Printer Friendly

The trends and outlook for foreclosure & delinquencies: problem loan rates are now nearly double what they were in the late 1960s.

THE TRENDS AND OUTLOOK FOR FORECLOSURE & DELINQUENCIES

Problem loan rates are now nearly double what they were in the late 1960s.

Servicing residential loans is the bread and butter of the mortgage banking business. When loans become delinquent, servicing income is undermined. When they go into foreclosure, servicing value is lost. Consequently, the outlook for delinquency and foreclosure rates is of critical importance to mortgage bankers.

During the 1980s and now in early 1990, there has been cause for concern. Problem loan rates are considerably higher now than they were in the 1960s or 1970s. Although the performance of many categories of problem loans appears to have shown some improvement in the past few years, only modest improvement has occurred in the most severe categories, that is, loans 90 days or more delinquent or in foreclosure.

This article analyzes trends in delinquency and foreclosure rates, examines the causes of defaults and discusses the outlook for problem loans.(1)

Some of the key findings and most important conclusions include:

* Problem loan rates (delinquencies

and foreclosures) rose steadily

from the mid-1960s to 1985, before

declining in the past few years.

Still, problem loan rates currently

are nearly double what they were in

the late 1960s. * From 1965 to 1989, the number of

delinquent loans rose 265 percent

and the number of loans in

foreclosure increased 590 percent. * FHA problem loan rates have been

consistently higher than

conventional or VA problem loan rates. * During much of the mid-1960s to

1989, problem loan rates were

highest in the Northeast, but

problems soared in the South during

the 1980s. * The collapse of energy prices

resulted in a sharp increase in

problem loan rates in energy states, but

was not sufficient to account for the

general rise in problem loan rates

in the U.S. * Higher problem loan rates on ARMs

during some years in the 1980s and

their growing share of total loans

outstanding drove up the national

problem loan rate in some years,

but was not sufficient to account for

the general rise in U.S. problem

loan rates. * Changes in 60-day and 90-day

delinquency rates provide useful guides

to future changes in the percentage

of loans in foreclosure. * The amount of home equity is a

critical determinant of default

rates--this was underscored by

statistical tests done for this article.

Because home price appreciation is

one of the most important

determinants of equity, it plays a key role in

explaining trends in default rates. * Changes in the unemployment rate,

consumer debt burdens, loan-to-value

ratios, and, perhaps, divorce

rates also help explain trends in

problem loan rates. * Personal bankruptcies are

significantly contributing to the costs of

loan problems. * Problem loan rates are not likely to

continue to fall in the early 1990s,

but they are not likely to reverse

course and rise either. Rather, they

are expected to remain roughly at

current levels.

National trends

During the late 1960s, the problem loan rate--delinquent loans and loans in foreclosure as a percentage of total loans serviced--stood at about 2 1/2 percent (See Table 1 and Chart 1)(2). Sharp increases during the first half of the 1970s moved the rate up to around 3 3/4 percent, where it remained during the latter half of the 1970s. Further sharp increases in the first half of the 1980s pushed the rate to about 5 1/2 percent in 1985 and 1986, before it fell to a little more than 4 1/2 percent by 1989. Even with the recent declines, however, the problem loan rate still is almost double what it was in the late 1960s.

Delinquencies comprised the bulk of problem loans throughout this period. The delinquency rate drifted up through the 1970s and early 1980s, peaked in 1985, then declined. To a large extent, this pattern reflected movements in 30-day delinquencies, which rose during most of the period, peaked in 1985, then declined sharply. Unfortunately, the most severe delinquency category--loans 90 days or more delinquent--did not fare as well. In the late 1960s, the rate was about 0.2 percent; by 1985, it had reached 0.8 percent--a fourfold increase. It was still relatively high in 1989 at 0.6 percent.

Foreclosure problems did not improve much either. In the late 1960s, only about 0.2 percent of loans serviced were in foreclosure. By 1987, the rate more than quadrupled to 0.9 percent, and was still 0.7 percent in 1989.

