The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Banking and Community ISSUE 1 2009 T Perspectives F e d e r a l Rese r v e B a n k o f “ he available evidence … does not lend support to the argument that CRA is to blame for causing the subprime loan crisis.” — Ben Bernanke, Chairman Federal Reserve Board Nov. 25, 2008 The CRA and Subprime Lending: Discerning the Difference D a l l a s ISSUE 1 2009 Banking and Community Perspectives Federal Reserve Bank of Dallas Community Affairs Office P.O. Box 655906 Dallas, TX 75265-5906 Gloria Vasquez Brown Vice President, Public Affairs gloria.v.brown@dal.frb.org Alfreda B. Norman Assistant Vice President and Community Affairs Officer alfreda.norman@dal.frb.org Wenhua Di Economist wenhua.di@dal.frb.org Julie Gunter Senior Community Affairs Advisor julie.gunter@dal.frb.org Jackie Hoyer Houston Branch Senior Community Affairs Advisor jackie.hoyer@dal.frb.org Roy Lopez Community Affairs Specialist roy.lopez@dal.frb.org Elizabeth Sobel Community Affairs Research Associate elizabeth.sobel@dal.frb.org Editor: Kathy Thacker Designers: Gene Autry and Samantha Coplen Researcher and Writer: Elizabeth Sobel A mid the current economic turmoil, fingers are pointing in all directions. What was responsible for the subprime mortgage debacle, which set off a chain of events that led to a financial crisis of global and historic proportions? Many are blaming the Community Reinvestment Act (CRA), alleging that it required banks to make risky mortgage loans to low- and moderate-income people and neighborhoods. The CRA, passed in 1977, followed similar laws enacted to reduce discrimination in credit and housing markets, including the Fair Housing Act of 1968, the Equal Credit Opportunity Act of 1974 and the Home Mortgage Disclosure Act of 1975. The CRA encourages depository institutions to meet the credit needs of the communities in which they operate—including low- and moderate-income neighborhoods—in a manner consistent with safe and sound banking operations. To enforce this statute, four federal regulatory agencies examine banking institutions for CRA compliance. In the interest of separating fact from fiction, the Federal Reserve did its own research and found that the CRA is unequivocally not to blame for the housing market’s March 2009 The views expressed are the author’s and should not be attributed to the Federal Reserve Bank of Dallas or the Federal Reserve System. Articles may be reprinted if the source is credited and a copy is provided to the Community Affairs Office. fall. The numbers just don’t add up. On the Cover Quote from letter by Federal Reserve Chairman Ben Bernanke to U.S. Sen. Robert Menendez, in response to Menendez’s Oct. 24, 2008, letter requesting the Board’s view on the Community Reinvestment Act’s role in the housing market meltdown. Difference,” an overview based on recent data. This publication and our webzine, e-Perspectives, are available on the Dallas Fed website, www.dallasfed.org. 2 To advance the conversation on how to build a stronger, more stable and inclusive financial system, we present “The CRA and Subprime Lending: Discerning the Alfreda B. Norman Assistant Vice President and Community Affairs Officer Federal Reserve Bank of Dallas Banking and Community Perspectives Federal Reserve Bank of Dallas The CRA and Subprime Lending: Discerning the Difference T he Community Reinvestment Act (CRA) has been under much scrutiny amid the subprime lending bust. Critics of the CRA contend that the law pushed banking institutions to undertake high-risk mortgage lending. A Federal Reserve Board staff analysis finds that the CRA was neither a source nor driver of the housing market’s collapse.1 All parties in the housing market—from consumers to mortgage brokers, credit rating agencies, banks, insurance companies and investors—had a hand in the crisis when they miscalculated risk and lost sight of the basic principles of responsibility, accountability and transparency. In total, their decisions created a perfect storm. This perfect storm also exposed the inadequacies of home mortgage regulation and how strictly it was enforced. The late Federal Reserve Board Governor Edward Gramlich cited the “hole in the supervisory safety net,” pointing out that banks and thrifts are subject to federal regulation, their subsidiaries and affiliates are lightly supervised, and independent mortgage companies are not supervised at all on the federal level.2 We examine the CRA and its role in the mortgage market and distinguish it from causes of the subprime failure. CRA: An Overview The Community Reinvestment Act of 1977 was created to combat redlining, a practice in which banks would not offer credit to specific neighborhoods regardless