From the Examiner's Desk... Two Years After: Assessing the Impact of the New HMDA Reporting Requirements
This regular feature focuses on developments that affect the bank examination
function. We welcome ideas for future columns, and readers can e-mail suggestions
Two years have passed since the Federal Reserve Board's (FRB) latest revisions to Regulation C,1 the implementing regulation for the Home Mortgage Disclosure Act (HMDA),2 became effective. Among the changes made by the FRB, the most notable required reporting on loan pricing data. This article discusses the impact of the latest changes to Regulation C on fair lending examinations and on the HMDA examination process. The article also provides information on some of the most common HMDA violations identified since the implementation of the new requirements.
With the advent of these changes, the FDIC now has additional information to consider in the scoping and focusing of fair lending examinations. According to a September 13, 2005, Federal Financial Institutions Examination Council (FFIEC) press release, 8,853 institutions reported HMDA data for 2004. Depository and for-profit nondepository institutions must report HMDA data depending on their asset size, extent of business in a metropolitan statistical area, and whether or not they offer residential mortgage lending. For 2004, the asset threshold3 for depository institutions was $33 million. The FDIC oversees approximately 2,800 HMDA reporters, or nearly 32 percent of all HMDA reporters.
HMDA was enacted in 1975 to provide the public with loan data that would assist in determining whether institutions were serving the housing needs of their communities. It also provided an enhanced tool for examiners and others to use in identifying discriminatory lending practices, and assisted public officials in distributing public-sector investments.4 Even with the new data, however, it is important to note that analysis of the data alone cannot identify discriminatory lending practices. The HMDA data, while providing some good red flag indicators, does not include information on the creditworthiness of borrowers or other criteria (such as loan-to-value ratios or credit scores) that a bank may use in pricing loans. The new collection and reporting requirements, however, do provide an improved starting point for identifying potential discriminatory practices.
Fair Lending and the New Data
The new requirements garnering the most attention and having the most effect on the fair lending examination process are the reporting of the following:
Home Ownership and Equity Protection Act (HOEPA)5 applicability
Modified racial/ethnic categories
Rate spreads have been added to the reporting requirements to help identify loan pricing practices that may warrant further investigation. Rate spreads are reported if the spread between the loan's annual percentage rate and the Treasury yield equals or exceeds 3 percentage points for first-lien loans or 5 percentage points for subordinate-lien loans.6 The data are reported for all originations of home purchase, dwelling-secured home improvement, and refinance loans.
HMDA reporters must also report the HOEPA status of loans. HOEPA loans have unusually high interest rates or fees. Identifying these loans helps examiners detect abusive practices that have accompanied some of these loans in the past.
The racial/ethnic categories have been revised to reflect recent changes to the Office of Management and Budget (OMB) racial and ethnic standards for federal statistics and administrative reporting, and to conform to Census Bureau definitions. Instead of five mutually exclusive categories that combine race and ethnicity, applicants now designate their ethnicity ("Hispanic" or "Not of Hispanic Origin") separately from race. Applicants may also indicate more than one racial category. Additionally, lenders must now ask applicants their ethnicity, race, and sex in applications received by telephone, mail, or over the Internet.7 These changes allow examiners to identify and compare applicants on the basis of race and ethnicity more accurately.
Table 1 provides summary information on changes to HMDA data collection and reporting requirements.
HMDA Data Changes, Effective 1/1/2004
Description of Change
Requires lenders to identify applications and loans that involve manufactured housing. It is anticipated that reporting these loans separately will help explain differences in denial rates and pricing.
Redefined the definitions of refinancing and home improvement loans to provide more consistency and reliability of data.
Preapproval requests (Preapprovals should be distinguished from prequalifications, which are not reported for HMDA purposes)
The revisions require lenders to report information on requests for preapprovals of home purchase loans. Data on denied preapprovals will provide more complete information on the availability of home financing.
Lenders now must report the lien status of applications and originated loans. These data will be used to help interpret rate spread data and to differentiate between secured and unsecured home improvement loans.
Type of purchaser
For loans originated or purchased and then sold within the same year, the type of entity that purchased the loan must be reported. The types of purchasers have been expanded to increase the usability of the data and provide information about the secondary market.
Coverage rule (Nondepository lenders)
Nondepository lenders must report if they originated home purchase loans, including refinances, equaling at least $25 million in the preceding calendar year.
Application information New ethnicity categories, "Hispanic" and "Not of Hispanic Origin," were created, and the race categories were revised to reflect changes to OMB standards. Lenders also must now ask for ethnicity, race, and sex in applications taken by telephone, mail, or Internet.