The upshot is that most of the improvement in the problem loan rate in recent years came in the least severe category of delinquencies--loans only 30 days delinquent. Significant problems remain with longer term delinquencies and loans in foreclosure.

The upward trend in delinquency and foreclosure rates coupled with the steady rise in the number of loans serviced caused a dramatic increase in the total number of problem loans. In 1965, about 465,000 loans were delinquent and another 45,000 were in foreclosure. By 1989, about 1.7 million loans were delinquent and another 300,000 were in foreclosure--increases of roughly 265 percent and 590 percent, respectively.

Trends in loan problems among conventional, Veterans Administration (VA), and Federal Housing Administration (FHA) loans were similar, but not identical. Conventional loans, which constituted about 75 percent of loans serviced in 1989, had lower problem loan rates throughout the period. Further, their improvement since peaking at 4.6 percent in 1985 has been greater than the improvement in either VA or FHA loans. Still, conventional loans, like their government guaranteed or insured counterparts, had significantly higher problem loan rates in 1989 than in the late 1960s. The conventional rate was about 1 3/4 to 2 percent in the late 1960s, but was about 3 1/2 percent in 1989. Much of the increase was due to rising delinquency rates, but loans in foreclosure also shot up from 0.1 percent to 0.5 percent by 1989.

The FHA problem loan rate in 1989 was more than twice the rate of the late 1960s despite some improvement since 1985. Throughout the period, the FHA problem loan rate exceeded the rates of VA and conventional loans, sometimes by very wide margins. This outcome is not surprising, however, because the FHA program is designed to serve a portion of the low and moderate income market that the private sector often will not serve as well as borrowers that may be ineligible for VA programs.

The improvement in the FHA problem loan rate since the 1985 was primarily due to a sharp improvement in delinquency rates. In 1989, loans in foreclosure remained high, with about 1 1/4 percent of FHA loans outstanding in foreclosure.

Problem VA loan rates did not deteriorate as much as FHA rates from the late 1960s, but they did not improve as much since 1985 either. The VA problem loan rate remained at about 7 1/2 percent from 1985 to 1989. The delinquency rate peaked in 1985, then came down, but the rate of loans in foreclosure continued to rise through 1988, before declining very modestly in 1989. Like FHA loans, nearly 1 1/4 percent of VA loans outstanding were in foreclosure in 1989.

For mortgage bankers, the problem is especially acute with respect to VA loans due to the "no-bid" situation. VA no-bids arise when VA opts to pay the guaranty--a maximum of $36,000 currently--rather than take possession of the property and pay off the full amount of the loan. VA takes such an action when it would lose less by doing so than by taking possession and selling the property. When a no-bid occurs, the servicer takes possession of the property and invariably loses money on the subsequent sale.

With about 2 1/4 percent of VA loans 90 days or more delinquent or in foreclosure, many additional claims are likely to arise. Increasingly during the 1980s, VA chose to respond to claims with a no-bid. Because mortgage loan servicers lose substantial amounts of money on virtually every no-bid, improvement in the performance of VA loans could significantly improve the profitability of these servicers. However, few signs of improvement in VA loan performance have emerged at present.

Regional trends

Patterns of change in regional delinquency and foreclosure rates and the levels of those rates underscore pronounced differences in the components that shape the national averages.

During most of the 1965-1989 period, the Northeast had the highest problem loan rates for all loan types. The region's problem loan rates peaked in 1983. The first quarter of 1983 was very near the November 1982 trough of the 1981-82 recession--then trended downward before turning up in mid-1988.

The improvement in Northeast problem loan rates following the recession until mid-1988 generally tracked patterns of economic growth in the region. Robust growth in the Northeast during the 1980s surprised many observers who expected the region's economic sluggishness during the 1970s to continue into the 1980s. Instead, the near boom conditions following the 1981-82 recession drove unemployment rates to below 4.0 percent in seven of the region's nine states, before growth slowed significantly in 1988. This slowdown is probably the primary cause for the upturn in problem loan rates in the region since mid-1988.