of residents’ creditworthiness. These neighborhoods were redlined largely because of residents’ race, ethnicity, income or a combination of these factors. The CRA requires federally regulated and Federal Reserve Bank of Dallas insured financial institutions to show they are lending and investing throughout their assessment areas, which are defined by the banks as areas in which they accept deposits and make a majority of their loans.3 One of the main principles behind the CRA is that banks and thrifts benefit from the deposits of low- and moderateincome households; in return, they should open access to credit in these communities. By opening access, the CRA enables creditworthy low- and moderate-income individuals to become part of the financial mainstream. Since its passage, the CRA has leveraged an estimated $4.5 trillion in these communities and helped to create jobs, develop small businesses and make mortgages accessible.4 In 1998, CRA-regulated institutions extended $172 billion in small-business and small-farm loans; in 2007, they almost doubled that amount to $342 billion. Their community development lending quadrupled over this period from $16 billion to $64 billion.5 The loan amounts do not include other community development loans, investments and services by public and private institutions that are not required to follow the CRA. In a Federal Reserve survey of CRAcovered financial institutions, most reported that CRA lending was profitable or marginally profitable. And, when banks have tested CRA special lending programs, most have reported low delinquency rates and net charge-off rates.6 How the CRA Works Banks’ lending records are evaluated under the CRA. If a potential borrower applies for a loan for a house, small business, small farm or other purpose, the bank is required to examine the applicant’s creditworthiness and determine if it can extend a loan in a safe and sound manner. The performance context for each bank is different and is a function of the local economy and a bank’s branching structure, business plan, community needs, financial condition and other factors. A bank’s compliance with CRA requirements is evaluated by its regulator, which assigns a rating of substantial noncompliance, needs to improve, satisfactory or outstanding. A bank can build goodwill with the community through a strong CRA rating. But, fundamentally, a bank has an incentive to earn at least a satisfactory rating because falling below that level may result in the denial or delay of applications to open a new branch, merge with another lending institution or expand in other ways. If a regulator has reason to believe that a creditor has engaged in a pattern or practice of discrimination under the Equal Credit Opportunity Act, the regulator is required by the statute to refer the matter to the Department of Justice (DOJ). Housing-related discrimination in violation of the Fair Housing Act that does not involve a pattern or practice, and is not referred to the DOJ, must be referred to the Department of Housing and Urban Development. A regulator does not specify which or how many loans, investments or services a bank has to make under the CRA. It assesses local economic and market conditions that might affect the bank’s income and the geographic distribution of its lending, identifies the number and dollar amount of loans to lowerincome borrowers or areas, and then judges the bank’s performance relative to its context.7 Banking and Community Perspectives 3 By 2006, the subprime market had surpassed $600 billion and accounted for one-fifth of mortgage originations. Independent mortgage companies made 46 percent of these loans. not make bankable loans. They failed to verify borrowers’ income, charged borrowers excessive interest rates and fees, and conducted other poor lending practices. In effect, they set up many borrowers for failure. It’s important to note that subprime loans are not necessarily nonbankable or “predatory” loans. A majority of subprime borrowers are making their payments, building wealth and participating in the American dream.10 Abuses in Subprime Housing Market Making Bankable Loans Rules and regulations require banking institutions, regardless of CRA standing, to make loans that are “bankable”—likely to be repaid according to the terms—and consistent with safe and sound banking practices. Past debt-to-income level, loan payment performance, collateral, net worth and liquidity are all part of a loan applicant’s qualities and determine the likelihood of payment. Race and ethnicity do not factor into creditworthiness. If the potential borrower does not meet all of the criteria, the bank must lower its risk exposure before making the loan—for example, by