Additional data items
For loan originations, lenders must now report the rate spread between the annual percentage rate and the yield on comparable Treasury securities, if the spread exceeds or equals 3 percentage points for first-lien loans or 5 percentage points for subordinate-lien loans. Lenders must also report whether a loan is subject to the Home Ownership and Equity Protection Act.
HMDA sampling and resubmission procedures
New HMDA fields to the list of key fields include the following:
Request for preapproval
Ethnicity, race, and sex of the applicant and co-applicant
Type of purchaser
These fields are considered critical to the integrity of analyses of the overall HMDA data.
Fair Lending Examination Procedures
Examiners consider pricing systems and discretionary pricing practices in conjunction with the new pricing data as a part of the scoping process whenever they examine any HMDA reporter. When significant disparities8 are found in a system that permits pricing discretion, a comparative loan file analysis is conducted to determine the reason for the pricing differences.
FDIC's headquarters staff9 reviews the data annually for all HMDA reporting institutions to identify institutions that appear to have particularly strong indicators of possible discrimination in the pricing of one or more loan products. These institutions undergo increased scrutiny and may receive an accelerated fair lending examination, including a comparative file analysis.
In conducting a comparative file analysis, examiners consider both race and ethnicity. Selecting a target group (the group suspected of receiving less favorable treatment) that will be the focal point of a fair lending review requires examiner discretion. Selection of both a target group and an appropriate control group (the group suspected to be receiving more favorable treatment) may incorporate both race and ethnicity. For example, a common control group would be non-Hispanic (ethnicity) whites (race). The addition of ethnicity information and the ability of applicants to select more than one race allow a more precisely targeted analysis.
The addition of rate spread and HOEPA information to the HMDA data provides examiners additional tools to scope and focus fair lending examinations. Examiners use the data to compare different lenders in the market and to more readily identify secondary market loans (see purchaser type code changes in Table 1).
In all comparisons, examiners look for differences in how certain target groups are treated compared to an appropriate control group. Appropriate racial/
ethnic/sex comparisons are made within each combination of loan type, property type, loan purpose, and lien type. For example, the percentage of loans to non-Hispanic whites for which rate spreads are reported are compared to the percentage of loans to Hispanics for which rate spreads are reported. The average spread for target and control group loans is also analyzed.
Examiners also use the race/ethnicity or sex HMDA data elements in conjunction with the pricing information to determine the need for a steering analysis. A steering analysis determines whether lending personnel guide, or "steer," applicants from a market-rate product for which the applicants may qualify to a less favorable alternative (e.g., a more expensive subprime mortgage product). While guiding an applicant to a loan product that meets that applicant's individual qualifications is not illegal, it can result in fair lending violations when the reason for the referral is not related to the applicant's creditworthiness, but rather to one of the prohibited bases.
For example, a bank may originate loans subject to HMDA reporting through both a mortgage division and a consumer loan division. Loans originated through a bank's internal consumer loan division are typically priced higher, in either rates or fees, than loans sold on the secondary market through its mortgage division. In such situations, examiners consider whether target group applicants are discriminatorily steered to the consumer loan division. In most institutions, part of the loan number on the HMDA-Loan Application Register (HMDA-LAR) will indicate which division originated the loan. The HMDA data field "Type of Purchaser" can also help distinguish between in-house loans, which are often originated out of a consumer loan division, and secondary market loans, which are sold to investors. In addition to the data analysis, customer interviews may be required to substantiate whether steering is occurring. A decision on whether to conduct customer interviews is made only after consultation with senior headquarters staff.
Throughout the fair lending examination process, examiners consult with Regional fair lending examination specialists and, in many cases, headquarters fair lending staff, to ensure that financial institutions receive consistent treatment on both a Regional and a national basis. If a pattern or practice of discrimination is identified, the violation is referred to the Department of Justice (DOJ). The referral provisions of the Equal Credit Opportunity Act (ECOA)10 require that the federal financial institution regulatory agencies make a referral to the DOJ "whenever the agency has reason to believe that one or more creditors has engaged in a pattern or practice of discouraging or denying applications for credit" in violation of ECOA's general rule prohibiting discrimination. At the FDIC, referral to DOJ is initiated through formal consultations with the Regional office and headquarters in Washington.
HMDA Examination Procedures
Interagency Examination Procedures11 were revised to address the new HMDA data requirements. Under the new procedures, and consistent with the FDIC's compliance examination focus,12 examiners now concentrate their review of HMDA compliance on determining the effectiveness of an institution's compliance management system with regard to HMDA data collection and reporting requirements. The data are tested as necessary to determine whether the compliance management system is adequate.
