Recent Changes in Household Finances and Home Lending

It is a pleasure for me to join you today. Since we last met--almost exactly three years ago--the U.S. economy has experienced substantial changes. Over the next few minutes, I will focus on recent changes in the economic status of minority Americans, drawing on information from the Federal Reserve's 2004 Survey of Consumer Finances. The data from the 2004 survey are not yet public, but I expect them to be released by the end of the first quarter of next year. I will then turn to some recently published findings from staff research on home lending that used the latest data collected under the Home Mortgage Disclosure Act. That discussion will concentrate on patterns of lending across racial and ethnic groups.

Economic Background
First, let me set the broad economic background. After a relatively mild economic downturn in 2001, the U.S. economy grew at moderate pace in 2002 and then more vigorously in the succeeding years. At first, much of the step-up in economic activity was achieved through productivity gains rather than through the addition of more workers. As a consequence, the unemployment rate continued to rise, peaking at 6.5 percent in mid-2003. Since then, however, employment has been expanding apace, and the jobless rate has declined to about 5 percent. In asset markets, most major stock market indexes have posted only small positive gains since 2001, while housing prices have risen substantially. The slow initial recovery in the labor market and the stock market, combined with rising house values, has led to an uneven distribution of the gains from recent economic growth.

Changes in Family Income
As part of our continuing effort to understand the American economy, my colleagues and I at the Federal Reserve take an active interest in tracking the changing financial circumstances of U.S. families. The Board's Survey of Consumer Finances (SCF), which is conducted every three years, provides the most comprehensive information available on the balance sheets of American families and is a fundamental source of information about many issues of personal finance.

In addition to collecting information on the assets and debts of families, the SCF collects information on family income. Data from the most recent survey show that, from 2001 to 2004, real (that is, inflation-adjusted) median and mean family incomes were little changed--a small gain in the median and a small decline in the mean. In particular, the gap between the median income of families in which the primary racial identification of the survey respondent was black and the median income of families in which the primary racial identification of the survey respondent was non-Hispanic white remained quite large. Specifically, the median income of black families in 2004 was about $29,000--only 58 percent of the median for non-Hispanic white families. Although this ratio has increased over the past decade, the level of inequality between the two groups continues to be a concern. Lower levels of income make it more difficult for black families to acquire assets and to create wealth.

Changes in Family Wealth
Income is an essential measure in gauging the ongoing health of family finances, but net worth is, in the long term, an even more important yardstick. Compared with the significant gains observed in real median and real mean net worth from 1998 to 2001, changes over the 2001-04 period were subdued. Over that more recent period, the median net worth of all families in real terms was essentially unchanged, while the mean net worth increased about 6 percent. A rise in the mean relative to the median indicates that the gap in net worth between households in the upper part of the wealth distribution and other households has widened. At the same time, the difference in median wealth between non-Hispanic white families and black families widened even further. For non-Hispanic white families, median net worth increased about 6 percent, to $136,000, while for black families the median was $20,000, a level essentially the same as it was in 2001. However, the mean net worth of black families rose substantially from 2001 to 2004, increasing 37 percent compared to an increase of 8 percent for non-Hispanic white families. I'll return to this difference in the mean and median in a moment.

A substantial part of the wealth gap between black families and non-Hispanic white families is associated with a difference between the two groups in their ownership of assets. In 2004, about 16 percent of black families had no type of financial asset--that is, no checking or savings account and no stocks, bonds, or retirement accounts; this percentage was essentially unchanged from 2001. The percentage of black families that have no type of nonfinancial asset--for example, that do not own a home, a vehicle, or a business--declined from about 24 percent in 2001 to about 20 percent in 2004. The comparable figures for non-Hispanic white families were below 5 percent in 2004 for both financial and nonfinancial assets.

The increase in ownership of nonfinancial assets for black families occurred primarily in residences, other real estate, and privately held businesses. Because none of these types of assets are owned by a large share of black families, any wealth gains arising from ownership of these assets will not be widely distributed across black families. The relatively narrow ownership of these assets may be one reason why we saw a large increase in mean net worth coupled with no change in median wealth from 2001 to 2004. Let me now provide more detail on two of these nonfinancial assets--businesses and residences.