obtaining credit enhancements.8 The subprime market took off in the late 1990s. By 2006, the market had surpassed $600 billion and accounted for one-fifth of mortgage originations. Independent mortgage companies made 46 percent of these loans; banks and thrifts made 29 percent. Affiliates and subsidiaries of banks and thrifts made the remaining 25 percent.11 While subprime lending existed before the 1990s, the flagrant and widespread abuses in this market did not occur until the late 1990s. The originate-to-distribute model presented the opportunity for independent lending institutions and mortgage brokers to Subprime Mortgage Loans Table 1 Subprime mortgage loans are designed for borrowers who do not qualify for prime mortgages (Table 1).9 Subprime loans are more expensive than prime loans because of the higher risk of default. Relative to borrowers who qualify for prime credit, subprime borrowers have lower FICO credit scores, higher debt-to-income ratios, insufficient cash for down payments or a combination of these risk factors. More than half of subprime mortgages have adjustable rates. Subprime adjustable-rate loans typically have an initial period of two to three years of fixed payments, followed by variable payments (for example, the so-called 2/28 and 3/27 mortgages). Not all borrowers of subprime loans qualified only for this type of loan. Some qualified for prime credit but were steered into subprime loans or chose them. Many mortgage lenders and brokers did Subprime Mortgages At a Glance 4 Banking and Community Perspectives Characteristic Number of loans Average balance (dollars) Average loan age (months) Average FICO make substantial profits. In this model, originators sell, or “distribute,” loans to the secondary market and have less incentive to scrutinize the riskiness of these loans than if they keep them. Their income and fees are based on volume of loans sold, so their focus is on quantity. Before the subprime market fell, securities backed by these loans yielded high returns and were thus appealing to many investors. In this market, serious delinquencies and foreclosures began to edge up nationwide in 2006 and shot up in 2007 and 2008 (Figure 1). There appears to be a direct correlation between the quality of subprime loans and the degree of regulatory oversight. Nondepository mortgage providers such as mortgage lenders and brokers are regulated by 50 different state banking supervisors instead of a federal body responsible for comprehensive oversight. Comptroller of the Currency John Dugan reported that these companies “originated the overwhelming preponderance of toxic subprime mortgages” and these loans “account for a disproportionate percentage of defaults and foreclosures nationwide, with glaring examples in the metropolitan areas hardest hit by the foreclosure crisis.”12 Total Adjustable rate Fixed rate 3,542,728 2,274,513 1,268,215 181,347 199,621 148,573 26 22 33 621 617 628 FICO < 580 (percent) 24.2 25.4 22.0 580 ≤ FICO < 620 (percent) 25.6 26.9 23.3 620 ≤ FICO < 700 (percent) 40.3 39.7 41.4 700 ≥ FICO (percent) 9.9 8.0 13.3 Percent with second lien 22.3 29.9 8.7 Percent with loan to value (LTV) > 90 percent 35.9 43.3 22.6 Percent with prepayment penalty (PPP) 72.6 74.4 69.4 30 26 37 Percent full documentation 66.4 62.4 73.6 Percent potentially prime 19.9 12.1 33.9 Initial interest rate 7.99 8.03 7.92 Current interest rate 8.62 9.01 7.92 Average PPP term (months) NOTES: Figures are as of Dec. 31, 2007. FICO scores, LTV ratios and second-lien percentages are at time of origination. Potentially prime mortgages are loans that at time of origination had less than 80 percent LTV, full documentation and a FICO score of at least 620 and were “owner occupied.” SOURCE: Federal Reserve Board staff calculations from First American LoanPerformance data. Federal Reserve Bank of Dallas Only 6 percent of higher-priced Figure 1 loan originations made by banks Subprime Loan Performance in U.S. Percent and their affiliates in 2005 and 12 2006 went to lower-income Total Serious delinquency Foreclosure started 10 borrowers or borrowers in lower- 8 income neighborhoods within CRA assessment areas. 