Mortgages

Subscribe to Mortgages

The CFPB’s Winter 2019 Supervisory Highlights discusses the Bureau’s examination findings in the areas of automobile loan servicing, deposits, mortgage loan servicing, and remittances.  We discussed the Bureau’s auto loan servicing findings in a separate blog post.  In this blog post, we focus on the Bureau’s additional findings.

Although issued under Director Kraninger’s leadership, the Winter 2019 Supervisory Highlights covers examinations generally completed between June and November 2018 when Mick Mulvaney was Acting Director.  It represents the CFPB’s second Supervisory Highlights covering supervisory activities conducted under Mr. Mulvaney’s leadership.  Like the Summer 2019 Supervisory Highlights, the Winter 2019 issue contains the following language in its introduction:

It is important to keep in mind that institutions are subject only to the requirements of relevant laws and regulations.  The information contained in Supervisory Highlights is disseminated to help institutions and better understand how the Bureau examines institutions for compliance with those requirements.  In addition, the legal violations described in this and previous issues of Supervisory Highlights are based on the particular facts and circumstances reviewed by the Bureau as part of its examinations.  A conclusion that a legal violation exists on the facts and circumstances described here may not lead to such a finding under different circumstances.

Also like the Summer 2019 Supervisory Highlights, the new issue’s introduction and the Bureau’s press release about the report does not include any statements touting the amount of restitution payments that resulted from supervisory resolutions or the amounts of consumer remediation or civil money penalties resulting from public enforcement actions connected to recent supervisory activities.  (The report does, however, include summaries of the terms of five consent orders entered into by the Bureau, including its settlements with Cash Tyme, a payday retail lender, and Cash Express, a small-dollar lender.)

Key findings include:

Deposits.  CFPB examiners found that one or more institutions engaged in deceptive acts or practices by representing that payments made through their online bill-pay service would be debited no sooner than the date selected by the consumer and failing to disclose (or disclose adequately) that the debit might occur earlier than that date when the payee only accepted a paper check.  Such paper checks were sent by the institution several days before the consumer’s designated payment date, at the institution’s discretion, and would be debited from the consumer’s account when presented and cashed by the payee.  As a result, the debit could have occurred earlier or later than the designated date.  In response to the Bureau’s findings, the institutions “undertook a revision” of their consumer-facing materials and “undertook a plan to remediate consumers” who were charged an overdraft fee due to a paper check being negotiated before the consumer’s designated payment date.

Mortgage Servicing.  CFPB examiners found:

  • Servicers engaged in unfair practices by charging late fees greater than those permitted by the mortgage notes.  In one example, the FHA mortgage note permitted late fees based on a percentage of the overdue principal and interest only.  However, the servicer charged late fees based on a percentage of the full periodic payment of principal, interest, taxes, and insurance.  The overcharges were the result of programming errors in the servicing platform and “lapses in service provider oversight.”  In response to the findings, servicers conducted a review to identify and remediate affected borrowers and changed their policies and procedures “to assist in charging the late fee authorized by the mortgage note.”
  • Servicers engaged in deceptive practices by misrepresenting the conditions for the cancellation of private mortgage insurance (PMI).  One or more servicers were found to have told borrowers requesting PMI cancellation that such requests were declined because the borrowers has not reached the 80% loan to value requirement for cancellation.  While the relevant amortization schedules did not yet reach 80% LTV, the borrowers had in fact reached 80% LTV by making additional principal curtailment payments.  Although the borrowers had not satisfied additional conditions necessary for PMI cancellation, the servicers did not identify those conditions as the reasons for denying the borrowers’ cancellation requests.  In response to the Bureau’s findings, the servicers “changed templates, as well as policies and procedures, to ensure that PMI cancellation notices state accurate denial reasons.”
  • One or more servicers were found not to have satisfied the Regulation X requirement for a servicer to exercise “reasonable diligence” in obtaining documents and information to complete a loss mitigation application.  The servicers offered short-term payment forbearance programs during collection calls to delinquent borrowers who had expressed interest in loss mitigation and submitted financial information that the servicer would consider in evaluating them for loss mitigation.  However, the servicers had not notified the borrowers that their forbearance offers were based on an incomplete application evaluation and did not contact the borrowers, near the end of the forbearance period, to inquire whether they wanted to complete the applications to receive a full loss mitigation evaluation.  In response to the Bureau’s findings, the servicers “used enhanced processes, such as a centralized queue, to track borrowers receiving short-term forbearance programs and subsequently notify them that additional loss mitigation options may be available and that they could apply for such options over the phone or in writing.”
  • In examinations reviewing servicing of Home Equity Conversion Mortgage loans, examiners criticized the notices sent by servicers to successors of deceased borrowers informing them that they could qualify for an extension of time to delay or avoid foreclosure to enable them to purchase or market the property and directing them to return a form indicating their intentions for the property.  While the notices listed several documents that might be needed to evaluate whether the successor qualified for an extension, it did not direct them to submit such documents within a certain timeframe to be eligible for an extension. The Bureau found that the servicers had assessed foreclosure fees, and in some instances had foreclosed on properties, where successors had returned the form indicating that they intended to purchase or market the property but had not returned the documents necessary for an evaluation. While examiners did not find this to be a legal violation, they observed that it could pose a risk of a deceptive practice by creating an impression that the statement of intent was all that was needed to delay foreclosure.  In response to the Bureau’s findings, the servicers “planned to improve communications with successors, including specifying the documents successors needed for an extension and the relevant deadlines.”