Accurate HMDA reporting is critical. HMDA data are made available to the public to help determine whether institutions are serving the housing needs of their communities and to identify potential discriminatory lending patterns. The new HMDA pricing information has been of significant interest to many public and private groups, including consumer groups, community groups, federal regulators, and congressional committees. In addition, the financial institution regulatory agencies use the data in conjunction with Community Reinvestment Act performance evaluations, as well as fair lending examinations. Inaccurate collection and reporting of HMDA data resulting in significant violations could subject an institution to civil money penalties. The FFIEC interagency examination guidance states that every bank, regardless of asset size, should have comprehensive audit and review procedures to verify the accuracy of its HMDA data.
Through management interviews and reviews of a bank's written policies, internal controls, and HMDA-LAR, examiners determine whether the bank has adopted and implemented comprehensive procedures to ensure adequate compilation of home mortgage disclosure information in accordance with Section 203.4(a-e). Examiners also interview the bank's frontline HMDA personnel and review training records to determine the effectiveness of a bank's policies and training program.
Examiners determine whether the bank has a system for tracking rate lock dates and rate spreads.13 Examiners also review for written procedures relating to the collection of ethnicity, race, and sex data for all applications received by telephone, mail, or Internet.
The latest changes to the HMDA data collection and reporting requirements provide examiners more readily available data for initial analysis, which should improve the efficiency and quality of the fair lending scoping process. Examiners' ability to identify loan pricing concerns that warrant further investigation should be substantially enhanced. Preliminary questions that examiners pose most often include the following:
To what extent are there disparate rates of higher-priced loans in minority communities, and why?
What pricing disparities exist among borrowers of different races, ethnicities, or genders, and why?
Do the disparities reflect important new homeownership opportunities for some borrowers that would not otherwise exist or unfair treatment?
To what extent do disparities exist among insured financial institutions, affiliated mortgage companies, or independent mortgage companies that focus on the subprime market?
Information on current HMDA violations indicates the continuing need for bank management to provide appropriate oversight of their banks' HMDA reporting systems to ensure accurate reporting. Institutions that have been successful with their HMDA programs provide effective training, a strong internal monitoring system, and audit procedures that identify and address the underlying causes of violations.
A limited review of HMDA examinations since the reporting of the new data revealed that errors in collecting and reporting data elements often resulted in violations of law. Deficiencies noted were similar to those addressed in the December 2005 FFIEC memorandum, with the most frequently cited violations pertaining to the HOEPA status and the rate spread information. For example, some banks incorrectly reported rate spread information for loans that were not subject to Regulation Z. Others inaccurately reported loans as being subject to HOEPA, had erroneous information pertaining to preapproval requests, or failed to collect the ethnicity of applicants.
While violations may reflect errors rather than willful violation of requirements, repeat violations of the same or similar nature in subsequent examinations can result in the assessment of civil money penalties. Further review indicated that the HMDA violations often stemmed from weaknesses in the banks' compliance management systems, including inadequate training, insufficient monitoring, and lack of appropriate audit procedures. Addressing these weaknesses can minimize the potential for future violations.
Julie V. Banfield
Field Review Examiner,
New York, NY
Elizabeth C. Borio
Field Review Examiner,
1 On January 23, 2002, the Federal Reserve Board (Board) adopted a final rule amending Regulation C, effective January 1, 2003. See 67 FR 7222 , February 15, 2002. The Board subsequently delayed the effective date of the amendments until January 1, 2004. See 67 FR 30771 , May 8, 2002.
3Section 203.2 (e)(1)(i) of Regulation C provides that the Federal Reserve Board will adjust the exemption threshold for depository institutions annually based on the year-to-year change in the average of the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPIW), not seasonally adjusted, for each 12-month period ending in November, rounded to the nearest million.
5HOEPA , contained in the Riegle Community Development and Regulatory Improvement Act of 1994, Pub. L. 103-325, was enacted in response to anecdotal evidence of abusive lending practices in the home-equity lending market. HOEPA imposes certain disclosure requirements, as well as some substantive limitations, on certain home-equity loans with rates and fees above a certain percentage or amount. HOEPA is implemented through the Federal Reserve Boards' Regulation Z, including 12 C.F.R. §226.31 , §226.32 , and §226.34.
7 Lenders were required to ask applicants their race, national origin, and sex in applications taken entirely by telephone effective January 1, 2003. The revised ethnicity and race categories did not take effect until January 1, 2004. See 67 FR 43217 and 67 FR 43218 .
8 A determination of significant disparities typically involves a statistical analysis conducted by Regional and headquarters fair lending specialists and statisticians.
9 The FDIC's Division of Supervision and Consumer Protection works closely with statisticians and economists in the Division of Insurance and Research to develop screening techniques to identify institutions that exhibit an unusually high risk of pricing discrimination against one or more racial/ethnic minority groups or women.