6 4 2 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 SOURCE: Mortgage Bankers Association/Haver Analytics. Table 2 Share of Higher-Priced Mortgage Originations in 2005–06 (percent) Banking institutions and affiliates Borrower characteristic In CRA assessment area Outside CRA assessment area Independent mortgage companies Total Lower income 6 17 22 45 Non-lower income 7 20 28 55 Total 13 38 50 100 NOTES: Totals may not add up due to rounding. Calculations based on first-lien, conventional, site-built home purchase and refinance originations reported as higher-priced under the Home Mortgage Disclosure Act (HMDA). SOURCE: HMDA data from the Federal Financial Institutions Examination Council. Other factors helped create the perfect storm. Credit scoring systems may have been inadequate to accurately assess credit risk.13 The unusually high rate of housing appreciation gave borrowers extra financial cushioning through increased equity. And the array of financing options was confusing for many borrowers trying to match a loan product to their financial situation. CRA Analysis The Federal Reserve Board researched whether the CRA played a substantial role in the subprime loan crisis. Its staff analysis of 2006 Home Mortgage Disclosure Act (HMDA) data and other sources concludes that the CRA did not contribute to or cause this crisis. Federal Reserve Bank of Dallas According to the analysis: • No major changes have been made to the CRA or its enforcement since 1995. The subprime crisis was triggered by poorly performing mortgage loans orig inated between 2004 and 2007. This chronological gap weakens the contention that the CRA is a major cause of the crisis. • Contrary to the widely held perception that most higher-priced loans were made to lower-income groups targeted by the CRA, 55 percent of higher-priced loan originations went to middle- and upper-income borrowers or borrowers in middle- and upper-income neighborhoods in 2005 and 2006.14 • Only 6 percent of higher-priced loan originations made by banking institutions and their affiliates in 2005 and 2006 went to lower-income borrowers or borrowers in lower-income neighborhoods within CRA assessment areas (Table 2).15 This was calculated by taking the number of higher-priced lowerincome loans made by banks and affiliates in their assessment areas and dividing it by the total number of higherpriced loans made by these institutions. If the proportion were high, it would suggest that banks were trying to originate a large percentage of higher-priced lower-income loans in areas that would earn them CRA credit. The result suggests that banks were not trying to target these areas. • Mortgage purchase data counter the notion that the CRA indirectly created an incentive for independent mortgage companies to make higher-priced lowerincome loans. In 2006, banking institutions bought only about 9 percent of independent mortgage companies’ loans; 15 percent of those loans were higherpriced loans to lower-income borrowers or neighborhoods.16 • The CRA does not appear to have an impact on delinquencies. The Board report compared 90-day-plus delinquency rates of subprime and Alt-A loans in ZIP codes just above and below the CRA eligibility threshold.17 If the rates were different between these types of ZIP codes, the data would suggest that the Banking and Community Perspectives 5 In the years ahead, Table 3 90-Day-Plus Delinquency Rates for ZIP Codes Just Above and Below CRA Threshold (percent) ZIP codes communities can benefit from Subprime Alt-A Total Just above threshold 24.9 15.4 21.0 Just below threshold 24.1 15.6 20.7 NOTES: Data based on mortgages originated between January 2006 and April 2008. Delinquency rates as of August 2008. Table 4 90-Day-Plus Delinquency Rates by ZIP Code (percent) Subprime Alt-A Total Lower income 25.0 16.1 21.5 Middle income 21.3 12.9 17.7 Higher income 19.5 10.9 14.5 NOTES: Data based on mortgages originated between January 2006 and April 2008. Delinquency rates as of August 2008. SOURCE: Delinquency data from First American LoanPerformance. CRA might have an effect on delinquency rates. The rates were almost identical (Table 3). • Delinquency rates were high across all neighborhoods, not just those that were lower income (Table 4). While 90-dayplus delinquency rates of lower-income neighborhoods were the highest, these ZIP codes accounted for a relatively small share of all households—about one-fifth.18 So, the incidence of foreclosure may be quite high in lower-income areas but not be a major contributor to the national foreclosure crisis.19 CRA’s Positive Role In the years ahead, low- and moderateincome households and the growing number of financially fragile households can benefit from the CRA because it helps attract safe and sound lending and spurs competition in their neighborhoods. University of Michigan law professor Michael Barr, who has written extensively on financial services and low- and moderateincome households, said in testimony before the U.S. House of Representatives’ Committee on Financial Services in 2008:20 “In some ways, CRA is well positioned 6 attract safe and sound lending and spurs competition in their neighborhoods. SOURCE: Delinquency data from First American LoanPerformance. ZIP codes the CRA because it helps to help overcome the