The Theory of Financial Literacy
I commend the Chairman for his long efforts towards increasing financial literacy. I believe we all share a desire to see consumers make better and more informed choices. We must, however, when evaluating public policy remember that intentions and outcomes are not the same thing.
Too often in Washington policy discussions confuse ends and means. Financial education is, at heart, a means to improving financial literacy, the purpose of which is not simply to increase knowledge but to improve household decision-making and behavior. We can, and do, spend considerable amounts on a variety of financial education efforts, as does the financial services industry. State and local governments also commit a considerable amount of resources to financial education, particularly in the form of classroom hours and the compensation and time of educators. Dollars, or hours, should not be our measure of success. They are a measure of cost (and only one measure as teaching hours spent on financial education are not spent on other teaching). The true measure of success is whether households are making good financial decisions and behaving in a responsible manner. The notion that a more informed consumer makes a better one is appealing, but it also a notion lacking in concrete guidance. Whether financial education and literacy programs actually make a “better” consumer is ultimately an empirical question.
Empirical Evaluation of Financial Literacy Programs
The good news is that a variety of financial education programs have received evaluation, even if most have not. There are also a small number of literature surveys providing an overview.1A recent literature review by economists at the Federal Reserve Bank of Cleveland provides what I believe is a fair and representative conclusion: “the literature does not succeed in establishing the extent of the benefit provided by financial education programs, nor does it provide conclusive support that any benefit at all exists.”2
Another review, focusing on financial education at the high school level, concludes, “The findings indicated that those who took the course were no more financially literate than those who had not. In addition, those who took the course did not evaluate themselves to be more savings-oriented and did not appear to have better financial behavior than those who had not taken the course.”3
While I am, admittedly, a skeptic of the effectiveness of government programs in general, even I have been surprised at the extent to which evaluations of financial education programs have generally failed to find significant effects.
Given the opportunity costs of financial education, both in time and money, the failure to find consistent positive effects would be bad enough, however a small number of studies have actually found negative effects. One researcher, for instance, found that among high school seniors who paid for their own car insurance, those who took a financial literary course actually did worse when tested about car insurance than students who did not take such a course.4
Studies have also found on adults that financial literacy courses can, for some, increase the consumer’s confidence without actually increasing their knowledge. This is perhaps my area of greatest concern regarding financial education. Ultimately we do not want consumers to be either too overconfident or too under-confident. Overconfident can reduce a consumer’s willingness to further investigate the characteristics and performance of various financial products. Overconfidence can also bias consumers toward under-estimations of risk. For instance researchers have found that overconfidence correlates with excessive stock trading, leading to lower investment returns.5Overconfidence likely also plays a role in the generally superior investment performance of women relative to men.
On the other hand, under-confidence can dissuade consumers from entering into financial transactions that would improve their welfare. What we ultimately want is for consumers to have an unbiased and accurate representation of the individual risks (and rewards) that they face when engaging in various financial transactions.
Given that consumers already appear to grossly exaggerate their own credit quality and financial literacy6, as a public policy matter, we need to be concerned about the impact of financial education on making consumers believe they are more knowledgeable then they actually are.
To summarize, despite some 56 programs running across 20 agencies, some of which have received funding for decades, there is little concrete evidence that said programs have improved consumer welfare.
Case Study: Housing Counseling
As the title of today’s hearing makes clear, one of the objectives of financial literacy could be to avoid financial crises. Despite the conventional wisdom, the last decade witnessed booms in a variety of asset classes, including various segments of the real estate market. The boom was not simply in housing. The commercial, retail and multifamily real estate markets also went boom and burst. These busts generally occurred before the decline in the housing market, removing any question as to causality. All that said, housing did play a special role in the financial crisis and the subsequent bailouts. Accordingly if there is one area where financial literacy could have helped mitigate the crisis, it is in the area of housing counseling.
Housing counseling has also been one of the most highly funded and researched areas of financial literacy. Congress first authorized funds for housing counseling in 1969. Funding grew significantly in the mid-1970s, then decline and stayed relatively flat until about 1991. HUD appropriations increased rather dramatically in 1990s, jumping almost 400% between 1991 and 2001. Funding continued to increase. In fact some of the largest increases were in the years just preceding the peak of the housing market. In just the fiscal year of 2003, HUD funding for housing counseling doubled from about $20 million to $40 million, later increasing to $50 million in FY 2008.