Remittances.  CFPB examiners found that one or more supervised entities violated the remittance rule’s error resolution requirements by failing to refund fees and, as allowed by law, taxes to consumers whose remitted funds were made available to designated recipients later than the date of availability stated in the entity’s remittance disclosures and the delay was not due to any of the rule’s excepted events. The CFPB cited the violations, even though the delays at issue were due to a mistake by a non-agent foreign payer institution.  The CFPB reminded companies that “neither the relationship between a remittance transfer provider and the institution disbursing the funds to the designated recipient, nor the particular entity that is at fault for the delayed receipt of funds, is relevant to whether the remittance transfer provider must refund fees and taxes to the consumer.”  In response to the Bureau’s findings, the entities are making the mandated refunds.

The CFPB has issued a report, “Mortgages to First-time Homebuying Servicemembers,” that focuses on mortgage loans made from 2006 to 2016 to first-time homebuyers who are serving in the military or are veterans.  In its press release, the Bureau states that the report represents “the first time researchers have been able to provide a description and analysis of servicemembers’ mortgage choices and mortgage performance, both during and after the housing crisis of the last decade.”

The report is part of the Bureau’s series of quarterly reports on consumer credit trends that uses a “longitudinal, nationally-representative sample of approximately five million de-identified credit reports from one of the three nationwide consumer reporting agencies.”  To distinguish credit reports for servicemembers from those of non-servicemembers, the consumer reporting agency providing the data matched credit records in the sample to the Department of Defense’s Servicemembers Civil Relief Act database.

The report’s key findings include:

  • The share of first-time homebuying servicemembers using loans partially guaranteed by the U.S. Department of Veterans Affairs (VA Loans) increased from 30 percent before 2007 to 63 percent in 2009.  Among non-servicemember first-time homebuyers there was a parallel increase in the use of FHA and Department of Agriculture mortgage loans.  However, in contrast to non-servicemembers whose reliance on FHA/USDA mortgages declined after 2009, servicemembers continued to increased their reliance on VA loans, with VA Loans comprising 78 percent of servicemember loans in 016.
  • The greater share of VA Loans among servicemembers was part of a larger shift among consumers (both servicemembers and non-servicemembers) away from conventional to government-guaranteed mortgage loans between 2006 and 2009.  Conventional (i.e. non-government-guaranteed) mortgage loans represented about 60 percent of loans among first-time homebuying servicemembers in 2006 and 2007, but declined to 13 percent by 2016.  In comparison, the share of conventional loans among non-servicemembers fell from almost 90 percent before 2008 to 41 percent in 2009, then increased back to 60 percent in 2016.  The combined share of FHA and USDA mortgages to these borrowers increased and then decreased accordingly.
  • The median VA Loan amount for first-time homebuying servicemembers increased from $156,000 in 2006 to $212,000 in 2016, closely tracking the median value of conventional home loans taken out by non-servicemembers.  In contrast, the median loan amounts for servicemembers who used conventional or FHA/USDA mortgages during this period were lower and increased more slowly.
  • Delinquency rates for nonprime servicemember borrowers with VA Loans peaked in 2007 at approximately 7 percent and fell to about 3 percent by 2016.  Delinquency rates between 2006 and 2008 among nonprime borrowers were significantly higher for FHA/USDA loans to non-servicemembers and for conventional loans to both servicemembers and non-servicemembers.  Conventional loans to nonprime borrowers, for both servicemembers and non-servicemembers, had delinquency rates greater than 13 percent in 2006 but were below two percent by 2016.  After 2009, conventional loans to all nonprime borrowers generally had lower delinquency rates than both VA Loan to nonprime servicemembers and FHA/USDA loans to nonprime non-servicemembers.

