Consumer Compliance Outlook > 2020 > Fourth Issue 2020

Consumer Compliance Outlook: Fourth Issue 2020

HMDA Data Collection and Reporting: Keys to an Effective Program

By Allison Burns, Senior Examiner, Federal Reserve Bank of Minneapolis, and Angelo Parker, Examiner, Federal Reserve Bank of Richmond

The Home Mortgage Disclosure Act (HMDA), as implemented by Regulation C, requires¹ financial institutions subject to the law and regulation (HMDA reporters) to collect and report certain data fields about applications, originations, and purchases of “covered loans.”² It is important that HMDA reporters accurately collect and report HMDA data because of the critical purposes for which the data are used. For example, examiners use HMDA data for fair lending examinations, compliance examinations, and Community Reinvestment Act (CRA) examinations, while public officials use the data for making decisions about distributing public-sector investments.³ Policymakers also review and analyze HMDA data for insights into the mortgage market. Given the importance of accurate data for these purposes, the tolerance threshold for the percentage of errors requiring the reporter to resubmit its data is low, ranging from 2.5 percent to 10 percent, depending on the number of HMDA loans reported.⁴ Errors exceeding the tolerance can result in data resubmission, examination delays, regulatory violations, and civil money penalties.⁵

HMDA data collection and reporting can be challenging. HMDA reporters must collect multiple data fields, some of which have nuanced requirements. For example, prequalifications are not reported, while applications for a home purchase loan in preapproval programs generally are. Similarly, the “action taken” field has multiple options with particular requirements, such as reporting the action taken as “withdrawn” only when the applicant withdraws the application before a credit decision is made to approve or deny the application or before the file is closed because it is incomplete.⁶ It is therefore not surprising that HMDA violations were among the top 10 compliance violations cited by Federal Reserve examiners even before the Dodd‒Frank Wall Street Reform and Consumer Protection Act (Dodd‒Frank Act) and its implementing regulations required HMDA reporters to begin collecting additional data fields beginning in January 2018 and made changes to some of the existing fields.

Technical changes, frequent violations, and the need for accurate data underscore the importance of discussing ways that reporting institutions can strengthen their HMDA data collection, verification, and reporting processes. This article discusses sound practices identified by examiners and suggests compliance management program improvements to help financial institutions strengthen their HMDA collecting and reporting practices. The article’s suggestions should be considered in the context of the size, complexity, and risk profile of the HMDA reporter. We begin by discussing the role of the board of directors and management, provide a brief overview of the law’s requirements, and finish by reviewing sound practices to improve compliance.

Role of Board of Directors and Senior Management

For HMDA reporters, management is responsible for ensuring procedures are in place to collect and maintain accurate data regarding each loan application, loan origination, and loan purchase that must be reported. Further, the board and management also need to ensure the institution provides individuals assigned responsibility for preparing and maintaining the data appropriate training regarding the regulatory requirements, resources, and tools needed to report complete and accurate HMDA data.

Sound practices for successful data collection include creating effective procedures, providing useful tools to staff members responsible for HMDA data collection, and delivering comprehensive training. The following strategies are suggested to help achieve an effective HMDA program.

Successful data collection starts with developing HMDA data collection procedures. Effective HMDA procedures document processes for identifying all HMDA reportable transactions and develop consistent approaches for identifying the underlying source of information for reporting data fields on the HMDA loan application register (LAR) to ensure accurate data collection.

Overview of the Law’s Requirements

Who Must Report

First, the institution must be a “financial institution” as defined by Regulation C. Although HMDA applies to both depository and nondepository financial institutions, this article only addresses the requirements for depository institutions.

• Asset-Size Threshold. On December 31 of the preceding year, the institution meets the requisite asset-size threshold, which is published annually in the Federal Register. For data collection in 2020, the current asset threshold is $47 million.
• Location Test. On December 31 of the preceding year, the institution had a home or branch location located in a metropolitan statistical area (MSA).
• Loan Activity Test. During the preceding calendar year, the institution originated at least one home purchase loan or refinance of a home purchase loan secured by a first lien on a one-to four-unit dwelling.
• Federally Related Test. The institution is (a) federally insured; (b) federally regulated; or (c) originated at least one home purchase or refinance of a home purchase loan that was secured by a first lien on a one-to four-unit dwelling and (i) was insured, guaranteed, or supplemented by a federal agency or (ii) was intended for sale to the Federal National Mortgage Association (Fannie Mae) or the Federal Home Loan Mortgage Corporation (Freddie Mac).
• Loan-Volume Threshold. Effective July 1, 2020, the institution meets or exceeds either the closed-end mortgage loan or the open-end line of credit loan-volume threshold in each of the two preceding calendar years. Currently, an institution that originated at least 100 closed-end mortgage loans in each of the two preceding calendar years,⁷ or originated at least 500 open-end lines of credit in each of the two preceding calendar years meets or exceeds the loan-volume threshold for the respective loan category. The open‑end line credit threshold of 500 is temporary with the permanent 200‑loan threshold effective on January 1, 2022 (i.e., data collection would occur in 2022, with reporting in 2023).⁸