The 1990s and 2000s also witnessed a significant increase in HUD-approved counseling agencies. One should bear in mind that not all organizations providing counseling are HUD-approved. For instance HUD does not approve for-profit or for-profit sponsored organizations. So the figures below would include housing counseling provided by financial institutions.
What should be clear from the preceding is that the years prior to the bursting of the housing bubble and the subsequent financial crisis, were years in which an ever increasing amount of resources were devoted to housing counseling. As Social Science lacks the luxury of conducting natural experiments, we cannot say with any certainty that the housing crisis would have been worse, or how much worse, if we had not spent $100s of millions in housing counseling. What we can say, with certainty, is that spending a few $100 million on housing counseling did not stop a financial crisis from occurring.
We also know that the several $100 million spent on housing counseling by HUD was only a small part of the funding for agencies receiving said funding. For HUD approved agencies, HUD counseling funds averaged 13.5% of their budgets in FY07. This would that in FY07, at least $400 million was spent in total on housing counseling from all sources. In the immediate years preceding the crisis, it is likely that total funding sources for housing counseling exceeded a $1 billion totaled over those years.
According to HUD the average cost of housing counseling was over $400 per person counseled in FY07. Ten percent of agencies actually had average per client costs in excess of $1,000. This figure is particularly astounding when one considers that only about half of the agencies used their own materials, relying instead on material and courses developed by others.
The intent of housing counseling should be helping potential homebuyers receive unbiased and accurate information. Housing counseling should also help potential homebuyers develop a plan to become ready for home-buying. In this sense, it is not clear that housing counseling is reaching the right consumers at the right time. Only about a fourth of clients were deemed to be “near ready” in terms of making a home purchase. Over forty percent of clients were deemed not ready for purchase for at least six months, raising the issue of how much material clients would retain six months hence.
A common rationale for the use of non-profit housing counseling is that such avoids the potential conflict-of-interest that can arise when financial education is being provided by a for-profit business entity. Unfortunately HUD surveys indicate that HUD-approved non-profit counselors were heavily dependent upon members of the real estate and mortgage industry.
Almost 80 percent of non-profit housing counselors used mortgage lenders in their workshops, while over 70 percent used real estate agents. While there is some obvious advantage to using knowledgeable industry representatives to educate, it does raise the critical question of whether housing counselors were doing little more than prepping and steering consumers toward select lenders and real estate agents.
In terms of effectiveness, evaluations of housing counseling have also been mixed, but have generally shown more success than other forms of financial education. While some researchers have found no effect of counseling on default rates, these researchers did some improvement in choice of mortgage characteristics, although their measure was somewhat subjective.7Other researchers have found that the form of counseling greatly matters, where intensive one-on-one counseling reduces default but soft-touch counseling was largely ineffective.8Of course part of the effect of intensive counseling could be driven by screening, that is marginal borrowers are dissuaded from the loan due to the time and cost of counseling.
To summarize, we spent a considerable amount on housing counseling for years prior to the crisis with no evidence that such minimized the severity of the crisis. There is actually some evidence to suggest it might have made the crisis slightly worse. There is also some evidence to suggest that housing counseling served more as a vehicle for connecting borrowers with the mortgage and real estate industry than as a method for arming borrowers with relevant knowledge. There is no evidence that counseling instilled potential borrowers with skepticism about homeownership.
Knowledge versus Incentives
My primary concern with linking financial literary to the recent financial crisis is that it distracts from much needed changes in our financial regulatory system, not to mention our monetary system. At the risk of over-generalizing, I do not believe we had a financial crisis due to a lack of financial literacy. I believe we had a financial crisis due to very perverse incentives in our financial system that encouraged excess risk-taking on the part of lenders and borrowers, while also reducing incentives for appropriate due diligence on the part of investors and creditors.
Going back to the mortgage market, when borrowers were required to put little, if any equity, into a home purchase, and the loans were generally non-recourse, is it any surprise that such borrowers defaulted when prices declined. In fact there is evidence that borrowers who had received counseling were more likely to engage in strategic default, ultimately increasing the level of foreclosures, rather than reducing it.9
In a well-functioning market, lenders have strong incentives to provide borrowers with the appropriate information that would reduce default. Unfortunately we do not have well-functioning financial markets. We have markets characterized by extensive government guarantees and moral hazard. If lenders do not face the true and full risk of their actions, then their incentives to appropriately manage risk and effectively educate consumers is reduced, if not eliminated. Whether it is the presence of deposit insurance or the ability to transfer mortgage credit risk to the taxpayer via the government sponsored enterprises and the Federal Housing Administration, lenders do not face appropriate incentives for risk-taking. These issues are only compounded when lenders face extensive penalties for not extending credit to risky borrowers if such borrowers are members of a protected class.