 

The CFPB has published its 2019 final lists of Rural and Rural or Underserved Counties on its website. The CFPB has previously posted lists of such counties for calendar years 2011-2018. The CFPB has also updated the rural and underserved areas website tool for 2019.

The lists and tool are relevant to exemptions from certain regulatory requirements of the Truth in Lending Act, including the following CFPB mortgage rules:

  • Escrows Rule, which requires a creditor to establish an escrow account for certain first-lien higher-priced mortgage loans (HPMLs), but exempts HPMLs consummated during a calendar year (or next-to-last calendar year for loans where the application was received before April 1 of the current calendar year) if the creditor extended a first-lien covered transaction in the preceding calendar year secured by a property located in a rural-or-underserved area, and meets certain additional conditions.
  • Ability to Repay and Qualified Mortgage Standards Rule, which treats certain balloon-payment mortgages as qualified mortgages if they are originated and held in portfolio by small creditors that meet the rural-or-underserved test above and certain additional conditions
  • Home Ownership and Equity Protections Act of 1994 (HOEPA) rule, which generally bans balloon payments for mortgages that fall within HOEPA’s high-cost mortgage coverage test, unless they meet the rural-or-underserved test and certain additional conditions
  • Appraisals for HPMLs rule, which exempts HPMLs made in “rural” counties from its additional appraisal requirement

Note that if a creditor makes a first-lien mortgage loan secured by a property located in a rural or underserved area during 2019, the creditor will satisfy the rural and underserved test for the rules noted above during all of 2020 and for loan applications received before April 1, 2021.

Last month, after more than three years of urging by the industry to provide written guidance, the CFPB issued four FAQs on its TILA/RESPA Integrated Disclosure (TRID) Rule.  In this week’s podcast, we take a close look at the FAQs and what they tell creditors, particularly its guidance on when a corrected Closing Disclosure and new three-day waiting period are required (with a caution for those selling to investors) and the safe harbor for using a model form.

Click here to listen to the podcast.

Federal financial institution regulators recently issued a joint final rule to implement provisions of the Biggert-Waters Flood Insurance Reform Act of 2012 (the “Act”) that require regulated financial institutions to accept private flood insurance policies. The regulators are the Farm Credit Administration, Federal Deposit Insurance Corporation, Federal Reserve Board, National Credit Union Administration, and Comptroller of the Currency. The final rule takes effect on July 1, 2019.

To qualify as private flood insurance under the final rule, a policy must be issued by an insurance company that meets certain conditions, and the policy must provide flood insurance coverage that is at least as broad as the coverage provided under a standard flood insurance policy (SFIP) issued under the National Flood Insurance Program (NFIP) for the same type of property, including when considering deductibles, exclusions, and conditions offered by the insurer. The final rule sets forth specific requirements that a policy must meet to be considered to provide coverage at least as broad as an SFIP.

The final rule requires that a regulated financial institution accept a private flood insurance policy that provides coverage at least as broad as an SFIP in satisfaction of the requirement for flood insurance under the applicable regulator’s flood insurance rules. The final rule allows a financial institution to determine that a private policy provides the necessary coverage if the following statement is included within the policy or as an endorsement to the policy: “This policy meets the definition of private flood insurance contained in 42 U.S.C. 4012a(b)(7) and the corresponding regulation.” A financial institution can simply rely on the statement, without independently assessing the policy. However, the final rule does not require insurance companies to include such a statement in a private policy, and a financial institution cannot reject a private policy simply because the statement is not provided.

The final rule also allows a financial institution to accept a private flood insurance policy that does not provide coverage at least as broad as an SFIP, if the private policy meets a more limited set of requirements, and the financial institution documents in writing its determination that the policy provides sufficient protection of the applicable loan, consistent with general safety and soundness principles.