Identifying HMDA Reportable Transactions

Once an institution confirms it is covered by HMDA, the next step in the data collection and reporting process is to identify all HMDA reportable transactions. The Dodd‒Frank Act amendments that became effective in January 2018 changed which transactions are deemed covered by HMDA and Regulation C and therefore reportable. Generally speaking, unless a transaction is expressly excluded under 12 C.F.R. §1003.3(c), an institution subject to HMDA must report all consumer closed‑end mortgage loans and open-end lines of credit secured by a dwelling.⁹ For business or commercial-purpose loans secured by a dwelling, the loan is reportable only if it can be categorized as a refinancing, home improvement, or home purchase loan.¹⁰ The 2018 amendments also modified the definitions of institutional coverage¹¹ and transactional coverage.¹² In addition, the 2018 Economic Growth, Regulatory Relief, and Consumer Protection Act (EGRRCPA)¹³ partially exempted some HMDA reporters from collecting most of the newer HMDA fields added by the Dodd‒Frank Act and implementing regulations.¹⁴

Over or underreporting transactions on the HMDA LAR can lead to noncompliance. Further, HMDA requires the collection of government monitoring information (GMI), which includes the race, ethnicity, and sex of the applicant(s) for the transactions reported on the HMDA LAR. An institution that incorrectly determines whether HMDA applies to a particular transaction could collect GMI when it is not required, or fail to collect it when it is, resulting in examination violations and potential fair lending issues.¹⁵

Transactional Requirements¹⁶

HMDA reporting requirements apply to loans or applications that satisfy the following requirements:

• A consumer open- or closed-end loan secured by a dwelling;
• A business-purpose loan secured by a dwelling that is a refinance, home purchase, or home improvement loan, as those terms are defined in §1003.2(i), (j), and (p), respectively; and
• The application or loan is not on the list of excluded transactions in 12 C.F.R. §1003.3(c), which includes:
  • Loans originated or purchased by the financial institution acting in a fiduciary capacity;
  • Loans secured by a lien on unimproved land;
  • Temporary financing;
  • Purchase of an interest in a pool of otherwise covered loans;
  • Purchase of solely the right to service loans;
  • Purchase of loans as part of a merger or acquisition;
  • Loans in which the total dollar amount is less than $500;
  • Purchase of partial interest in an otherwise covered loan;
  • Loans used primarily for agricultural purposes;
  • Loans made primarily for business or commercial purposes, unless the transaction is also a home improvement loan, home purchase loan, or refinancing; and/or
  • A transaction that proposed to provide or did provide new funds in advance of being consolidated in a New York State consolidation, extension, and modification agreement.¹⁷

Partial Exemption¹⁸

The EGRRCPA partially exempted certain HMDA reporters. In particular, reporters that were insured depository institutions or insured credit unions and originated fewer than 500 closed‑end mortgage loans or open-end lines of credit in each of the two preceding calendar years do not have to collect and report most of the new fields added by the CFPB’s 2015 final rule implementing the Dodd‒Frank Act amendments to HMDA. But even if a HMDA reporter is partially exempt, it must still collect and report the nonexempt data points. We list the exempt fields in Table 1 (The Effect of EGRRCP Act’s Partial Exemption for Certain HMDA Data Points) along with the fields that all HMDA reporters are still required to collect.

Table 1: The EGRRCPA’s Partial Exemption for Certain HMDA Data Points

* The Age field was added by §1094 of the Dodd‒Frank Act, but the EGRRCPA amendment did not include this field in the partial exemption, so even banks qualifying for the partial exemption must still collect and report this new field.