Problem loan rates for all loan types in the Midwest and West trended upward during most of the period. VA and FHA problem loan rates peaked in 1985 in both regions, and conventional loan rates peaked in 1983 in the West and in 1986 in the Midwest. Even after substantial improvement, these regions' conventional problem loan rates in early 1989 were still 70 to 80 percent higher than during the late 1960s. The problem loan rates of VA and FHA loans ranged from 80 to 160 percent higher.

The extent of improvement in the two regions since 1985 differed by loan type. The improvement in rates for conventional loans was about the same in the two regions, but improvement in rates for government guaranteed or insured loans was slightly better in the Midwest than in the West. One possible explanation is that home prices reached levels in some markets in the West--notably in California--where homes in the middle price range locally required loans beyond the FHA loan limits or the secondary market ceiling for VA loans. If default rates on mid-range homes are relatively favorable (low), as some evidence suggests, then the more modest improvement in problem loan rates in the West may have reflected the failure of government loan limits to keep pace with home prices.

In some ways, the situation in the South is the most troubling of the four regions. Like the other regions, problem loan rates climbed during most of the period, but, atypically, problem loan rates for VA and FHA loans in the South did not turn downward after 1985 or 1986. Further, although conventional problem loan rates in the South declined slightly after 1986, the rate remained the highest conventional problem loan rate of all of the regions in 1989, after having been the lowest as recently as 1984.

The severity of problems with VA and FHA loans in the South was exemplified by the surge in foreclosures. The percentage of both VA and FHA loans in foreclosure jumped 0.9 percentage points there from 1983 to 1989, nearly double that of the next largest increase of any other region. In short, loan performance in the South deteriorated more significantly than anywhere else during the 1980s.

Part of the South's problems, of course, stemmed from movements in energy prices. These prices began a slow decline starting around the end of the 1981-82 recession, then collapsed in 1986. This weakness in prices undercut business activity and employment in extraction industries and contributed to major economic downturns in several Southern states, notably Texas, Louisiana and Oklahoma. Because of Texas' size, the South felt the energy price collapse particularly hard; but it was not alone in absorbing the aftershocks.

Charts 2 and 3 show the sum of the conventional 90-day or more delinquency rates and the percentage of conventional loans in foreclosure by state for 1980 and 1988, respectively. The charts illustrate two important points. First, a comparison of the rates in 1980 with those in 1988 dramatically shows the extremely large relative increases in problem loan rates in energy states, such as Texas, Oklahoma, Louisiana, Colorado, Alaska, West Virginia and Wyoming, compared to most other states. Second, however, is that problem loan rates rose in many states not tied to energy production in any important way. In short, weaknesses in energy-producing states contributed to the rise in national problem loan rates, but did not account for most of the deterioration.

Carving out the energy states results does not fundamentally alter the national pattern of problem loan rates in the 1980s. Chart 4 illustrates the point using VA loans as an example. Further, Table 2 shows changes in problem loan rates in the energy states, in the non-energy states and for the U.S. as a whole from 1980 to 1985 (the latter year is approximately the peak of problem loan rates) and from 1985 to 1989. As Table 2 and Chart 4 clearly illustrate, excluding the energy states produces series that still track national problem loan rates closely. Consequently, the secular rise in problem loan rates was not caused by specific regional problems, but by broad-based national influences. A review of some recent research is a helpful way to identify some of these influences.

Results from previous studies on delinquency and foreclosure

Table 3 summarizes some of the common variables considered in delinquency and foreclosure research and the expected relationships with default rates. The table does not distinguish explicitly between delinquency and foreclosure. Some researchers have emphasized that the same factors that affect delinquency rates may not affect foreclosure rates. The opposite also holds true. While both points of view have merit, underwriters and loan administrators must focus on all factors and relationships associated with loan problems during the life of the loan. Virtually all loans are delinquent for a substantial period of time before a foreclosure proceeding is initiated. During that time, the cumulative interest in arrears may offset any equity the borrower has in the property, making it a potentially rational choice for the borrower to allow the property to go into foreclosure even if the property value exceeds the loan amount by a substantial margin.