Additionally, the final rule permits a financial institution to accept a plan issued by a mutual aid society in satisfaction of the requirement for flood insurance under the applicable regulator’s flood insurance rules, subject to conditions. The conditions are that the applicable regulator has determined that the plan qualifies as flood insurance for purposes of the Act, the plan provides specified coverage for the mortgagor and mortgagee as loss payees, and the financial institution documents in writing its determination that the plan provides sufficient protection of the applicable loan, consistent with general safety and soundness principles. For purposes of the final rule, a “mutual aid society” is defined as an organization (1) whose members share a common religious, charitable, educational, or fraternal bond, (2) that covers losses caused by damage to members’ property pursuant to an agreement, including damage caused by flooding, in accordance with this common bond, and (3) that has a demonstrated history of fulfilling the terms of agreements to cover losses to members’ property caused by flooding.

The CFPB recently issued A Regulatory and Reporting Overview Reference Chart for HMDA Data Collected in 2019. As previously reported, the Economic Growth, Regulatory Relief, and Consumer Protection Act created an exemption from the reporting of the new HMDA data categories for smaller mortgage loan volume depository institutions and credit unions. The 2019 Chart is updated from the 2018 version to include guidance on how to address data categories that do not have to be reported under the exemption. The 2019 Chart also provides how a lender that elects not to report a Universal Loan Identifier for an application or loan under the exemption would report a Non-Universal Loan Identifier for the application or loan.

Unrelated to the exemption, the 2019 Chart also includes additional guidance on the reporting of the Credit Scoring Model and the reporting of the Automated Underwriting System result.

Last week, the CFPB issued a Complaint Snapshot analyzing mortgage complaint data from November 1, 2016 to October 31, 2018. The report indicates an overall reduction in complaint volume for mortgage products over the time period analyzed.

Mortgage complaints declined by 15%, comparing the 3-month average of complaints from August to October 2017 with the 3-month average from August to October 2018. Similarly, mortgage complaints were down 18% in October 2018, as compared to the rolling 24-month average of complaints from November 1, 2016 to October 31, 2018.

Mortgage complaints declined from 2017 to 2018 in the majority of states (again, when comparing the 3-month average of August-October 2017 with that of August-October 2018). The states which saw an increase in mortgage complaints by this metric were Iowa (+48%), New Hampshire (+38%), New Mexico (+21%), Oregon (+17%), Florida (+7%), Oklahoma (+3%), and Connecticut (+1%). We note that the report excluded data for 12 states in which less than 25 mortgage complaints were received during the time frame covered in the report.

During the time period covered by the report, the CFPB received approximately 71,000 mortgage complaints through its Consumer Response system. As with prior data regarding mortgage complaints, the report shows that the majority of mortgage complaints involve servicing.

The two most common complaint categories were: (1) “trouble during payment process” – 42% of mortgage complaints; and (2) “struggling to pay mortgage” – 36% of mortgage complaints. Complaints classified as “trouble during payment process” alleged a wide range of problems, including issues with periodic statements, incorrect application of payments, incorrect shortages in escrow account analyses, and mishandled payoff statement requests. In the category of “struggling to pay mortgage”, complaints alleged issues such as difficulty receiving assistance following financial hardship, confusing denials of loan modifications, and unresponsiveness from the single point of contact.

The full report can be viewed here.

For years, the mortgage industry has urged the CFPB to issue informal written guidance on the TILA/RESPA Integrated Disclosure (TRID) Rule, as well as other rules.  The CFPB resisted, providing most guidance in the form of actual rules, webinars or oral statements.  The industry believed that it would be a cold day in Hades if the CFPB ever issued guidance similar to the FAQs that the Department of Housing and Urban Development issued to provide guidance on the 2010 Good Faith Estimate rule.  It may be more than a coincidence that the CFPB has issued the first four FAQs ever addressing TRID Rule issues on the heels of a severe cold wave.

Three of the four FAQs relate to corrected Closing Disclosures and the three business-day waiting period before consummation, topics that have proven a source of much confusion for creditors. The first FAQ provides that, in the event that there is a change to the disclosed terms after a creditor provides the initial Closing Disclosure, the creditor is required to ensure that the consumer receives a corrected Closing Disclosure at least 3 business days before consummation if:

  1. the change results in the APR becoming inaccurate;
  2. if the loan product information required to be disclosed under the TRID Rule has become inaccurate; or
  3. if a prepayment penalty has been added to the loan.