To qualify for the partial exemption, an institution must:

• Be an insured credit union or an insured depository institution¹⁹ that has not been rated either “needs to improve record of meeting community credit needs” during each of its two most recent CRA examinations or not rated “substantial noncompliance in meeting community credit needs”²⁰ on its most recent CRA examination;
• Originate fewer than 500 closed-end mortgage loans in each of the two preceding calendar years (to be partially exempt for its closed-end loans);²¹ or
• Originate fewer than 500 open-end lines of credit in each of the two preceding calendar years (to be partially exempt for its open-end lines of credit).²²

The open- and closed-end exemptions operate independently from each other; an institution may qualify for one partial exemption but not the other.

Further, business lines other than those offering traditional residential mortgages may offer credit extensions that require the institutions to collect and report HMDA data. For example, the commercial loan department may originate purchase-money loans for multifamily buildings such as apartment, cooperative, or condominium buildings. Originating HMDA-reportable transactions in multiple business lines makes identifying and collecting data more challenging, and staff in nonmortgage origination business lines may not be as mindful of HMDA requirements in day-to-day operations. Table 2 (Business Lines and Reportable Transactions) depicts the business lines that often offer loans subject to HMDA as well as the types of HMDA-reportable transactions often found within each business line. As a reminder, this table is illustrative; other types of loans may be HMDA reportable.

Table 2: Business Lines and Reportable Transactions

* The prior version of Regulation C used a “purpose” test (refinance, purchase, or home improvement loan) for all loans to determine if a loan was HMDA reportable. The purpose test now only applies to dwelling-secured loans with a business or commercial purpose.

Institutions have different methods of ensuring that they accurately identify HMDA-reportable transactions. At some institutions, lenders are initially responsible for identifying HMDA-related applications, and the compliance department confirms lenders identified all covered applications by comparing the new loan list with the HMDA LAR. Larger reporters often use automated systems to identify HMDA-reportable transactions. It is also important that financial institutions have a process to track nonoriginated loan applications, such as denied, withdrawn, approved but not accepted, or incomplete applications that have a HMDA purpose. If an institution has a largely manual HMDA process, a centralized review of all nonoriginated loan applications can help ensure the institution reports nonoriginated applications appropriately.

The Bureau has published a chart to help clarify when a loan is secured by a lien on a dwelling, which we list in Table 3 (HMDA Transactional Coverage Chart):²³

Table 3: HMDA Transactional Coverage Chart

†Dwelling means a residential structure, whether or not attached to real property. See §1003.2(f) and comments 2(f)-1 through-5.

Sound Practices

Ensuring Accurate Data Collection

After identifying HMDA reportable transactions, an institution’s next step is to collect accurate data. This step requires attention to detail because of the large number of data fields collected for each application and a solid understanding of HMDA’s requirements, given the complexity of the regulation and certain HMDA transactions. Additionally, depending on the capabilities of the bank’s application and loan systems, data are not always readily accessible to be collected, especially when multiple business lines and staff are involved in providing data. Because of these factors, some banks have found that developing staff with specialized HMDA proficiency improves HMDA compliance.

Deciding how an organization will handle certain HMDA scenarios, such as determining the specific information to report (for which the regulation allows some latitude), eliminates guesswork and ensures consistency across business lines. Some examples of situations in which the bank should determine in advance how it will respond include:

• Using the date of the adverse action notice to determine the action taken date for denied applications (rather than using the date of the decision to deny the application, which could be a different date);
• Using the date the lender receives the application as the application date for originated loans (rather than the date on the application which could be different);
• Determining which credit bureau score to report if more than one score was used in making the credit decision because the regulation requires that only one of the scores used be reported.

Centralizing data collection can be an effective way to reduce reporting errors by reducing the number of people in the data collection process. As part of the centralized process, financial institutions may designate a HMDA subject matter expert (SME) to serve as the central point of contact for data collection and reporting. A well-trained SME can serve as a reliable resource for all individuals involved in HMDA data collection processes.

Additionally, the institution’s procedures should help ensure compliance with Regulation C’s requirement that a covered institution record a transaction on the LAR within calendar days after the end of the calendar quarter, in which it takes final action on the transaction (such as origination or purchase of a loan or denial of an application).

Tools

Providing tools for staff, such as flow charts, worksheets, and industry materials, can also aid in the collection process. Flow charts may include guidance that helps staff decide whether a transaction is HMDA reportable. HMDA worksheets are an effective way for helping staff collect data on all key fields during the loan application process. Worksheets may include references on where to find information in the loan file or reminders about HMDA’s requirements. For example, the worksheet may indicate where to find gross income in the file, depending on the loan type, and could include a reference of when income should be reported as not applicable. Cheat sheets may remind staff how to geocode the collateral securing the loan. Finally, providing copies of industry guidance, such as the HMDA Getting it Right! booklet or the HMDA Small Entity Compliance Guide, also helps staff understand HMDA data collection requirements, especially when they encounter unfamiliar or complex transactions.