A common point of departure among theoretical models used for studying defaults is that the amount of equity is crucial. One typical theoretical model is set up to view mortgage default as a put option, giving the borrower the right to sell the house to the lender for the principal balance. Wealth-maximizing borrowers will be increasingly likely to exercise this option whenever the market value of the house, net of all relevant transactions costs (e.g., selling costs, moving expenses, etc.), is less than the market value of the mortgage.

An analysis of the basic components of net equity (derived using the market rather than the par value of mortgages) suggests that initial loan-to-value ratios, home price appreciation rates, current versus contract mortgage interest rates and the age of loans should all influence default rates. The lead-lag relationships, however, between these variables and default rates are complex. For example, frequently there's a long lag from the time the fundamental conditions leading to default occur, and the time the loan goes into foreclosure. Also, some research suggests that current, negative net equity may not trigger a default if borrowers expect significant equity gains from future appreciation. In short, borrowers are concerned with maximizing their long-term wealth position, not just their wealth at any point in time. Consequently, borrowers' expectations play a key role in the outcome of a problem loan.

Defaults are most common three to five years after origination, or perhaps a bit earlier for FHA loans. This lag relationship suggests that developments during the first few years of the loan contribute to defaults and have an important impact on the loan's future.

Aside from net equity, studies have tested the effects that the following have on default rates: age, sex, occupation, payment burden (typically measured as mortgage payment-to-income ratios, with variants including inflation adjusted and nominal measures, pre-and after-tax measures and assorted combinations), marital status, number of dependents, source of income (e.g., wage and salary, commission, child support, alimony, etc.), unemployment rates and other borrower characteristics. Many of these variables are viewed as proxies for the reliability of income streams over time. In some studies, they are found to be important influences, but in others, they're not significant.

The increasing importance of adjustable rate mortgages (ARMs) also has been viewed by some as a fundamental cause for the sharp rise in default rates from the 1970s to the 1980s. The rationale behind this view is that payment shock results in defaults as income growth falls behind payment adjustments. This would be more credible in an environment of continuously increasing interest rates. However, interest rates fell during most years from 1981 to 1985--a period when problem loan rates were rising. Further, FHA and VA problem loan rates rose sharply even though no VA loans are ARMs and only a handful of FHA loans are ARMs.

Still, ARMs appear to have greater delinquency and foreclosure problems than fixed-rate loans during some periods. These problems are magnified by the sharp rise in the ARM share of loans outstanding, from a very small share in the late 1970s, to about 10 percent in 1984, to roughly 26 percent by the end of 1988. Chart 5 shows the total problem loan rate quarterly from the first quarter of 1984 to the final quarter of 1988. It also shows the problem loan rate excluding ARMs. Both measures declined sharply between 1985 and 1988, but it is clear that ARMs kept the total problem loan rate much higher during late 1985 and early 1986 than it would have been without them. The wide gap between the two measures reflects the relatively more severe loan problems of ARMs. However, the two measures ultimately converged as ARM delinquency and foreclosure rates improved significantly in 1987 and 1988. The bottom line is that ARMs played a role in increasing loan problems during the mid-1980s, but the general patterns of problem loan rates remain even if ARMs are excluded from the mix. Therefore, ARMs alone do not account for the broader trends.

In addition to borrower and loan characteristics, the institutional framework of a state for handling foreclosure proceedings can influence the decision of a lender to foreclosure. For example, if the circumstances are the same, the probability of foreclosure is lower in states requiring judicial foreclosures than in states with much less costly power-of-sale foreclosures.

The institutional framework also can influence a borrower's decision to default. Deficiency judgments are permitted in most states and allow lenders to recoup losses from the sale of foreclosed properties with negative equity by attaching claims to borrowers' other personal assets. The threat of deficiency judgments in most states almost certainly holds down foreclosure rates. California is the largest state that does not permit deficiency judgments.