Changes in Business Ownership
The proportion of black families that own a business is only about one-third that for non-Hispanic white families, but the proportion for black families did increase somewhat from 2001 to 2004. As evidenced by the data in the SCF, business ownership is an important avenue for wealth creation. For example, the median net worth of black families with business assets was about $174,000 in 2004, a level more than eight times the median net worth for all black families. Similar results hold for income: The median for black families with business assets was $68,000 in 2004, a level more than twice that for black families as a whole. As further evidence of the importance of business assets in wealth creation, the data show that there is far less inequality in the median net worth and income between black and non-Hispanic white business owners than between black and non-Hispanic white families overall.

Changes in Homeownership
Black families made progress in homeownership from 2001 to 2004. Indeed, for most homeowning households, regardless of race or ethnicity, the home is their largest and most important asset. Homeownership is one of the cornerstones of wealth creation and is generally associated with a range of socially desirable outcomes, including better schools, less crime, and neighborhood stability. For these and other reasons, increasing the rate of homeownership has been a longstanding national priority.

Over recent years, factors such as low interest rates, a growing economy, and rising employment have helped spur homebuying by all segments of society. In addition to favorable economic conditions, the goal of increasing homeownership has been helped by introduction of new information processing technologies, including credit scoring, that have reduced the costs of obtaining a mortgage and by the introduction of a host of new, more affordable mortgage products.

Data from the 2004 SCF reveal a milestone for homeownership among black families--the rate of homeownership surpassed 50 percent. For black families that owned homes, the median home value in 2004 was about 18 percent larger than the median in 2001. About 75 percent of non-Hispanic white families own a home, but the increase in the rate of homeownership for black families outpaced that for non-Hispanic white families by a factor of two to one from 2001 to 2004.

The rise in the homeownership rate is important for wealth creation, but there has been a concern that increased levels of home-secured debt--such as mortgages, home equity loans, and home equity lines of credit--could negate any gains from the rise in house values that occurred over the past few years. One way to address this concern is to examine the amount of equity that families have in their homes--that is, the difference between the value of the home and any debt secured by it. For black homeowners, the median value of home equity increased 24 percent from 2001 to 2004, to $45,000; the median also increased for non-Hispanic white homeowners, but the percent change was about one-half of the increase of black homeowners; nonetheless, the level of home equity is much higher for non-Hispanic white homeowners than for other homeowners.

I do not have time today to cover the full range of findings on asset and debt holdings from our 2004 Survey of Consumer Finances. But, clearly, the survey is an important source of information on how changes in financial markets, innovations in lending, and other factors affect the financial situation of families. Our staff is preparing an article for publication in the Federal Reserve Bulletin early next year, and it will provide a fuller review of the results. Publication of the article will be followed shortly thereafter by the public release of the data.

Development of the Higher-Priced Segment of the Mortgage Market
Given the importance of homeownership for wealth creation and for other desirable economic and social outcomes, I would like to spend my remaining time on some interesting developments in mortgage lending. As I noted, advances in information processing technology have played an important role in expanding access to credit and homeownership opportunities. In the past, individuals seeking credit, whether to purchase a home or for some other reason, either did or did not meet the specific underwriting criteria for a particular loan product; all who did meet the criteria paid about the same price--that is, interest rate--for the loan. Today, in part because of advances in credit scoring technology, lenders use explicit risk-based pricing, charging different borrowers different prices depending on their estimated creditworthiness. Now, less-creditworthy borrowers, or those unwilling or unable to document their incomes, can qualify for a loan, but at a higher price.

The higher-priced segment of the mortgage market, often referred to as the subprime market, has been growing rapidly, from less than 5 percent of all mortgage lending in 1994 to about 19 percent in 2004. As significant price variability has emerged in the mortgage market, so have concerns about the fairness of pricing decisions by lenders and the adequacy of competition in the subprime market.