Is it irrational or uninformed for lenders and borrowers to become highly leveraged when our tax code subsidized debt relative to equity? Or when the Federal Reserve maintains negative real rates for several years, as was the case in 2002 to 2005? An extremely steep yield curve, as engineered by the Federal Reserve, also encourages maturity mismatch, both on the part of borrowers and lenders, which increase financial fragility.
We should also be clear that the Dodd-Frank Act does not fix our financial system. Too-big-to-fail and moral hazard are bigger problems today then before the financial crisis. So while I again commend the Chairman’s efforts, we must not lose sight of the urgent need for reforming our financial regulatory system.
Substance over Form
Financial education is only going to be as good as the information that is imparted. A very basic question should be: what is it that consumers do not know? If financial education focuses on minor or irrelevant issues, such as the impact of “pulling” a credit report on one’s credit score, to the exclusive on central issues, like the impact of timely debt payment on one’s credit score, then consumers could easily be worse off from counseling.
Counseling also runs the risk of having its substance driven by the bias of both providers and government. Take the largely positive image of homeownership presented by many housing counselors or the negative image presented of mortgage prepayment penalties. Both images are far more driven by bias than fact. From my own experience at HUD, I watched lawyers drive mortgage disclosure in such a way that harm consumers because the government lawyers were convinced that mortgage brokers were inherent “bad”. Efforts at financial education have to devote more attention to the substance of such, rather than the form.
Conclusions
The federal government, along with state/local governments and the private sector, fund a variety of financial education efforts. The research and evaluation literature has failed to find strong, consistent effects for these efforts. In some circumstances even negative effects have been found. Housing counseling has been a particularly well funded and researched area. Even here the rests, while better than most counseling, are mixed. It should also be clear that significant funding for housing counseling did not help us avoid a financial crisis in which housing finance played a unique role. It is my contention that financial literacy, at least on the part of consumers, was at most a minor factor in the recent financial crisis and that failings in our monetary and regulatory systems played much greater roles. These failings have not been addressed and continue to pose significant risk to our financial markets, broader economy and ultimately the taxpayer.
I thank the Subcommittee for your attention and the opportunity to offer my perspective.
Notes
1See generally, Angela Lyons et al., “Are We Making the Grade? A National Overview of Financial Education and Program Evaluation,”. Journal of Consumer Affairs 40(2006):208. and Tzu-Chin Martina Peng et al., “The Impact of Personal Finance Education Delivered in High School and College Courses,” Journal of Family and Economic Issues 28(2007) and Lauren Willis, “Evidence and Ideology in Assessing the Effectiveness of Financial Literacy Education,” San Diego Law Review 46(2009).
2Ian Hathaway and Semeer Khatiwada, Do Financial Education Programs Work? Working Paper 08–03. Federal Reserve Bank of Cleveland. April 2008.
3Lewis Mandell and Linda Klein, “The Impact of Financial Literacy Education on Subsequent Financial Behavior,” Journal of Financial Counseling and Planning Volume 20, Issue 1 (2009). http://www.afcpe.org/assets/pdf/lewis_mandell_linda_schmid_klein.pdf
4Lewis Mundell, “Does Just-in-Time Instruction Improve Financial Literacy?,” Credit Union Magazine, January 2006.
5Brad Barber and Terrance Odean, “Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,” Journal of Finance 55(2000).
6Venessa Gail Perry, “Is Ignorance Bliss? Consumer Accuracy in Judgments About Credit Ratings,” Journal of Consumer Affairs 42(2008).
7Jonathan Spader and Roberto Quercia, “Does Homeownership Counseling Affect the Prepayment and Default Behavior of Affordable Mortgage Borrowers?” Journal of Policy Analysis and Management 27(2008).
8Abdighani Hirad and Peter Zorn, Prepurchase Homeownership Counseling: A Little Knowledge Is A Good Thing, in LOW-INCOME HOMEOWNERSHIP: EXAMINING THE UNEXAMINED GOAL. 2002.
9Valentina Hartarska and Claudio Gonzales-Vega, “Credit Counseling and Mortgage Termination by Low-Income Households,” Journal of Real Estate Economics and Finance 30(2005).