For other types of changes, the creditor can consummate the loan without waiting three business days after the consumer receives the corrected Closing Disclosure.

The FAQs also address how to handle a situation where an APR decreases (i.e., the previously disclosed APR is overstated). The FAQs state that if the overstated APR is accurate under Regulation Z (i.e., the difference between the disclosed APR and the actual APR for the loan is within an applicable tolerance in Regulation Z), the creditor must provide a corrected Closing Disclosure at or before consummation but a new three-business day waiting period is not required. An inaccurate APR, on the other hand, will trigger a new three-business day waiting period. The FAQ provides additional information related to APR accuracy, including a reference to the Federal Reserve’s Consumer Compliance Outlook. Please note that while the FAQs indicate that there is an APR overstatement tolerance that is tied to an overstated finance charge, various loan investors may not permit correspondent lenders to rely on the tolerance. It is advisable to check with your investor regarding their policy on this issue.

The CFPB’s FAQ clarifies that Section 109(a) of the Economic Growth, Regulatory Relief, and Consumer Protection Act (the “Act”) does not affect the requirement for providing a revised Closing Disclosure with another three-business day waiting period in cases in which the APR in the prior Closing Disclosure becomes inaccurate based on a decrease in the APR. This section provides that if a creditor extends a second offer of credit to a consumer with an APR that is lower than the APR disclosed in the prior Closing Disclosure, the transaction may be consummated without regard to the 3-day waiting period with respect to the second offer. However, as previously reported, and as noted by the CFPB, Section 109(a) did not create an exception to the TRID rule waiting period because Section 109(a) amends TILA Section 129(b), which only applies to high cost mortgage disclosures. The CFPB adds that TILA Section 128 sets forth a waiting period requirement for other credit transactions, and that such section was not amended by the Act. Clearly, Congress intended to modify the waiting period under the TRID rule, and simply made a technical error. The CFPB should act immediately to reflect the intent of Congress by proposing a rule to eliminate the need for a second three-business day waiting period when the APR in the prior Closing Disclosure becomes inaccurate because of a new offer of credit with a lower APR.

The last of the CFPB’s FAQs states that a creditor’s use of a model form will provide a safe harbor even if the model form does not reflect the changes to the regulatory text and commentary that were finalized in 2017. An appropriate model form must be properly completed with accurate content in order to meet the safe harbor for TRID Rule compliance. In July 2016 the CFPB had proposed to amend the TRID rule to indicate that the industry could not rely on various sample forms included in Appendix H to Regulation Z, but did not adopt the proposal based on significant opposition from the industry, as well as Fannie Mae and Freddie Mac.

For more information on these FAQs, stay tuned for an upcoming episode of the Consumer Finance Monitor Podcast.

On January 10, the CFPB published a report containing the results of its assessment of the Ability-to-Repay and Qualified Mortgage Rule (“ATR/QM Rule”) issued in 2013. The assessment was conducted pursuant to the Dodd-Frank Act, which requires the Bureau to review each significant rule it issues and evaluate whether the rule is effective in achieving its intended objectives, and the purposes and objectives of Title X of the Dodd-Frank Act, or whether it is having unintended consequences. The Bureau based the report on information gathered from a variety of sources, including:

  1. Loan origination and performance data from the National Mortgage Database (NMDB), Black Knight, CoreLogic, and HMDA
  2. Desktop Underwriter and Loan Prospector submissions and acquisitions data provided by Fannie Mae and Freddie Mac
  3. Application-level data from nine lenders covering over 9 million applicants
  4. Survey results from 190 lenders
  5. Supervision Data
  6. Residential mortgage backed securities (RMBS) data from IMF, Bloomberg, L.P., and SEC
  7. Cost data from the Mortgage Bankers Association’s (MBA) Annual Mortgage Bankers Performance Reports between 2009 and 2018
  8. Conference of State Bank Supervisors’ (CSBS) 2015 Public Survey data
  9. Evidence from comments received in response to the 2017 RFI concerning the ATR/QM assessmentThe ATR/QM Rule, which came into effect in January 2014, prohibits a lender from making a closed-end residential mortgage loan unless before closing the lender makes a reasonable and good faith determination, based on verified and documented information, that the consumer has a reasonable ability to repay (ATR). Qualified Mortgage (QM) loans are presumed to comply with the ATR requirement, except in the case of “higher priced” mortgage loans, where this presumption is rebuttable.Based on its survey of lenders, the Bureau found that a majority of respondents changed their business model due to the ATR/QM Rule in the form of increased income documentation, increased staffing, or adopting of a policy of not originating non-QM loans. The Bureau concluded that among the nine lenders that provided data, the changes resulted in lost profits of between $20 and 26 million per year. The Bureau also found that over the period of 2014 to 2016 the ATR/QM Rule eliminated between 63-70% of non-GSE eligible home purchase loans with debt-to-income (DTI) ratios above 43%. This impact did not carry over to refinance transactions, where lenders are more likely to extend credit due to a demonstrated ability to repay. The Bureau admits that because credit standards were already tight when the ATR/QM Rule took effect, “it is possible that the impacts would be different during times when credit is more abundant.”We note that the mortgage industry did not believe that a robust non-QM market would develop, and that the temporary GSE QM would be relied on heavily by lenders. And this is no surprise. The potential liability for violating the rule is significant, and based on the general standards for a non-QM loan there is no way for a lender, a due diligence firm or other party to conclusively determine if a given non-QM loan complies with the rule. As a result, a robust QM market did not develop, and it will never develop based on the current statute and rule. Based on the presumption of compliance with the rule, which is conclusive for non-higher priced loans, mortgage lenders mainly will originate a QM loan when possible. And based on the familiarity of the industry with Fannie Mae and Freddie Mac underwriting requirements, and the relative inflexibility of the standard QM based on the strict 43% DTI ratio limit and Appendix Q, the temporary GSE QM is favored by the industry. While the Bureau noted that the temporary GSE QM will expire no later than January 10, 2021, it did not address the fact that should the QM expire, mortgage lending would be severely constrained. Congress and/or the Bureau must act to prevent another mortgage crisis.The Bureau further concludes that the ATR/QM Rule does not appear to be constraining the activities of smaller lenders, who may originate QM loans that have DTI ratios above 43% without following Appendix Q, and may also originate QM loans that have balloon payments if various conditions are met, as long as such loans are held in portfolio for at least two years after the origination. In March 2016, the definition of a small creditor was amended to increase the loan threshold from 500 to 2,000 loans per year. According to the Bureau, this amendment had ameliorative effects – (1) the geographic market coverage of small creditors increased substantially with the new threshold, increasing access to credit for borrowers in rural and underserved areas who have DTIs above 43%; and (2) the share of loans made by depository institutions that were small creditors almost doubled.
  10. We note that in May 2018, Senate Bill 2155, the Economic Growth, Regulatory Relief, and Consumer Protection Act, was signed into law, creating a new QM category for insured depository institutions and insured credit unions that have, together with their affiliates, less than $10 billion in total consolidated assets. The Act provides that the new QM loan is deemed to comply with the ATR requirements if the loan: (1) is originated by and retained by the institution, (2) complies with requirements regarding prepayment penalties and points and fees, and (3) does not have negative amortization or interest-only terms. Furthermore, the institution must consider and verify the debt, income, and financial resources of the consumer. Beyond a minor footnote, the Bureau does not address this amendment in its report, likely because it still needs to implement the law.
  11. The Bureau also notes that innovation is occurring in the Temporary GSE QM space in the area of income verification and calculation, because compliance with Appendix Q is not required (loans made under the standard QM that is based on the strict 43% DTI ratio limit must follow Appendix Q). While the innovation is positive, it does not address the underlying need to continue, or find a suitable alternative, for the Temporary GSE QM.
  12. Although the rule includes a standard QM that is based on a strict DTI ratio limit of 43%, the Bureau created a temporary QM for loans eligible for sale to Fannie Mae or Freddie Mac “to preserve access to credit for consumers with debt-to-income ratios above 43 percent during a transition period in which the market was fragile and the mortgage industry was adjusting to the final rule.” The Bureau notes that it “expected that there would be a robust and sizable market for non-QM loans beyond the 43 percent threshold and structured the Rule to try to ensure that this market would develop.”
  13. In its assessment, the Bureau found that the introduction of the ATR/QM Rule was generally not correlated with an improvement in loan performance (as measured by the percentage of loans becoming 60 or more days delinquent within two years of origination). Rather, the Bureau concludes that delinquency rates on mortgages originated in the years immediately prior to the effective date of the ATR/QM Rule were historically low, “as credit was already tight at that time.” Moreover, although the performance of non-QM loans did not improve in absolute terms under the ATR/QM Rule, it has improved relative to the performance of comparable QM loans.
  14. The CFPB states that the report does not include a cost-benefit analysis of the ATR/QM Rule or its provisions, but that “each report does address matters relating to the costs and benefits.” The CFPB indicated that going forward, it will reconsider whether to include such an analysis in its assessment.