Origins and Recent Changes in the Home Mortgage Disclosure Act
The Congress enacted the Home Mortgage Disclosure Act (HMDA) thirty years ago. This law requires most home lenders to disclose selected information about the applications they receive and the loans they extend each year. HMDA was enacted to address concerns about redlining and unfair treatment of mortgage borrowers. Importantly, HMDA is a disclosure law. It does not direct lenders to make loans nor does it proscribe lender behavior. Rather, it helps to identify potential market failures, including racial, gender, or other illegal discrimination, by shining a bright light on lender behavior and by facilitating enforcement of the nation's fair lending laws.

In 2002, the Federal Reserve Board amended its regulations that implement HMDA to expand the types of information that lenders must disclose to the public about their lending activities. The amendments were intended to improve the quality, consistency, and utility of the reported data and to keep the regulation in step with recent developments in home-loan markets, particularly growth in the subprime portion of the market. Most prominently, the new information provides the first publicly available loan-level information on loan pricing in the higher-priced segment of the home-loan market. Data released to the public this past September covering lending activity in 2004 are the first to reflect the changes in the reporting rules.

As you may have seen in recent news accounts, much of the initial public interest in the data has focused on the loan pricing information, particularly on who receives higher-priced loans and on a comparison of prices paid by borrowers of different races and ethnicities.

Under the Fed's regulations, lenders do not report the actual interest rate paid by a borrower but instead report whether and by how much the loan rate exceeds designated thresholds established by the Federal Reserve Board. Loans with rates above the thresholds are referred to as higher-priced loans. These thresholds were carefully selected to limit data reporting, which is costly, to that segment of the home-loan market that has been raising the biggest concern.

Findings from the Expanded HMDA Data
I would like to share with you some of the more interesting and important findings from the recently released HMDA data. These findings were highlighted in an article by Federal Reserve researchers that was published in the Summer 2005 issue of the Federal Reserve Bulletin and is available on the Fed's web site.

The new HMDA data show that most home lenders make few if any higher-priced loans; only 500 out of the 8,850 reporting institutions made as many as 100 or more higher-priced loans last year. As a measure of the concentration of the business of higher-priced lending, the 10 lenders with the largest volume accounted for about 40 percent of all such loans.

The data also show that 16 percent of borrowers took out higher-priced loans in 2004 in the nation as a whole, but the proportion of borrowers obtaining higher-priced loans varied widely by region and by city. For example, across many of the metropolitan areas in the south and southwest, 30 percent to 40 percent of homebuyers taking out conventional loans last year took out higher-priced loans, but in other areas of the country, the proportion was much smaller. For example, in San Francisco only 2 percent of homebuyers obtained higher-priced loans; in the Washington, D.C., area the proportion was 8 percent.

Racial and Ethnic Differences in the Incidence of Higher-Priced Lending
The incidence of higher-priced lending also varies substantially across racial and ethnic lines. Blacks and Hispanics are much more likely than non-Hispanic whites, to receive higher-priced loans, and Asians are less likely than non-Hispanic whites to receive such loans. For example, in 2004, 32 percent of black borrowers received higher-priced home-purchase loans, but only 9 percent of non-Hispanic white borrowers did. In other words, black homebuyers received higher-priced loans more than three times as often as non-Hispanic white homebuyers.

Certainly, differences of this magnitude are eye-opening and have led some to conclude that racial discrimination must play an important role in the pricing of home loans. However, before one reaches such a conclusion, one must take note of the inherent limitations of the reported data. Although the HMDA data include information on borrower income, loan amount, and property location, the data do not include many factors that lenders routinely consider in loan underwriting and pricing--because lenders are not required by the law to report those factors. Perhaps the most important credit-risk factors that are not included in the HMDA data are credit scores, loan-to-value ratios, and debt-to-income ratios.