bifurcation between the prime and subprime markets by enhancing competition from banks and thrifts … [this] would improve market efficiency, reduce racial discrimination, and speed the process of correcting other market failures. Competition … can help to drive out abusive practices and improve price transparency in these markets.” From a broader perspective, the CRA can help stabilize and strengthen the economy. For example, small-business loans reported under the CRA totaled $2.5 trillion from 1998 through 2007. According to the Small Business Administration, firms with fewer than 500 employees accounted for more than half of nonfarm private gross domestic product and 60 to 80 percent of net new jobs annually over the past decade.21 Moreover, data from the Board’s staff report suggest that the CRA prevented the subprime situation from being more severe. As shown in Table 2, only 6 percent of higherpriced loans were made by CRA-regulated lenders to lower-income borrowers or neighborhoods inside their assessment areas, in contrast to 17 percent outside of these areas. A recent analysis of 2006 HMDA data from the country’s 15 largest metropolitan areas compared loans originated by banks in Banking and Community Perspectives their CRA assessment areas with loans made by other lenders in each of these markets.22 Among the findings, these banks were significantly less likely to make high-cost loans—and high-cost loans to low- and moderate-income borrowers—than other lenders. Banks lending in their CRA assessment areas were twice as likely as other lenders to keep the loans they originated. And there was a strong negative correlation between a metropolitan area’s concentration of bank branches and its foreclosure rate: the higher the concentration, the lower the rate. Together, these findings suggest that the CRA helped deter irresponsible lending. Comptroller Dugan concludes that the CRA can continue to play a positive role in the housing market. As the credit market stabilizes, he says, CRA-driven initiatives can help with such challenges as the preservation of homeownership opportunities and rental housing development. Opportunities also lie ahead for bank partnerships with nonprofits to help mitigate the impact of foreclosures in communities across the country.23 The economic crisis unveiled the vulnerabilities of the nation’s financial system. Many laws and regulations—including the CRA—are under review as officials question how effective they are and can be in making the system stronger, more resilient and inclusive. In an effort to foster constructive debate, the Federal Reserve Banks of Boston and San Francisco recently issued a report, Revisiting the CRA: Perspectives on the Future of the Community Reinvestment Act, that examines the CRA’s evolution and highlights possible reforms. The report can be found at www.bos.frb.org/commdev/cra/index.htm. Federal Reserve Bank of Dallas Notes “Did the CRA Cause the Mortgage Market Meltdown?” by Neil Bhutta and Glenn B. Canner, Community Dividend, March 2009, www.minneapolisfed.org/research/pub_display. cfm?id=4136. 2 Subprime Mortgages: America’s Latest Boom and Bust, by Edward M. Gramlich, Washington, D.C.: The Urban Institute Press, 2007. 3 For a comprehensive definition of assessment area, see “A Banker’s Quick Reference Guide to CRA: As Amended Effective September 1, 2005,” Federal Reserve Bank of Dallas, www.dallasfed.org/ca/pubs/quickref.pdf. 4 For more on the CRA, see the National Community Reinvestment Coalition website at www.ncrc.org/index. php?option=com_content&task=view&id=101&Itemid=122. 5 These loan totals are minimums, as not all financial institutions report their data. See nationwide summary statistics, 1998 and 2007, Federal Financial Institutions Examination Council (FFIEC), www.ffiec.gov/reports.htm. 6 “The Performance and Profitability of CRA-Related Lending,” Board of Governors of the Federal Reserve System, report to Congress, July 17, 2000, www.federalreserve.gov/boarddocs/ surveys/craloansurvey/cratext.pdf. For a given loan, the net charge-off is the total dollars owed at default minus any recoveries. An institution’s net charge-off rate is calculated by summing its loan-level net charge-offs over a period of time and dividing this amount by the average outstanding loan balances over the period. 1 A common reference on bank assessment is the FFIEC’s “Uniform Bank Performance Report” at www.ffiec.gov/ubpr.htm. For CRA performance standards by bank size, see source in note 3. 8 Examples of credit enhancements are a Small Business Administration guarantee or a soft second loan from a downpayment assistance program. 