The conclusion that emerges from research on defaults is that net equity appears to be the most crucial determinant. Experience shows that if income is inadequate to meet payments, movement toward foreclosure is inevitable. However, most studies are unable to find clear statistical links between payment-income imbalances and foreclosures. Finally, the institutional characteristics of state foreclosure procedures appear to have some influence on the decisions made by borrowers or lenders.

Relationships among problem loan rates

On a quarterly basis, longer term delinquencies are expected to follow the path of shorter term delinquencies with a lag. For example, FHA 60-day delinquencies should reflect movements in FHA 30-day delinquencies in prior months; 90-day delinquencies should reflect movements in 30-day delinquencies with even a longer lag. The logic, of course, is that unremedied delinquent loans move through this sequence of categories to end up in the 90-day or more category and ultimately in foreclosure.

Statistical tests confirm these expected relationships. For all three loan types, changes in 90-day delinquency rates are significant indicators of changes in the percentage of loans in foreclosure, and changes in 60-day delinquencies reliably portend changes in 90-day late loans, but changes in 30-day delinquency rates do not appear to be as reliable an indicator. In some cases, these shorter duration delinquency rates reflect future patterns of longer duration rates, but in other cases they do not. It is likely that there is more meaningless movement in 30-day delinquency rates than in the other rates, because normally reliable borrowers miss a single payment for one reason or another at times, but quickly become current again.

There is a very high correlation among default rates across loan types. Movements in FHA delinquency rates tend to mirror movements in VA delinquency rates. Movements in VA delinquency rates, as well as the percentage of VA loans in foreclosure, tend to mirror movements for both these categories for conventional loans, and so forth.

Statistical tests

Given the findings of previous studies and the statistical relationships among default rates of various durations and across loan types, it is likely that a common set of factors can help to explain the variation in default rates over time. The factors tested here include: changes in home prices, loan-to-value ratios, current versus historical mortgage interest rates, unemployment rates, income growth, divorce rates, consumer debt burdens, changes in interest rates affecting ARM payments and growth in self-employment. The first three factors tested are determinants of net equity; the next six affect income flows and/or reflect the ability to pay.

Changes in home prices--Changes in home prices provide nearly all equity gains after origination until very late into the term of the loan. Early in the life of a loan, very little principal is repaid. Generally, the relationship of home price changes to defaults is inverse. For example, if prices appreciate and equity builds, borrowers will likely make every effort to avoid delinquency problems and are less likely to face foreclosure, because they are more likely to be able to sell their home for more than the amount due on the loan after expenses. Several measures of home price changes were tested, including the National Association of Realtors' (NAR) median price for existing homes sold (both adjusted and unadjusted for inflation) and the Census Bureau's index of constant quality new home prices (again, both adjusted and unadjusted for inflation).

Loan-to-value ratios--The loan-to-value (LTV) ratio determines the initial equity. The expected relationship with default rates is positive; if the LTV ratio is high and the initial equity is thin, the financial stake may not be large enough to compel borrowers to make their loan payments on time.

Current versus historical interest rates--The difference between current and historical interest rates provides a guide to the market versus par value of a loan. The expected relationship is inverse for delinquencies, but ambiguous for foreclosures. If the current market rate is higher than the historical rate, borrowers holding loans at the historical rate are certain to try to avoid loan problems because their payments would rise if they were forced to acquire new loans (if possible at all) at the current rates following a default.

Lenders may have a different response. Lenders have limited influence over delinquencies, but do have choices concerning foreclosures. If the current market rate is greater than the historical rate, portfolio lenders may push forward with foreclosure proceedings. Because borrowers and portfolio lenders may react to the difference between current and historical interest rates in opposite ways, the relationship of this difference to the percentage of loans in foreclosure is ambiguous. Further, it is likely that expectations about changes in the difference are critical because mortgage loans are such long-term instruments. Consequently, it may be that what matters are the changes in the rate differential rather than the disparity itself.