Because the HMDA data lack such information, any analysis that relies on the HMDA data alone and fails to recognize the data's inherent limitations could lead to inaccurate conclusions about the causes of the racial gaps I just mentioned. It would be unfortunate, not just for lenders but also for potential borrowers, if unwarranted accusations of discrimination, stemming from improperly analyzed pricing differences, discouraged lenders from making higher-priced loans to higher-risk borrowers. Risk-based pricing has greatly expanded the availability of credit to borrowers who, because of weaknesses in their credit profiles, had previously been unable to qualify for home loans. Unwarranted tarnishing of the reputations of lenders operating lawfully and in good faith would run the risk of curtailing that progress.

Despite the limitations of the data, they do tell us something significant about the racial gap in the rate of receiving higher-priced loans. They tell us that a major reason that black and Hispanic borrowers are much more likely than non-Hispanic white borrowers to obtain higher-priced mortgage loans is the simple fact that black and Hispanic borrowers are much more likely to obtain mortgage loans from institutions that tend to specialize in higher-priced, or subprime, lending.

On one hand, this segmentation of the market may reflect objective differences among borrowers in product preferences or in the sort of indicators of credit risk, such as credit scores, that I have just explained are not included in the HMDA data. On the other hand, the segmentation may be more troubling. Lenders that focus most of their business on prime, or lower-rate, lending may be either less willing or less able to serve minority borrowers or neighborhoods. An equally troubling possibility is that the segmentation may stem, at least in part, from a steering of borrowers to lenders that charge higher prices than the credit characteristics of the borrowers warrant.

Of course, other explanations for the segmentation also need to be considered. Some borrowers lack financial education, and some may be unable or unwilling to shop around for the best loan or to negotiate the terms of their loans.

It is important to determine which of these alternative explanations, or perhaps other explanations, account for the fact that blacks and non-white Hispanics are more likely than others to obtain their home loans from higher-priced lenders. I hope that the new data stimulate more research into the workings of the higher-priced segment of the mortgage market.

Enforcement of Fair Lending Laws
Of course, the data are used not just for public-policy research but also for enforcement of the fair lending laws, which prohibit racial and ethnic discrimination in lending. The new data provide regulators with valuable information that they can use to screen institutions for possible violations of the law. For example, our analysis of the data at the level of the individual lender suggests that, after accounting for factors included in the HMDA data, about 2 percent of the 8,850 lenders covered by HMDA in 2004 were more likely, by a statistically significant amount, to extend higher-priced loans to black and Hispanic borrowers than to non-Hispanic white borrowers. The names of these institutions have been passed along to the appropriate regulatory agencies for their follow-up. In the case of the Federal Reserve, each institution we are responsible for examining that fell into that group has already been contacted, and their lending activities and loan files will be reviewed in depth. Examiners will determine whether the statistical differences can be explained by legitimate differences among borrowers that cannot be detected with the HMDA data or whether the statistical differences amount to violations of the fair lending laws.

Though much of the focus right now is on the higher-priced segment of the mortgage market and on lenders that make higher-priced loans, the agencies examine all institutions for compliance with the fair lending laws. They examine them for unlawful price discrimination in the prime segment as well as in the subprime segment of the mortgage market. The agencies also examine mortgage lenders for redlining and for unlawful steering of potential borrowers to higher-priced loans.

Clearly, enforcement of our nation's fair lending laws is critical. However, I would like to emphasize that in the long run the best defense against unfair practices is ensuring vigorous competition for the business of knowledgeable consumers. Increased competition will best ensure that informed consumers obtain the loan products most appropriate to their circumstances. In this regard, I believe that enhanced efforts to promote financial education and literacy are crucial steps in preparing individuals for the homebuying process. Equally important are efforts to promote financial literacy among our children so they realize at an early stage the importance of managing their finances properly and accumulating the savings they will need to meet life's ups and downs.

Conclusion
In closing, I think that all of us understand that the issues I have been discussing go to the heart of efforts to achieve equal opportunity for all Americans. We have made progress, we must make more. I hope that my remarks have helped clarify some important points. Once again, it has been my pleasure to meet with you today.

Return to top

2005 Speeches