Many banks find that using an automated collection process reduces the burden of compiling HMDA data. Automated collection offers a consistent process, using the information entered during loan origination as source documentation for HMDA data. The level of automation may vary from bank to bank, usually depending on factors such as origination volume and institutional complexity. Some financial institutions use their loan processing system to determine geocodes. Other institutions use data collection software to compile the entire LAR. Examples of automated processes for the new HMDA data fields include calculating the applicant’s age based on birthdate as opposed to staff manually entering the date or using software that automates the process of extracting HMDA data from the lender’s origination software to ensure the information is in the correct format for the HMDA LAR prior to submitting it.

Training

Regular in-depth training is an effective tool to ensure staff understand HMDA data collection requirements. Whether using a centralized or decentralized process, or a hybrid of both, it is important that all staff members involved in the process understand reporting requirements and that the bank applies collection procedures consistently. Effective training reflects each individual’s role in the collection process and provides sufficient detail to aid staff in identifying the transactions to be reported and the data to collect. Effective training also helps staff understand regulatory requirements and internal HMDA procedures. Regular training helps staff stay up to date on the rules and helps create consistency among business lines and staff involved in the HMDA process.

Training is particularly beneficial for some of HMDA’s more complicated requirements in which data reporting errors are more common. Some training topics that could be addressed are:

• How to properly report denials, withdrawals, and multiple use loans;
• The nuances in reporting data fields that depend on specific calculations, such as borrower age, borrower credit score, and origination fees/closing costs; and
• The interface between the core system and the automated collection software.

Data Verification

Before submitting its HMDA data, an institution can perform a comprehensive review to verify the accuracy of the data collected compared with the source documentation within the loan files to identify and correct any errors and increase the accuracy of the reported information. Depending on the volume of data collected, this process may involve testing through sampling. An effective verification process gives the financial institution an opportunity to measure the accuracy of its collection and reporting processes and identify weaknesses that may exist. The verification should also test the effectiveness of processes used to identify all applicable HMDA loans and nonoriginated applications.

The data review can be conducted internally or by a reputable third-party vendor. The strength of the institution’s data collection processes should determine the scope and frequency of the review. The risk of HMDA noncompliance may be greater for institutions with a high origination volume or a decentralized collection process. Reviews may uncover errors that can range from simple typographical errors to more significant procedural errors that could lead to systemic reporting violations, data scrubs, and resubmission. If the review identifies errors, the institution should correct the data prior to submission. When weaknesses are noted, the severity of the weaknesses should be assessed and appropriate corrective actions taken to address the root cause. A thorough data verification process provides a much-needed last line of defense for HMDA reporters.

Data Reporting

In addition to data collection, institutions can also develop procedures and training for individuals responsible for reporting collected HMDA data. An institution reporting fewer than 60,000 covered loans and applications in the preceding calendar year must submit its prior year’s LAR to its primary federal regulator by March 1.²⁴ Institutions reporting more than 60,000 covered loans and application in the preceding calendar year must submit the data within 60 calendar days after the end of each quarter, except the fourth one.²⁵ But note, as discussed in endnote 27, the Bureau and the Federal Reserve have temporarily relaxed enforcement of the quarterly reporting requirement. A separate and complete LAR must be transmitted for each covered institution. For example, one LAR must be submitted for a bank and a separate LAR for a subsidiary of the bank. A number of tools are available to ensure the LAR meets submission standards.

The LAR must be submitted electronically as a text file using the HMDA Filing Platform (the platform) from the Federal Financial Institutions Examination Council (FFIEC).²⁶ The platform will automatically check the file for syntactical, validity, quality, and macro edits. If there are any errors, the system will notify the institution immediately. Any errors must be corrected at the source level, and the entire LAR must be uploaded again. Once the completed LAR has been uploaded, an approved representative must certify it and mark it as complete. Questions around the filing process can be answered at the FFIEC’s website at https://ffiec.cfpb.gov/, which provides a number of tools to assist institutions, including the Filing Instruction Guide, the Supplemental Guide for Quarterly Filers, and the Self-Service Knowledge Portal, which answers frequently asked questions on HMDA.

Institutions that report a small volume of covered loans can use the LAR Formatting Tool to help create an electronic file for submission.²⁷

Navigating the nuances of HMDA collection and reporting can be challenging, especially because of the changes that became effective in 2018. Table 4 (Sound HMDA Practices) lists the processes we have observed at institutions with effective HMDA data collection and reporting processes.

