Berson's Weekly Commentary
March 21, 2005
Mortgage delinquencies drop in 2004, but will that trend continue?
The Mortgage Bankers Association (MBA) just released its National Delinquency Survey, covering data as of the end of 2004. Among the important findings in the survey were:
The share of total loans past due fell by 18 basis points to 4.23 percent in the fourth quarter, the lowest level since the second quarter of 2000. The entire drop was in the 30- and 60-day past due segments, however, as serious delinquency rates (those loans 90 days or more past due plus those going into foreclosure) edged up by 7 basis points.
As usual, the lowest delinquency rates were in the prime conventional market, with a total rate of 2.22 percent, down by 10 basis points from the prior quarter. Serious delinquency rates were little changed for this segment.
There was a sharp drop in the total delinquency rate for subprime loans (although the level was still about 4.5 times that of the prime market), with a decline of 51 basis points to 9.88 percent -- the first time this measure has been below double-digits since it became a separate category in the survey in 1998. Even with this improvement, however, serious delinquency rates rose by 25 basis points.
The worst delinquency performance continued to be with FHA-insured loans, with virtually no change in the total delinquency rate -- keeping it at a near-record 12.21 percent. As with the other categories, serious delinquency rates rose in the fourth quarter -- up by 39 basis points.
The increase in serious delinquency rates was a surprise, as the general trend in 30- and 60-day delinquency rates has been gradually downward over the past few years -- and these tend to be leading indicators of 90-day delinquencies and foreclosures started. This may be a result of survey noise, however, and so may not represent a new upward trend.
Looking ahead, what should the trends be for mortgage delinquency rates -- especially for serious delinquency rates? There are a number of crosscurrents in place that could move these rates either upward or downward. Among the most important is what happens to unemployment rates -- as there is a clear positive correlation between movements in unemployment rates and, with a lag, movements in delinquency rates. We project that continued modestly above trend economic growth will bring national unemployment rates down, but only slowly. As a result, this would suggest that serious delinquency rates should soon resume their downward movement -- but probably at a slower rate than usual, given the slow pace of decline in the unemployment rate.
Another important determinant of serious delinquency rates is the age of the mortgage. Borrowers rarely become delinquent early in the life of the mortgage (as long as it was underwritten appropriately), but the longer the mortgage is in place, the greater the chance that a negative shock could hit the household (e.g., job loss, serious illness, divorce, etc.). If the household has enough equity in the home, then a delinquency caused by one of these factors should not lead to a foreclosure -- as the household typically sells the house in order to recover its equity. Historically, the probability of default has reached a peak after about five years. The very young age of the mortgage stock suggests that serious delinquency rates should rise in coming years -- although the rapid price growth over the past several years in much of the country may mitigate this trend.
The behavior of home prices is another important determinant of serious delinquency rates, independent of the role it plays in the preceding paragraph. Specifically, a sharp decline in regional home prices could put some households in those regions "under water" -- that is, with a mortgage balance significantly in excess of the value of the home.
Finally, interest rate changes could cause a rise in serious delinquencies. That is, if rising interest rates cause payments on adjustable-rate mortgages (ARMs) to rise sharply relative to incomes, then an increasing number of households might have problems making their monthly mortgage payments. While only about 35 percent of the flow of prime conventional mortgages were ARMs last year (and about two-thirds of those were hybrid ARMs, and so won’t adjust soon), the ARM share was much higher in the subprime and ALT-A markets. Especially in the former category, a sharp rise in mortgage rates on ARMs (in many cases equal to the usual cap rate of 2 percentage points) could result in mortgage payments significantly in excess of income gains. For example, a $75,000 subprime mortgage at 7.5 percent that adjusts by a full 2 percentage points to 9.5 percent this year would result in a more than 20 percent increase in payments. With overall disposable income climbing by only about 4 percent (and with subprime borrowers probably seeing a smaller rise than the average -- in addition to the impact of higher energy prices), this could create a problem.
What would the impact of all of these factors be on the rate of serious delinquencies this year? The continued growth in home prices (albeit slower) and further declines in unemployment rates should mean that overall serious delinquency rates will edge down over 2005 (as should the rate for the prime mortgage market). But in the subprime arena, where the majority of loans are short-term ARMs and where the age of the mortgage stock is especially young, serious delinquency rates could rise in 2005.
This will be an important week for economic releases, with new data on inflation and home sales.
On Tuesday, the producer price index (PPI) for February is projected to rise by 0.4 percent, mostly because of higher oil prices. Excluding food and energy, the core rate is projected to be unchanged -- after last month’s surprising surge.
On Wednesday, the consumer price index (CPI) is projected to increase by 0.3 percent in February -- also with a boost from oil prices. The core rate is projected to increase by 0.2 percent.
Also on Wednesday, existing home sales should fall by about 3 percent to an annual rate of 6.60 million units for February -- consistent with the sharp decline in purchase applications in the MBA survey for January (but still strong applications in December).
On Thursday, new orders for durable goods are expected to rise by about 1.0 percent in February -- with the key orders and shipments for core capital goods categories continuing their strong upward moves, indicating that the business expansion remains in place.
Also on Thursday, initial unemployment claims are projected to slip to around 315,000 for the week ending March 19th -- continuing to show strength.
Finally, on Thursday, new home sales are projected to edge up to around an annual rate of 1.12 million units in February -- rebounding from a weather-induced sharp decline in the prior month. The trend in sales, however, should be down in future months, given the lower MBA purchase applications.
David W. Berson
Fannie Mae Economics
fanniemae.com |
