This is the most thorough analysis of the mortgage meltdown I have found. While it does include “Speculators”, I left them out of the extractions below because the author ultimately concluded that they were just using the system made available to them to try to make a buck.
The last defense of banks trying to defend themselves against charges of engaging in biased mortgage lending appeared to fall when the Federal Reserve Bank of Boston (Boston Fed) conducted an apparently careful statistical analysis in 1992, which purported to demonstrate that even after controlling for important variables associated with creditworthiness, minorities were found to be denied mortgages at higher rates than whites.
In fact, the study was based on such horribly mangled data that the study’s authors apparently never bothered to examine them. Every later article of which I am aware accepted that the data were badly mangled, even those authored by individuals who ultimately agreed with the conclusions of the Boston Fed study. The authors of the Boston Fed study, however, stuck to their guns even in the face of overwhelming evidence that the data used in their study was riddled with errors. Ex post, this was a wise decision for them, even if a less than honorable one.
The winds were behind the sails of the study. Most politicians jumped to support the study. “This study is definitive,” and “it changes the landscape,” said a spokeswoman for the Office of the Comptroller of the Currency. “This comports completely with common sense,” and “I don’t think you need a lot more studies like this,” said Richard F. Syron, president of the Boston Fed (and former head of Freddie Mac). One of the study’s authors, Alicia Munnell, said, without any apparent concern for academic modesty, “The study eliminates all the other possible factors that could be influencing [mortgage] decisions.” When important functionaries make quotes like these, you know that the fix is in and that scientific enquiry is out.
My colleague, Ted Day, and I only decided to investigate the Boston Fed study because we knew that no single study, particularly a first study, should ever be considered definitive and that something smelled funny about the whole endeavor. Nevertheless, we were shocked at the poor quality of the data created by the Boston Fed. The Boston Fed collected data on approximately three thousand mortgages. Data problems were obvious to anyone who bothered to examine the numbers. Here is a quick summary of the data problems: (a) the loan data that Boston Fed created had information that implied, if it were to be believed, that hundreds of loans had interest rates that were much too high or much too low (about fifty loans had negative interest rates according to the data); (b) over five hundred applications could not be matched to the original HMDA data upon which the Boston Fed data was supposedly based; (c) forty-four loans were supposedly rejected by the lender but then sold in the secondary market, which is impossible; (d) two separate measures of income differed by more than 50 percent for over fifty observations; (e) over five hundred loans that should have needed mortgage insurance to be approved were approved even though there was no record of mortgage insurance; and (f ) several mortgages were supposedly approved to individuals with a net worth in the negative millions of dollars.
When we attempted to conduct a statistical analysis removing the impact of these obvious data errors, we found that the evidence of discrimination vanished. Without discrimination there would be no reason to try to “fix” the mortgage market. Nevertheless, our work largely evaporated down the memory hole as government regulators got busy putting the results of the Boston Fed study to use in creating policy. That policy, simply put, was to weaken underwriting standards. What happened next is nicely summed up in an enthusiastic Fannie Mae report authored by some leading academics (Listokin et al., 2002).
2. Relaxed Lending Standards—Everyone’s Doin’ It
Within a few months of the appearance of the Boston Fed study, a new manual appeared from the Boston Fed. It was in the nature of a “Nondiscriminatory Mortgage Lending for Dummies”3 booklet. The president of the Boston Fed wrote in the foreword:
The Federal Reserve Bank of Boston wants to be helpful to lenders as they work to close the mortgage gap [higher rejection rate for minorities]. For this publication, we have gathered recommendations on “best practice” from lending institutions and consumer groups. With their help, we have developed a comprehensive program for lenders who seek to ensure that all loan applicants are treated fairly and to expand their markets to reach a more diverse customer base.
Even the most determined lending institution will have difficulty cultivating business from minority customers if its underwriting standards contain arbitrary or unreasonable measures of creditworthiness.
The document continues:
Management should be directed to review existing underwriting standards and practices to ensure that they are valid predictors of risk. Special care should be taken to ensure that standards are appropriate to the economic culture of urban, lower–income, and nontraditional consumers.
Credit History: Lack of credit history should not be seen as a negative factor. Certain cultures encourage people to “pay as you go” and avoid debt. Willingness to pay debt promptly can be determined through review of utility, rent, telephone, insurance, and medical bill payments. In reviewing past credit problems, lenders should be willing to consider extenuating circumstances. For lower–income applicants in particular, unforeseen expenses can have a disproportionate effect on an otherwise positive credit record. In these instances, paying off past bad debts or establishing a regular repayment schedule with creditors may demonstrate a willingness and ability to resolve debts. Successful participation in credit counseling or buyer education programs is another way that applicants can demonstrate an ability to manage their debts responsibly.
Obligation Ratios: Special consideration could be given to applicants with relatively high obligation ratios who have demonstrated an ability to cover high housing expenses in the past. Many lower–income households are accustomed to allocating a large percentage of their income toward rent. While it is important to ensure that the borrower is not assuming an unreasonable level of debt, it should be noted that the secondary market is willing to consider ratios above the standard 28/36.
Down Payment and Closing Costs: Accumulating enough savings to cover the various costs associated with a mortgage loan is often a significant barrier to home ownership by lower–income applicants. Lenders may wish to allow gifts, grants, or loans from relatives, nonprofit organizations, or municipal agencies to cover
part of these costs. Cash–on–hand could also be an acceptable means of payment if borrowers can document its source and demonstrate that they normally pay their bills in cash.
Sources of Income: In addition to primary employment income, Fannie Mae and Freddie Mac will accept the following as valid income sources: overtime and part–time work, second jobs (including seasonal work), retirement and Social Security income, alimony, child support, Veterans Administration (VA) benefits, welfare payments, and unemployment benefits.