Unemployment rates--The unemployment rate affects income flows. Default rates generally increase with rising unemployment. The variable chosen to represent unemployment is the Bureau of Labor Statistics' (BLS) civilian unemployment rate for 25- to 54-year old males. This unemployment rate provides a better gauge of cyclical movements in the economy over the past 25 years than the overall unemployment rate, because the latter rate is strongly influenced by shifts in the age and sex composition of the labor force.

Income growth--The expected relationship of income growth with default rates is inverse; as the ability to pay increases with income, loan problems are likely to diminish. The income variable is measured by changes in the Bureau of Economic Analysis' (BEA) real disposable personal income per capita.

Divorce rates--Divorces can affect both the ability to pay and the willingness to pay. The expected relationship with default rates is positive; as couples divorce, income that previously supported one household must now often support two. Further, in hotly contested divorces, disputes over property and financial responsibilities can result in the failure to make mortgage payments by either party.

Consumer debt--Consumer debt burdens can restrict a borrower's ability to pay. The expected relationship with default rates is positive. As consumer debt (excluding mortgage debt) rises relative to income, borrowers may be unable to meet all their financial obligations, including their mortgage payments. The variable is measured as the ratio of the Federal Reserve's consumer installment debt outstanding to BEA's personal income.

ARM rate changes--Changes in interest rates affecting ARM payments impact the ability to pay. The expected relationship with default rates is positive; as ARMs are adjusted upward, loan problems may rise. The variable is measured as the change from one year ago in the one-year constant maturity Treasury rate--the most common index used to adjust ARM rates. The variable enters the analysis only in the 1980s as ARMs grew in volume.

Self-employment growth--The growth in self-employment affects the stability of income flows. Self-employment income tends to be much more volatile than the generally stable flow from wage and salary employment. Therefore, the relationship with default rates is expected to be positive; as self-employment increases as a percentage of total employment, loan problems may rise. The variable is measured as the ratio of nonagricultural self-employed workers to all nonagricultural workers (BLS household survey).

Obviously, all of the measures of the variables as discussed are proxies. Further, practically all of these measures have statistical problems of their own limiting their reliability in measuring what they purport to measure. Even so, the results of the tests are quite interesting.

Results

Separate tests were done on conventional delinquencies of 60 days or more and on the percentage of conventional loans in foreclosure. Tests were also done on comparable measures for VA and FHA loans, resulting in a total of six separate tests. In all cases, the factors tested (taken together) explained virtually all the variation in the respective default rates by loan type.

Across all tests, the unemployment rate, prices and consumer debt appear to be among the most important factors tied to higher default rates. The unemployment rate showed the expected relationship and was statistically significant in explaining default rates in four of the six tests; while price increases showed the expected relationship and were statistically significant in all six tests. (Here, the NAR nominal median price measure performed best among the measures of price change.) Finally, the consumer debt variable had the expected relationship with defaults and was highly significant in five of the six tests.

The LTV and ARM variables were tested for conventional loans only. The LTV variable demonstrated the expected relationship with defaults and was highly significant in the conventional delinquency and foreclosure tests. The ARM variable demonstrated the anticipated connection with problem loan rates and was statistically significant in the delinquency test, but not in the foreclosure test.

The remaining variables--income, divorce rates, self-employment, and current versus historical mortgage rates--showed mixed results, but did not appear to be as important as the variables already discussed. The divorce variable showed the expected relationship and was highly significant in the VA and FHA delinquency tests, but was not significant in any of the other tests. Similarly, the self-employment and current versus historical interest rate variables appeared to influence conventional delinquency rates in the anticipated ways, but did not appear to matter much in the other tests. The income variable was not significant in any of the tests.

Although obviously not conclusive, the tests do produce some interesting results consistent with prior research and common sense. Trends in unemployment, home prices, loan-to-value ratios and consumer debt appear to be very significant contributors to the secular rise in default rates.

Institutional factors

These tests do not capture some institutional developments that also may have contributed to rising problem loan rates. The first has been discussed briefly: the effects of state foreclosure laws. Another is personal bankruptcy.