The Bureau did not announce any further action relative to the ATR/QM Rule but did indicate that reactions from stakeholders to the reports’ findings and conclusions would help inform future policy decisions. The Bureau concurrently released a report assessing the RESPA Mortgage Servicing Rule, which we will analyze separately.

As previously reported, in September 2017 the CFPB proposed policy guidance regarding what application-level Home Mortgage Disclosure Act (HMDA) data would be disclosed to the public based on the significant expansion to the HMDA data reporting items that the CFPB adopted in October 2015. Calendar year 2018 was the first year that reporting institutions collected data under the expanded requirements, and that data must be reported to the government by March 1, 2019. Not long after Kathy Kraninger became the new CFPB Director, the CFPB announced the final policy guidance regarding the application-level HMDA data that will be made available to the public. Unfortunately, by adopting final guidance that is very similar to the proposed guidance, the CFPB is emphasizing public disclosure over consumer privacy concerns.

HMDA requires the modification of data released to the public “for the purpose of protecting the privacy interests of mortgage applicants”. Under the prior HMDA requirements, the application or loan number, the date the application was received and the date the institution took final action on the application were removed from the application-level data that was released to the public. However, even under the prior HMDA requirements, there were concerns that by combining the publicly available HMDA data with other data sources, the identity of each applicant can be determined. With the significant expansion of the HMDA data items, as well as the overall increase in data on consumers that is available in the marketplace, the privacy concerns are even greater. For example, the revised HMDA data items include, among other items, the applicant’s age, income (which is currently reported), credit score, and debt-to-income ratio; the automated underwriting results; the property address; loan cost information; and, for denied applications, the principal denial reasons.

When the CFPB adopted the October 2015 revisions it deferred making a decision on which elements of the expanded HMDA data would be reported on an application-level basis. However, the CFPB indicated that it would use a balancing test to decide what information to disclose publicly, and would allow public input on the information that it proposed to disclose. At that time the CFPB advised that
“[c]‍‍‍onsidering the public disclosure of HMDA data as a whole, applicant and borrower privacy interests arise under the balancing test only where the disclosure of HMDA data may both substantially facilitate the identification of an applicant or borrower in the data and disclose information about the applicant or borrower that is not otherwise public and may be harmful or sensitive.” The CFPB echoed the balancing test approach in adopting the final guidance. However, the approach is off balance, as it sides too much on the side of public disclosure, ignoring valid consumer privacy concerns.

Under the final policy guidance, the CFPB will make all of the HMDA data available to the public on an application-level basis, except as follows:

