A pattern of high delinquencies is starting to emerge for affordable lending mortgages made early in this decade. Lenders are concerned about the potential fallout, as delinquencies mean higher servicing costs. The trend casts a shadow over future profits.
Good household credit has traditionally been one of the foundations of the American economy. Indeed, this economy is driven more by growth in consumer spending than business spending. But bad household credit is slowly being woven into the cultural fabric of the 1990s. Even as relaxed lending standards have yielded a bumper crop of bankruptcies, a parallel decline has occurred in mortgage credit. The ultimate impact of this recent erosion in mortgage credit quality has yet to be seen. But it hangs over the mortgage industry like a looming cloud of red ink.
Yet, perhaps the bigger story is not the deteriorating ability of Americans to pay their household and mortgage debt, but the fact that the indicators suggest the problem is not spread evenly throughout the broad population. The problem of bad household and mortgage credit is, instead, occurring more often in households with low and moderate income and in households across the income spectrum with little or no equity in their homes.
This suggests that the blossoming of bad credit may not so much represent a new casualness among all Americans about the obligation to repay their debts, as it represents a household response to real economic stress - and inadequate savings to cushion those financial blows.
Indeed, lenders of all kinds used to take extraordinary steps to ensure that the cushion was there in advance of offering to extend credit. But today new forces drive lenders to attempt to extend credit and underwriting considerations are no longer in the driver's seat. Today the onus is on lenders to make the loan, do the business, get the young homeowner in the home - overcome the down payment obstacle. In short, banking regulators, Congress, community activists, local governments and public opinion are telling lenders to find a way to lend the money. Whether the cushion is there is the lender's problem - and it's become secondary at best.
This zeal to walk the tightrope without a net and beat down real odds like lack of funds to close and lack of cash reserves to make routine household repairs has led the mortgage industry into one of the most aggressive chapters in credit risk experimentation since the invention of the ARM loan. Today, roughly three years into this experiment, the mortgage industry is now encountering delinquency data that suggests the envelope was pushed so far it has started to tear at the seams.
The most visible example of the experiment is the affordable housing programs developed by the two government-sponsored enterprises (GSEs), Fannie Mae and Freddie Mac. Directed by Congress by a law passed in 1992, Fannie and Freddie embraced their new goals to have a sizable piece of their business come from low- and moderate-income borrowers and certain underserved housing markets. Because these two corporations are so huge and represent an immense slice of the mortgage market, their goals generated a sizable quantity of affordable lending originations. And those loans are today sitting in the servicing portfolios of mortgage bankers across the country.
To generate this segment of their business, the GSEs positioned their underwriting guidelines in a new light for lenders. Their mantra in the early '90s thus became: "There's flexibility in the guides." But when early evidence of delinquency problems associated with this new underwriting flexibility began to emerge, the new mantra became "no loans with credit scores under 620." The new technology of credit scoring had come onto the scene midway through the experiment, and it gave Fannie and Freddie a way to look into the future for the loans they bought. Clearly they saw a lot of bad loans coming.
Then the pendulum began to quickly swing from openly encouraged flexibility to pinpoint precision numbers (or credit scores), and suddenly there was no flexibility. Not only wasn't there any flexibility, it was at the lenders' own financial risk if they sold to the GSEs loans that didn't fall in a specified range of credit scores. All this was precipitated from the loan performance fallout following the grand affordable lending experiment.
What have we learned from this chapter? It becomes crucial for the industry and its policymakers to answer this question when you examine the demographic trends. The demographics suggest that there will be even greater need for affordable lending programs in the near future if America is to preserve its cornerstone tradition of widespread homeownership.
This article examines some of the most recently available loan performance data, attempts to isolate how affordable loans have behaved and outlines the lessons industry experts say can be learned from all this.
Which borrowers are having difficulties?
The difference in credit quality in loans with various levels of borrower equity can be seen in data collected by the Mortgage Information Corporation (MIC) of San Francisco. The company tracks the performance of more than 20 million outstanding loans. Its data base captures approximately one-half the total dollar volume of mortgage loans outstanding in the nation.
MIC found that loans made in 1991 with more than 80 percent loan-to-value (LTV) had a seriously delinquent rate (90-plus days late or in foreclosure) of 34 basis points after seasoning to June 1992. That rate fell to 15 basis points for the 1992 book of business and rose to 17 basis points for 1993 loans with similar seasoning. It then jumped to 32 basis points for 1994 loan before rising to a very worrisome 50 basis points for 1995 originations seasoned through June 1996.
This pattern of deterioration of mortgage credit since 1993, however, is not apparent in loans with 80 percent LTV or less. Seriously delinquent lower LTV loans represented 19 basis points of the 1991 book of business after seasoning to June 1992. That fell to 6 basis points in 1993 and rose to only 15 basis points for the 1995 originations seasoned to June 1996, making 1995 a better year than the recession year of 1991 for such loans.
The data suggests that mortgage holders with a fair amount of equity are improving their household finances while credit continues to deteriorate in households with little or no equity.
The divergence in household credit performance between lower and higher income households is captured in the most recent Federal Reserve Surveys of Consumer Finance, which are conducted every three years. Debt burdens for households earning $30,000 to $50,000 rose steadily from 11 percent in 1983 to 17 percent in 1983 and then rose still more to 20 percent in 1992. (The 1995 survey has not been released by the Fed.)
During the 1980s, the debt burden for households earning $50,000 to $100,000 at first rose in a similar manner to the lower income households, rising from 12 percent in 1983 to 17 percent in 1989. But, in 1992 the trends between the two income groups diverged. In that year the debt burden for $50,000 to $100,000 group declined modestly even as it was continuing to rise significantly for the $30,000 to $50,000 income group.
The income gap
The debt burden for lower income families has increased primarily because real earnings for Americans in the two lowest quintiles of the population have declined during the last 20 years while those in the two highest quintiles have risen during the same period, widening the gap, observes the Joint Center for Housing Studies at Harvard University in its The State of the Nation's Housing for 1996.
The gap is expected to widen in the future, according to Mark Zandi, the chief economist at Regional Financial Associates in West Chester, Pennsylvania. This is the case, he says, because incomes for lower socioeconomic groups in the United States will continue to come under downward pressure while the best-educated in society continue to improve their incomes.
Zandi attributes the pressure on lower incomes to the increasing internationalization of the economy, where declining tariffs and trade barriers mean that American workers compete more directly with third-world workers in the same industry. It also is influenced by the continuing high levels of legal and illegal immigration that swell the labor pool of unskilled workers in the United States. A greater supply of labor drives down the price of labor. Even if illegal immigration is curtailed, legal immigration is expected to continue at record numbers, Zandi says, with 10 million this decade and 9.5 million in the next decade. These forces will keep the downward pressure on the wages of unskilled American workers and continue to expand the income and credit quality gap that is emerging, he adds.
This ongoing skewing of income between lower and higher income Americans is expected to increase the need for more affordable housing programs as fewer and fewer young families are expected to be able to qualify for conventional mortgages. This dilemma has been carefully delineated in the Joint Center's 1996 housing report. The center found evidence of the growing socioeconomic gap in American society in the statistics that track a broad secular decline in homeownership among younger families and nonwhite minorities during the past two decades.
Between 1975 and 1982, the median income of young (aged 25 to 29) married-couple renters fell from $27,139 to $23,866, according to the Joint Center's report, which contains the center's tabulations of data from the American Housing Survey adjusted by the Current Population Survey of the U.S. Census Bureau. After 1982, income for young married couples aged 25 to 29 slowly rose to $26,600 in 1989. Then, after declining to $23,700 in real income in...