ÀÖÓ㣨Leyu£©ÌåÓý¹ÙÍø

ÀÖÓ㣨Leyu£©ÌåÓý¹ÙÍø Insights:

• Model Anti-bias/Trust: Long-awaited rule adds to evolving regulations for anti-bias and trust in model/algorithm development and use.
• Quality Controls: Focus on quality control factors (including estimate confidence and fair lending considerations), in areas of model risk management, third party oversight and “automated systemâ€� risk management.
• Hurdles Notwithstanding: ÌýDespite potential difficulties to accessing a model’s data and design and/or the use of third parties, regulators will expect sound anti-bias/anti-discrimination in the institution’s model risk management controls.
• Institutions Responsible: ÌýRegulators will expect institutions that utilize AVMs to “own the riskâ€�.

_______________________________________________________________________________________________________________________________________

June 2024 (Updated July 2024)

Six federal agencies jointly release a to implement quality control standards for automated valuation models (AVMs) used by mortgage originators and secondary market issuers (each as defined in the rule) to determine the value of mortgage collateral securing a consumer’s principal dwelling.

The final rule is substantially the same as proposed in June 2023. It will become effective the first day of the calendar quarter following the date that is 12 months after publication in the Federal Register. The FDIC and OCC are the first of the six participating agencies (FRB, OCC, FDIC, CFPB, NCUA, and FHFA) to adopt the final rulemaking.

Note: The quality control standards are mandated by Section 1473(q) of the Dodd- Frank Act, which added a new Section 1125 to FIRREA.

Quality Control Standards

Under the final rule, mortgage originators and secondary market issuers that engage in credit decisions or covered securitization determinations themselves, or through or in cooperation with a third-party or affiliate, must adopt and maintain “policies, practices, procedures, and control systems� to ensure that AVMs used in these transactions adhere to quality control standards, or “factors�, designed to:

• Ensure a high level of confidence in the estimates produced by AVMs.
• Protect against the manipulation of data.
• Seek to avoid conflicts of interest.
• Require random sample testing and reviews.
• Comply with applicable nondiscrimination laws.

Note:Ìý The Dodd-Frank Act provided that the agencies could add other factors, as appropriate, to the quality control standards and, for purposes of the current rule, they have added the standard to “comply with applicable nondiscrimination lawsâ€�.

The final rule does not set specific requirements for how entities are to structure these policies, practices, procedures, and control systems though the agencies expect institutions to establish quality controls based on their size and the risk and complexity of transactions for which they will use AVMs covered by the rule. The agencies� existing guidance related to AVMs remains applicable.

Applicability

Section 1125 requires financial institutions, or subsidiaries owned and controlled by a financial institution and regulated by a Federal financial institution regulatory agency, to comply with regulations issued under the subsection as appropriate. For these purposes, AVMs are defined as “any computerized model used by mortgage originators and secondary market issuers to determine the value of a consumer’s principal dwelling collateralizing a mortgage.� The quality control standards apply as follows:

Ìý

Reconsiderations of Value - Final Guidance

In separate but related actions, five federal agencies (FRB, FDIC, OCC, CFPB, and NCUA) jointly release on ROVs (reconsiderations of value) for residential real estate valuations. The agencies state the final guidance, which has been adopted “largely as proposedâ€�, is intended to highlight risks associated with “deficient valuationsâ€� and describe how financial institutions may incorporate ROV processes and controls into established risk management functions.Ìý

In particular, the final guidance:

• Clarifies that the scope of the guidance is limited to real estate-related transactions secured by single 1-to-4 family residential property.Ìý
• Describes the “deficient valuationsâ€� to include “prohibited discrimination; errors or omissions; or valuation methods, assumptions, data sources, or conclusions that are otherwise unreasonable, unsupported, unrealistic, or inappropriate.â€�
• Identifies the risks of “deficient valuationsâ€� to include “loan losses, violations of law, fines, civil money penalties, payment of damages, or civil litigation.â€�
• Notes that “deficient valuationsâ€� may be identified through an institution’s valuation review processes or through consumer provided information.
• Outlines applicable statutes, regulations, and existing guidance that govern ROVs and collateral valuations, such as ECOA, Regulation B, Fair Housing Act, Truth-in-Lending Act, Regulation Z, and USPAP.
• Explains how ROV processes and controls can be incorporated into existing risk management functions, such as third-party risk management, appraisal review, and consumer complaint management.
• Provides examples of ROV policies, procedures, and controls that financial institutions may adopt to identify, address, and mitigate the risk of “deficient valuationsâ€�. Examples of topics include:
  • Roles and responsibilities for business units processing an ROV request.
  • Processes for identifying, managing, analyzing, escalating, and resolving valuation-related complaints and inquiries across relevant lines of business and from various channels and sources.
  • Consumer communication on how to raise concerns about valuations sufficiently early in the underwriting process for errors or issues to be resolved before a final credit decision is made.
  • Training, for relevant staff and third parties, to identify “deficienciesâ€� throughout the valuation review process.

Effective Date. The guidance is final upon publication in the Federal Register.