It has been suggested that state foreclosure laws influence the willingness of lenders to pursue foreclosure aggressively. If the process is extremely time-consuming, it may be more profitable for lenders to make financial concessions in order to resolve the deficiency rather than pursuing foreclosure. Because shifts in the nation's population with time contribute to a gradual shift in the geographical distribution of the housing stock and related mortgages, the national foreclosure rate could rise solely due to increased housing market activity in states with laws more conducive to foreclosure. Indeed, some of the fastest growing states during the 1970s and the 1980s, such as California, Texas and Florida, are also states that process foreclosures in shorter periods than the national average. As the share of loans increases in such states, the national foreclosure rate may drift up as a consequence.

To examine this possibility, a variable was created that measures the effects of shifts in housing market activity among states on the national average number of weeks required to complete foreclosure. Tests indicate that the expected inverse relationship with default rates was confirmed for all three loan types; as the length of time increases, some lenders may attempt to avoid the foreclosure process as much as possible. Consequently, the shift in the nation's population toward states with foreclosure laws that are simpler and less costly for lenders may have contributed to the secular rise in the percentage of loans in foreclosure. However, this conclusion is somewhat speculative.

Another institutional factor that may influence default rates is personal bankruptcy. A bankruptcy filing temporarily halts collection efforts by creditors; therefore, mortgage defaults may rise as bankruptcies rise. Unfortunately, testing the impact of personal bankruptcies on mortgage default rates is difficult. Research by the Administrative Office of the U.S. Courts--the organization that compiles data on bankruptcy filings--suggests that the consumer debt-to-income ratio is a critical determinant of bankruptcies. However, because a similar variable is probably an important cause of mortgage defaults, it is difficult to prove that bankruptcies per se contributed to the rise in mortgage default rates.

The significant liberalization of bankruptcy laws during the late 1970s made it easier to declare bankruptcy. As a result of that liberalization, the stigma attached to bankruptcy greatly diminished during the 1980s. These changing attitudes and increasing willingness to use bankruptcy made it likely that mortgage problems could increase as a result.

There is no doubt that bankruptcy filings are on the rise. In 1978, there were about 175,000 personal bankruptcies filed in the United States, or about 0.08 per capita; by 1988, there were about 750,000 filed, or about 0.30 per capita--nearly a four-fold increase per capita.

Outlook for delinquency and foreclosure problems

The decline in the delinquency rate since 1985 may have reached a trough in the short-run. The total delinquency rate for all three loan types was slightly higher in 1989 than in 1988. On the other hand, the earlier declines in delinquency rates are still contributing to declines in the percentage of loans in foreclosure. For all of 1989, the percentage dropped about 5 to 10 basis points from 1988 for conventional and VA loans and an impressive 20 basis points for FHA loans.

However, with delinquency rates showing few signs of decreasing further in the short run, the prospects for further declines in the percentage of loans in foreclosure are limited. It is more likely that the percentage will stabilize around current levels.

Over the long-run, the outlook for default rates hinges on the fundamental contributing factors discussed earlier. The most important ones likely include changes in home prices, consumer debt burdens, LTV ratios, the impacts of institutional factors, trends in unemployment rates, and, perhaps, changes in divorce rates. The direction each will take, in turn, is far from certain.

Nominal home prices are likely to increase at a more modest rate over the next decade than during the late 1970s and some of the 1980s. Price appreciation in these latter periods was strongly influenced by the explosive increase in household formations, coupled with extraordinary low real after-tax mortgage interest rates. The demand pressures on prices from demographic changes will be much less pronounced in the 1990s. Further, real after-tax mortgage interest rates are likely to remain relatively high. Reduced demand from these sources--if not offset by other sources of demand or supply pressures--will likely mean slower price appreciation and slower equity accumulation. Because equity growth is critical in holding down default rates, modest price appreciation in the 1990s is not likely to be a strong force in reducing default rates. But equity growth will probably be strong enough to contribute to stability in those rates.