Credit scores. While credit scores can be an analytical tool with conforming loans, their effectiveness is limited with CRA loans. Unfortunately, CRA loans do not fit neatly into the standard credit score framework . . . Do we automatically exclude or severely discount . . .loans [with poor credit scores]? Absolutely not.
Payment history. While some credit-score purists might take issue with our comments in the preceding section, payment history for CRA loans tracks consistently close to the risk curves of conforming loans . . . In many cases, purchasing a home puts the borrower in a more favorable financial position than renting. It is quite common for a first-time homebuyer using a CRA loan to have been shouldering a rent payment that consumed 40 percent to 50 percent of his or her gross income.
When considering the credit score, LTV, and payment history, we put the greatest weight by far on the last variable . . . Payment history speaks for itself. To many lower-income homeowners and CRA borrowers, being able to own a home is a near-sacred obligation. A family will do almost anything to meet that monthly mortgage payment.
Where do most payment problems occur? Usually, the problems stem from poor upfront planning and counseling. Hence, one of the key factors we look for in a CRA portfolio is whether the borrower completed a GSEaccredited homebuyer education program. The best of these programs help the individual plan for emergencies that can arise with homeownership.
There are two points that need to be kept in mind. First, preliminary evidence (Mian and Sufi, 2008) indicates that the recent increase in defaults has been dominated by those areas populated by poor and moderate-income borrowers. Further, figure 9 (Share of Speculative and Subprime Loans by Census Tract Income), which will be seen later in this report, and the discussion surrounding it show that poor and moderate-income areas had the largest share of speculative home buying, and speculative home buying will be seen, later in this report, to be the leading explanation for home foreclosures.
Thus the evidence is that the foreclosures are disproportionately a problem of the poor and moderateincome areas, which is entirely consistent with the weakened underwriting standards discussed above. The fact that foreclosures among poor and moderate homeowners are not receiving the greatest amount of newspaper attention doesn’t mean that they are not at the epicenter of the foreclosure problem.
Second, although the original mortgage innovations were rationalized for low-and middle-income buyers, once this sloppy thinking had taken hold it is naive to believe that this decade-long attack on traditional underwriting standards would not also lead to more relaxed standards for higher-income borrowers as well. When everyone cheers for relaxed underwriting standards, the relaxation is not likely to be kept in narrow confines.
3. Empirics of the Current Crisis
The immediate cause of the rise in mortgage defaults is fairly obvious—it was the reversal in the remarkable price appreciation of homes that occurred from 1998 until the second quarter of 2006. Since then prices have sharply declined. The housing price bubble can be easily seen in figure 1, which shows inflation-adjusted housing prices since 1987.
If relaxed lending standards allowed more households to qualify for financing, basic economics also says that housing prices would have risen as the demand for homes increased. Some portion of the housing price bubble, perhaps a large portion, must have been caused by the relaxed lending standards.
Of course, relaxed lending standards, or underwriting innovations as it is euphemistically put, were so successful that standards were loosened across the board so that even a prime loan applicant could avoid making virtually any down payment by taking out a piggyback second mortgage to cover the down payment required by the first mortgage (often both mortgages were made by the same lender).
4. Problems with the Subprime Bogeyman Hypothesis
The bogeyman in the mortgage story is the unethical subprime mortgage broker who seduced unwary applicants out of their hard-earned, sacredly treated assets. This subprime bogeyman charged usurious rates for his mortgages and bamboozled his clients with artificially low teaser rates that allowed them to purchase homes that were unaffordable at realistic interest rates. This character has been pilloried by all manner of politician and pundit. Although a convenient scapegoat, this character does not actually appear to be responsible for the main part of the mortgage meltdown. This is not to say that there are not lying and cheating mortgage brokers—there are. But every profession, including economics, has its share of liars and cheaters.
There is an important problem with the hypothesis that evil subprime lenders caused the mortgage meltdown. That problem is the fact that subprime loans did not perform any worse than prime loans.
So, if there is no subprime bogeyman on whom the mortgage meltdown can be blamed, what’s a politician to do?
6. Conclusions
The “mortgage innovations” that are largely the federal governments responsibility are almost completely ignored. These “innovations,” heralded as such by regulators, politicians, GSEs, and academics, are the true culprits responsible for the mortgage meltdown. Without these innovations we would not have seen prime mortgages made with zero down payments, which is what happens when individuals use a second mortgage to cover the down payment of their first. Nor would we have seen “liar loans” where the applicant was allowed to make up an income number, unless the applicant was putting up an enormous down payment, which was the perfectly reasonable historical usage of no-doc loans (which require minimal financial documentation).
The political housing establishment, by which I mean the federal government and all the agencies involved with regulating housing and mortgages, is proud of its mortgage innovations because they increased home ownership. The housing establishment refuses, however, to take the blame for the flip side of its focus on increasing home ownership— first, the bubble in home prices caused by lowering underwriting standards and then the bursting of the bubble with the almost catastrophic consequences to the economy as a whole and the financial difficulties being faced by some of the very homeowners the housing establishment claims to be trying to benefit.
Hindsight is the best sight, they say. Unfortunately, the housing establishment and our political leaders seem intent on not learning from the past. Hopefully this report can help move the debate in a direction that will allow for more productive learning.
*Stan J. Liebowitz is Research Fellow at the Independent Institute and the Ashbel Smith Distinguished Professor of Managerial Economics at the University of Texas at Dallas. Anatomy of a Train Wreck is included in the forthcoming Independent Institute book, Housing America, Building out of a Crisis, edited by Randall G. Holcombe and Benjamin W. Powell(http://www.independent.org/store/book_detail.asp?bookID=76).