9 “A Snapshot of Mortgage Conditions with an Emphasis on Subprime Mortgage Performance,” by Scott Frame, Andreas Lehnert and Ned Prescott, Federal Reserve System, Aug. 27, 2008, http://federalreserveonline.org/pdf/MF_ Knowledge_Snapshot-082708.pdf. 10 “It’s Not Your Parents’ Mortgage Market Anymore,” by Edward Gramlich, Urban Institute, April 6, 2007, www.urban.org/publications/901063.html. 11 The $600 billion data point is from Inside Mortgage Finance. The other data points are HMDA figures from the FFIEC. Calculations are based on first-lien, conventional, site-built home purchase and refinance originations reported as higher-priced under HMDA. 12 Comptroller of the Currency John C. Dugan (Testimony before the Senate Committee on Banking, Housing and Urban Affairs, March 19, 2009), www.occ.treas.gov/ftp/ release/2009-24b.pdf. 13 See “Credit Scoring and Fair Mortgage Lending,” Federal Reserve Bank of San Francisco Community Investments Online, vol. 15, no. 1, 2003, www.frbsf.org/publications/ community/investments/0303. 14 See note 1. Higher-priced loans are defined as those with a price spread exceeding certain thresholds set by the Federal 7 Federal Reserve Bank of Dallas Reserve Board. The price spread is the difference between the annual percentage rate on a loan and the rate on Treasury securities of comparable maturity. For first-lien loans, the threshold is 3 percentage points higher than the Treasury security of comparable maturity; for second-lien loans, which tend to have higher prices, the threshold is 5 percentage points higher. The Board chose the thresholds in the belief that they would exclude most prime-rate loans and include most subprime-rate loans. See note 1. Lower-income borrowers are those whose incomes are below 80 percent of area median family income. Area refers either to metropolitan statistical area (MSA) or to non-MSA counties of the state of origination. 16 See note 1. 17 See note 1. ZIP codes below the threshold have median family income greater than 75 percent and less than 80 percent of area median family income, while ZIP codes above the threshold have median family income between 80 and 85 percent of area median family income. An Alt-A mortgage is considered less risky than a subprime mortgage and riskier than a prime mortgage, so its price usually falls between the two. Loans marketed in Alt-A securities are typically higherbalance loans made to borrowers whose credit problems aren’t severe enough to drop them into subprime territory, or to those who cannot or choose not to document income to qualify for a prime mortgage. See “Nonprime Mortgage Conditions in the United States,” Federal Reserve Bank of New York, www.newyorkfed.org/regional/techappendix_ spreadsheets.html. 18 According to the Federal Reserve Board’s estimations of census and Claritas data, 21 percent of all ZIP codes are lower income, and these ZIP codes account for 19 percent of all households. ZIP code income is based on the ZIP code’s median family income. 19 See note 1. 20 Michael S. Barr (Testimony before the House Committee on Financial Services, hearing on “The Community Reinvestment Act: Thirty Years of Accomplishments, but Challenges Remain,” Feb. 13, 2008), http://financialservices.house.gov/ hearing110/barr021308.pdf. 21 The $2.5 trillion figure is a minimum because it does not cover all lenders. Data are from FFIEC, www.ffiec.gov/reports. htm, and the Small Business Administration, www.sba.gov/ advo/stats/sbfaq.pdf. 22 “The Community Reinvestment Act: A Welcome Anomaly in the Foreclosure Crisis, Indications that the CRA Deterred Irresponsible Lending in the 15 Most Populous U.S. Metropolitan Areas,” Traiger & Hinckley LLP, New York, January 2008, www.traigerlaw.com/publications/traiger_ hinckley_llp_cra_foreclosure_study_1-7-08.pdf. 23 Comptroller of the Currency John C. Dugan (Speech given at Enterprise Annual Network Conference, Nov. 19, 2008), www.occ.treas.gov/ftp/release/2008-136a.pdf. 15 Banking and Community Perspectives 7 PRSRT STD U.S. POSTAGE Federal Reserve Bank of Dallas P.O. Box 655906 Dallas, TX 75265-5906 PAID DALLAS, TEXAS PERMIT NO. 151 The Changing Economy: The New Community Development Lending Environment June 10, 2009 Federal Reserve Bank of Dallas, Houston Branch Amid a rapidly changing economic environment, this conference will help Texas financial institutions and their community development partners identify ways to maintain lending and be responsive to the growing need for community development. Watch for details at www.dallasfed.org. Sponsored by: Federal Reserve Bank of Dallas Texas Mezzanine Fund Office of the Comptroller of the Currency Federal Deposit Insurance Corporation In Partnership with: Office of Thrift Supervision Federal Home Loan Bank of Dallas Texas Association of Community Development Corporations
@FedFRASER