Subprime mortgages dominate the economic news. They have pushed $3-a-gallon gas off the front pages, and completely obscured the 4.6% national unemployment rate and the economic expansion that has now lasted five years and is the fourth longest since World War II.
The common view seems to be that the roof is falling in—not just in the subprime market, but in the entire housing sector and beyond into the American, and even the world economy. As the financial markets worry about who holds what tranches of subprime mortgage-backed securities, we are losing track of the underlying problem. What is actually happening to subprime borrowers, and what is likely to happen?
The subprime market is new, and it has grown very fast. It barely existed 20 years ago. It accounted for 20% of mortgage originations last year. As it grew, it changed. Ten years ago, subprime loans were basically either refinances or debt consolidation. Fewer than 5% were used to buy homes. In addition, loan-to-value ratios were substantially lower than in the conventional prime market, typically no more than 70%. Subprime lenders knew they were taking a great deal of risk and they wanted protection against loss, as much as possible. Five years ago, home purchase loans were about one-third of subprime originations, and LTVs were higher, but still nearly all were 90% or less. In the last few years, as the home purchase share has risen to 44%, subprime lenders have relaxed underwriting standards and offered mortgages with very high LTVs, mortgages with little or no documented evidence of the borrower's ability to pay, and adjustable rate mortgages with built-in large increases in the monthly payment after two or three years.
Those risks are now coming home to roost.
The risks are limited, however. They are concentrated in the mortgages originated during 2004-2006. Loans originated in 2003 and earlier pose no unusual problem. Those loans were less risky, and those borrowers have enjoyed the double-digit home price appreciation of 2004 and 2005. Many have been able to refinance into conventional mortgages with better terms.
Recent history suggests that these risks are manageable. Subprime delinquencies and foreclosures have risen sharply this year, but they are still below the rates reached during 2000-2002. The Mortgage Bankers Association reports that about 6% of subprime loans were seriously delinquent (90 days or more) in 2005 and the first half of 2006. So far this year, about 8% are delinquent. The foreclosure rate has climbed from 1.5% at the end of 2005 to 2.5%. During 2000-2002, serious delinquencies were about 12%, and foreclosures ranged from 2.5% to 3%. Rates could go higher over the next couple of years. The foreclosures in 2000-2002 occurred on loans that were less risky than the loans of the last few years. The point is that we have been here before, not that long ago.
Most importantly, delinquencies and especially foreclosures on prime mortgages remained low during 2000-2002—consistently around 1% for serious delinquencies and 0.2% for foreclosures. The subprime spike had no impact on the prime market then. There is no reason to expect one this time. The delinquency rate has again been stable, so far, while foreclosures have edged up to 0.25% from 0.2%.
As the subprime problem has worsened, the financial regulators have been encouraging forbearance. Their joint statement in April asked lenders "to work constructively" with homeowners who cannot make their mortgage payments, and to "consider prudent workout arrangements that increase the potential for financially stressed residential borrowers to keep their homes."
The regulators are right. There is evidence that forbearance works. The Federal Housing Administration, which insures mortgages to many borrowers with less-than-perfect credit, began a forbearance program in 1999. Lenders try to work out a modified repayment plan, if there seems to be a realistic chance that the borrower can catch up. Often there is a realistic chance. FHA's experience has been that about half of borrowers in the program do become current within a year. If the lenders and the regulators follow through, they can ease the problem substantially.
In addition, the underlying economic trends are favorable to homeowners. This may seem a surprising statement when interest rates have been rising and housing starts have fallen by one-third in a year—the sharpest drop since the Census Bureau began publishing monthly data in 1959. But the relevant indicator is house prices. If they are rising, subprime borrowers have a chance to build some equity in their home before they confront higher mortgage payments.
Everyone knows that the "housing bubble" has burst, like the Tulip Bubble of the 17th century, the South Sea Bubble of the 18th century and of course the dot-com bubble. In fact, prices are still rising, on average. The best measure of house prices is the index produced by the Office of Federal Housing Enterprise Oversight, based on the millions of loans purchased by Fannie Mae and Freddie Mac over more than 30 years. The index shows that prices continue to increase, as they have in every quarter since 1993. Over the past year, prices have risen by about 4%—much more slowly than the double digits of 2004-2005, but still an increase.
Four percent is a national average, and all housing markets are local. About 10% of U.S. metropolitan areas (30 out of 282) have seen a price decline of 1% or more—a slow leak, at most.
This is an odd sort of bubble, with prices rising in most places and falling slightly in some. It is certainly not the pattern of the Nasdaq Composite Index between 1999 and 2001.
The rise in subprime foreclosures is real and likely to be large. Many homeowners will find themselves in over their head, with loans whose terms they do not really understand. These are real families who are losing their homes. If they have been making their payments for a couple of years, they are also losing the equity they have built up, plus any price appreciation.
From a policy standpoint, the subprime problem is certainly serious, but it is a short-term problem, mainly involving mortgages originated in the last three years and playing out over the next two years or so. It will be mitigated by public and private efforts to help the families in distress. There is no reason for it to spread to the prime market. Homeowners who are able to make their payments are not going to lose their homes because other homeowners are losing theirs. The roof is not caving in on housing.