Studies Confirm Sponsors' Claims - 7/13/2005 - Mortgage Loan Refinance Debt Equity

Studies Confirm Sponsors' Claims

by Lew Sichelman

Sing along with me now ...

Of course, when reading these things, you have to pay close attention to who sponsors them. This is not to cast aspersions on anyone. Heaven forbid. But sometimes you get what you pay for. Or at the very least, you don't publish what you pay for if the conclusions aren't to your liking.

With that in mind, here are the results of three pieces of research that crossed this humble typist's desk recently:

 

• The rise of "piggyback loans" in which a borrower takes out an 80 percent first mortgage and a second mortgage to cover the difference between the first lien and the borrower's down payment, all to avoid paying private mortgage insurance, is placing undue risk on the mortgage banking system, according to a study entitled, "The Hidden Risks of Piggyback Lending."

Prepared for the PMI Mortgage Insurance Co., a major private mortgage insurer, the study found that such loans have become quite common in recent years as housing has appreciated. Some 42 percent of home purchase mortgage loan dollars involved piggyback loans during the first half of 2004, up from 20 percent in 2001.

Piggyback lending is especially popular in high-cost areas, many of which are at risk of house-price declines. And overlaying concentrations of such loans on top of PMI's assessment of the likelihood of depreciation in the top 50 metropolitan statistical areas reveals a strong positive correlation between the two.

"In fact, among the MSAs ranked in the top 10 in terms of market risk, seven regions -- all of them in California -- had more than half of their mortgage lending for home purchases in piggybacks during the first half of 2004," says the firm's chief risk officer, Mark Milner.

According to the study's author, Charles Calhoun, a widely respected independent consultant and researcher specializing in economics of housing and mortgage finance, piggybacks benefit both the borrower and the lender. The borrower can buy more expensive houses with less cash, and lenders get to generate two loans -- and two sets of fees.

But neither may be prepared for the one-two punch of rising interest rates and declining house price appreciation, the analyst warns.

"I expect that as interest rates rise and house price appreciation slows or declines, defaults will rise and borrowers could lose their homes. It's particularly worrisome given that borrowers may not fully understand the risks they face," Calhoun says.

He also says that the rapid growth of loans without mortgage insurance has increased the overall level of uninsured and lender-insured credit risk in mortgage markets, and that raises a range of capital pricing, and reserving questions.

 

• A second study, this one by two University of Virginia professors, has found that risk factors do, indeed, drive mortgage loan pricing.

In others, consumers who do not measure up to the highest credit standards are not being overcharged. Rather, the interest rate a borrower pays is directly related to his or her credit risk profile.

The new study was commissioned by the law firm of Sirote and Permutt P.C. of Birmingham at the request of the National Home Equity Mortgage Association, the only trade association solely representing the subprime wing of mortgage business. Authored by Professors Richard F. DeMong and James E. Burroughs of UVA's McIntire School of Commerce, they are based on a detailed analysis of approximately 1 million mortgage loans made in all 50 states and the District of Columbia during 2004 by multiple lenders that specialize in sub, or nonprime, mortgage lending.

The question of how certain risk factors and the presence of prepayment fee clauses affect mortgage pricing came up when DeMong testified on Capital Hill last year. He didn't have any hard data, so he decided to look into whether or not the mortgage market is efficiently pricing loans based on risk.

And guess what! After looking at a range of borrower characteristics on a huge number of loans, DeMong and Burroughs found that the level of risk these characteristics represented, along with the presence of a prepayment fee clause, directly related to the price the borrower paid for the loan, as measured by annual percentage rate, or APR.

In other words, higher credit ratings and a borrower's willingness to accept a prepayment clause translated into lower interest rates.

Specifically, their analysis found that:

 

1. The higher the credit (or FICO) score, the lower the interest rate on the loan. FICO scores, which measure risk based on a consumer's financial history, have the largest influence on mortgage prices. On average, each 10-point increase in a borrower's FICO score reduces the annual percentage rate by 10 basis points, or 0.10 percent.

    

2. Borrowers with more secure income pay less. Those who can fully document their incomes pay lower APRs on their mortgages than borrowers who only "state" their earnings but don't document them.

    

3. Buyers who live in their homes pay less than those purchasing property for rental or other investment. The average APR for owner-occupied homes is 62 basis points, or 0.62 percent, less than for investor loans.

    

4. Borrowers with lower loan-to-value ratio loans generally pay less. On average, for every one-percentage-point increase in LTV, the APR increases by 0.6 basis points, or 0.006 percent.

    

5. Borrowers pay less for loans in which they agree to pay a small fee if they pay off or refinance their loan before a certain amount of time has passed; usually two or three years. On average, a loan with a prepayment fee has an APR 38 basis points, or 0.38 percent, lower than a loan with no prepayment fee. Furthermore, borrowers with higher FICO scores are actually more likely to opt for a prepayment fee than borrowers with lower FICO scores.

 

• Finally, a third study found that private down payment-assistance programs tend to lead to lower quality federally insured mortgages.

Prepared by Concentrance Consulting for the Department of Housing and Urban Development, the research found that Federal Housing Administration-backed loans in which the seller indirectly gives the buyer money for a down payment often have inflated property values and higher defaults, largely because the purchaser has little or no money in the deal.

The Washington, D.C.-based consulting firm recommended that HUD take several steps to mitigate the risk of DPA programs, including implementing its own zero-down payment initiative. Wouldn't you know it; HUD has been having trouble getting lawmakers to fund a pilot nothing-down program.