Table 4: Sound HMDA Practices

While this list is not exhaustive, most institutions can implement these practices, regardless of the size and structure of the HMDA program. It is important to determine the institution’s risk profile, assess the level of knowledge within the institution, commit the necessary resources to the process, and apply the practices best suited for the level of risk and resources.

Conclusion

Implementing sound practices can help improve the HMDA data collection and reporting process. Whether the process is centralized or decentralized, establishing and consistently applying collection, verification, and reporting processes will give bank staff a solid foundation for ensuring complete and accurate data collection and reporting. Combined with adequate training, effective job aids and timely HMDA data reviews, the institution can leverage these sound practices and develop a HMDA process that will strengthen its compliance management program. Specific issues and questions related to Regulation C should be raised with the institution’s primary regulator.

ENDNOTES

¹ See 12 C.F.R. §1003.4(a).

² See 12 C.F.R. §1003.2(e).

³ See 12 C.F.R. §1003.1(b); 12 U.S.C. §2801(b).

⁴ See Federal Reserve CA letter 17-2 discussing revised FFIEC HMDA Examiner Transaction Testing Guidelines and thresholds for when data must be resubmitted.

⁵ See 12 C.F.R. §1003.6(a).

⁶ See Comment 4(a)(8)(i)-5.

⁷ See 12 C.F.R. §1003.3(c)(11). The change to the threshold to 100 HMDA loans was announced in April 2020 and made effective July 1, 2020. 85 Federal Register 28364 (May 12, 2020).

⁸ See 12 C.F.R. §1003.3(c)(12).

⁹ See 12 C.F.R. §1003.2(e). Before 2018, the regulation limited HMDA reporting to purchase, refinance, and home improvement loans secured by a dwelling. Effective January 1, 2018, the regulation expanded the coverage. Because consumer dwelling-secured loans can now have other purposes and still be covered, the purpose field was amended to add a new option “or for a purpose other than home purchase, home improvement, refinancing, or cash-out refinancing.” 12 C.F.R. §1003.4(a)(3).

¹⁰ See 12 C.F.R. §1003.3(10).

¹¹ See 12 C.F.R. §1003.2(g) and the related commentary. See also A Guide to HMDA Reporting: Getting it Right! p. 1.

¹² See 12 C.F.R. §1003.2(e) and the related commentary. See also A Guide to HMDA Reporting: Getting it Right! p. 10.

¹³ See Public Law 115–174, 132 Stat. 1296 (2018) at Section 104. The open- and closed-end exemptions are independent of each other. For example, an institution reporting more than 500 closed-end HMDA loans and 100 open-end loans would be required to report the closed-end loans but would be exempt from reporting the open-end loans.

¹⁴ See A Guide to HMDA Reporting: Getting It Right! (2020), p. 29.

¹⁵ Outlook reviewed the interplay between HMDA and the Equal Credit Opportunity Act (ECOA) in the Fourth October 2013 issue: “Government Monitoring Information Requirements Under the HMDA and the ECOA.”

¹⁶ See A Guide to HMDA Reporting: Getting It Right! (2020), p. 29

¹⁷ For additional information on this exclusion, see comment 3(c)(13)-1.

¹⁸ See A Guide to HMDA Reporting: Getting It Right! (2020), p. 29.

¹⁹ See 12 C.F.R. §1003.3(d)(1)(iv)(2), (3).

²⁰ See 12 C.F.R. §1003.3(d)(6).

²¹ See 12 C.F.R. §1003.3(d)(3) and Comment 3(d)(3)-1.

²² See 12 C.F.R. §1003.3(d)(3) and Comment 3(d)(3)-1.

²³ See CFPB’s HMDA transactional coverage.

²⁴ See 12 C.F.R. §1003.5(a)(1(i).

²⁵ See 12 C.F.R. §1003.5(a)(1)(ii). Note, however, that in March 2020, the Consumer Financial Protection Bureau announced that in response to the COVID-19 pandemic, “until further notice, the Bureau does not intend to cite in an examination or initiate an enforcement action against any institution for failure to report its HMDA data quarterly” for institutions reporting 60,000 or more covered loans.” The Federal Reserve Board made a similar announcement. See CA letter 20-6.

²⁶ The HMDA Filing Platform is available on the FFIEC website.

²⁷ The LAR Formatting Tool is available on the FFIEC website.