  • The following information would not be disclosed to the public (the non-disclosure of the first three items is consistent with prior public disclosure practices):
    • The universal loan identifier.
    • The date the application was received or the date shown on the application form (whichever was reported).
    • The date of the action taken on the application.
    • The property address.
    • The credit score(s) relied on.
    • The NMLS identifier for the mortgage loan originator.
    • The automated underwriting system result.
    • The free form text fields for the following (the standard fields reported would be disclosed):
      • The applicant’s race and ethnicity.
      • The name and version of the credit scoring model.
      • The principal reason(s) for denial.
      • The automated underwriting system name.
  • The CFPB will disclose in a modified format the loan amount, age of the applicant, the applicant’s debt-to-income ratio, the property value, the total individual dwelling units in the property, and for a multi-family dwelling the individual dwelling units that are income-restricted.
    • For the loan amount, the CFPB will disclose:
      • The midpoint for the $10,000 interval into which the reported value falls, such as $115,000 for amounts of $110,000 to less than $120,000. (Under prior requirements the loan amount was reported to the nearest $1,000, and the reported amount was disclosed to the public.)
      • Whether the reported loan amount exceeds the Fannie Mae and Freddie Mac conforming loan limit.
    • For the age of the applicant, the CFPB will disclose:
      • Ages of applicants in the following ranges: Under 25, 25 to 34, 35 to 44, 45 to 54, 55 to 64, 65 to 74, and over 74.
      • Whether the reported age is 62 or over. For purposes of the Equal Credit Opportunity Act, a person is considered elderly if they are age 62 or over.
    • For the debt-to-income ratio, the CFPB will disclose:
      • The reported debt-to-income ratio for reported values of 36% to less than 50%, and other debt-to-income ratios in the following ranges: under 20%, 20% to less than 30%, 30% to less than 36%, 50% to less than 60% and 60% or higher. As proposed, the reported debt-to-income level would have been disclosed for reported values of 40% to less than 50%.
    • For the property value, the CFPB will disclose the midpoint for the $10,000 interval into which the reported value falls, such as $115,000 for amounts of $110,000 to less than $120,000.
    • For the total individual dwelling units in the property, the CFPB will disclose:
      • The reported number of units for reported values below 5, and other unit numbers in the following ranges: 5 to 24, 25 to 49, 50 to 99, 100 to 149 and over 149. The CFPB had proposed to disclose the actual number of units reported.
    • For the individual dwelling units in a multifamily property that are income-restricted, the CFPB will disclose the reported value as a percentage, rounded to the nearest whole number, of the value reported for the total individual dwelling units. The CFPB had proposed to disclose the actual number of units reported.

Although the loan amount will now be reported in the applicable $10,000 interval and not to the nearest $1,000, the concern is that the totality of the information that is publicly available will make it easier than it is today to determine the identity of the applicant. Thus, based on the final policy guidance, there is a risk that a significant amount of information that consumers view as being confidential will become publicly available. The CFPB essentially dismissed most privacy concerns raised by parties commenting on the proposed policy guidance.

While the final policy guidance favors public disclosure over consumer privacy concerns, the final policy guidance may have a limited life. As previously reported, the CFPB has indicated that it plans to reopen HMDA rulemaking to reconsider various aspects of the revised HMDA rule, including the discretionary data points that were added by the CFPB. We noted previously that the CFPB’s Fall Rulemaking Agenda has a target of May 2019 for the issuance of a notice of proposed rulemaking. Potentially, the CFPB may eliminate entirely, or modify, one or more data categories that are included in the information disclosed publicly. The CFPB states as follows in the supplementary information to the final policy guidance: “The Bureau intends to commence a rulemaking in the spring of 2019 that will enable it to identify more definitively modifications to the data that the Bureau determines to be appropriate under the balancing test and incorporate these modifications into a legislative rule. The rulemaking will reconsider the determinations reflected in this final policy guidance based upon the Bureau’s experience administering the final policy guidance in 2019 and on a new rulemaking record, including data concerning the privacy risks posed by the disclosure of the HMDA data and the benefits of such disclosure in light of HMDA’s purposes.” Thus, in addition to reconsidering the HMDA rule itself, the CFPB already plans to reconsider the approach to public disclosure taken in the final policy guidance.

The adoption of policy guidance that favors public disclosure over consumer privacy appears contrary to the approach of former Acting Director Mulvaney. Perhaps the action reflects that Director Kraninger will take a different approach. Or perhaps Director Kraninger simply decided to initially punt on the issue. The guidance needed to be adopted in view the impending March 1 filing deadline for HMDA data. But as noted above, the CFPB intends to revisit both the HMDA rule itself and the policy guidance. Perhaps Director Kraninger will reassess the approach during the upcoming rulemaking.

Note that while the approach to the public disclosure of HMDA data is adopted in the form of public guidance and not a formal rule, the CFPB advises in the supplementary information to the guidance that pursuant to the Congressional Review Act it will file the guidance with Congress. As we previously reported, the Government Accountability Office determined that the CFPB bulletin on “Indirect Auto Lending and Compliance with the Equal Credit Opportunity Act” was a rule subject to the Congressional Review Act, and subsequently under the Act Congress passed, and President Trump signed, legislation disapproving the bulletin. Presumably, this result influenced the decision of the CFPB to file the policy guidance with Congress under the Act.