The outlook for consumer debt is less favorable. Consumers continue to amass debt at a rapid rate. Concerns about consumer debt accumulation have been raised many times in the past 50 years, but the economy has not faced the dire consequences often predicted. For lenders, though, there is reason for concern. Large debt burdens coupled with bankruptcy filings are hurting mortgage lenders. These factors are likely to continue to put upward pressure on default rates in the 1990s.

While mortgage lenders cannot control price appreciation, they can partially respond to debt burdens through careful underwriting and/or quality control. Loans very close to maximum ratios could be rejected. This would not prevent some borrowers from increasing their debt burdens sharply after a loan is made, but it would eliminate some high-risk borrowers that show a clear propensity to accumulate debt.

Lenders can also control LTV ratios. This factor will become even more important in the 1990s because price appreciation is likely to be much more modest. Stricter underwriting guidelines and quality control standards can be used to compensate for slower appreciation and to reduce defaults. Of course, reducing risk this way could be costly and involves business decisions. Some risk must be incurred to make a profit.

The outlook for the unemployment rate is for a modest increase in the next year or so, as the economy grows below its potential growth rate. Over the longer term, however, the outlook is much better than 15 years ago. The surge in new workers from the baby boom generation is over. As the economy continues to grow in the 1990s, a labor shortage is more likely to develop than a labor surplus. This can be mitigated by immigration or other factors, but the outlook for the long-run unemployment rate is favorable.

Finally, the divorce rate appeared to peak in the early 1980s and is now coming down. This trend is likely to continue into the 1990s and should help put downward pressure on default rates.

Taken together, there are obviously many conflicting factors that are likely to play a role in determining default rates in the 1990s. The possibility of a recession sometime within the next several years, which can't be ruled out, would almost certainly result in a temporary rise in default rates. Matters are complicated further by the ongoing resolution of the thrift crisis. Disposition of properties may depress appreciation rates in some areas, thereby increasing the probability of default, or may affect defaults in ways that are difficult to anticipate currently. Still, there are not compelling reasons to believe that default rates will rise sharply in the 1990s. On the other hand, there are not compelling reasons to believe that they will continue to come down either.

Lenders will play a key role. Tighter underwriting standards and quality control guidelines can make a difference. Whether such standards and guidelines justify their added costs may well determine the future course of default rates. [Charts 1 to 5 Omitted] [Tabular Data 1 to 3 Omitted]

(1)See Thomas M. Holloway and Robert M. Rosenblatt, "Problem Loans: Trends, Causes, and Outlook for the Future," economics department, Mortgage Bankers Association of America (MBA), April 1990 (mimeo) for an expanded version of this article with much more technical and statistical detail. (2)The estimates of "total" problem loans rates shown in Table 1 and Chart 1 and elsewhere in the article differ from estimates for the "all loans" category published in MBA's National Delinquency Survey (NDS) because the NDS includes a disproportionate number of mortgage bankers relative to other types of lenders. The estimates here adjust the NDS numbers to better reflect national averages. See the Holloway-Rosenblatt paper cited in footnote 1 for details.

Thomas M. Holloway is senior economist at the Mortgage Bankers Association of America (MBA). Robert M. Rosenblatt is an economist at MBA.
COPYRIGHT 1990 Mortgage Bankers Association of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Holloway, Thomas M.; Rosenblatt, Robert M.
Publication:Mortgage Banking
Date:Oct 1, 1990
Words:6231
Previous Article:Buying an umbrella.
Next Article:Detente & development.
Topics:


Related Articles
Economic trends.
Economic trends.
RESCUING Latepayers.
Mortgage delinquencies fall, foreclosure inventory rises. (Business Alert).
Delinquencies and foreclosures increase. (Business Alert).
Mortgage delinquencies decrease in first quarter. (Business Alert).
California mortgage delinquencies fall in QI.
N.H. mortgage delinquencies rise in 2Q.
Delinquencies, foreclosures increase during fourth quarter.
The big picture: everyone has been taking a glass-half-empty perspective on rising delinquency and foreclosure numbers. The real story is a bit more...

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters