher credit risk. But as home prices rose throughout the U.S. in the early 2000s, lenders grew more willing to let high-rate borrowers get bigger loans as measured against their annual incomes. In 2005, borrowers who got high-rate mortgages to buy one-to-four-family homes were loaned 2.1 times their reported annual income, on average, according to the data. That was 4% higher than regular borrowers.

Lenders also extended more “second-lien” mortgages — many of them “piggyback” second loans that borrowers used to cover down payments. Such second-lien loans climbed to 22% of all mortgages last year, up from 12% in 2004. Piggybacks are considered far more likely to default than a standard mortgage.

Lenders did little to discourage speculation by real-estate investors, which contributed to rising home prices. Last year, 13% of all high-rate home loans were for properties not occupied by owners, up from about 9% in 2004, the data show. Experts say such properties are higher foreclosure risks than homes lived in by their owners….

About 63% of high-rate mortgages originated in 2004 were sold that same year, compared to 68% of all home loans, the data indicate. Last year, about 73% of new high-rate loans were sold, compared to 67% rate of all home loans. Last year, the average high-rate loan carried an interest rate that was 5.6 percentage points higher than a Treasury security of comparable maturity — up from 5.3 points in 2005 and 4.8 points in 2004.