February 2025 Newsletters

New Year, New Thresholds

In case you missed any of these updates, annual threshold adjustments have been made for HMDA, Reg Z, Reg M, and HOEPA as of the time of this publication.

HMDA Thresholds for 2025

Annual adjustments to these asset-size thresholds are based on the change in the average of the Consumer Price Index (CPI) for Urban Wage Earners and Clerical Workers, not seasonally adjusted, for each 12-month period ending in November, with rounding to the nearest million.

For 2024, the threshold was $56 million. During the 12-month period ending in November 2024, the average of the CPI-W increased by 2.9 percent. As a result, the exemption threshold is increased to $58 million. Thus, banks and savings associations with assets of $58 million or less as of Dec. 31, 2024 are exempt from collecting data in 2025. That being said, a bankā€™s exemption from collecting data in 2025 does not affect its responsibility to report data it was required to collect in 2024.

Reg Z and Reg M Thresholds for 2025

Dodd-Frank Act amended TILA and CLA by requiring annual adjustments to the thresholds based on the annual percentage increase in the consumer price index for urban wage earners and clerical workers. Only those transactions at or below the thresholds are subject to the protections of the regulations.

Based on the annual percentage increase, TILA and CLA will apply to consumer credit transactions and consumer leases of $71,900 or less in 2025.

Keep in mind, this particular exemption threshold is not applicable to some types of credit, as listed below:

  • Loans secured by any real property, or by personal property used or expected to be used as the principal dwelling of the consumer; or private education loans.
  • Private education loans and loans secured by real property (such as home mortgage loans) or the consumerā€™s principal dwelling remain subject to TILA regardless of the amount of the loan.

HOEPA and QM Threshold Adjustments

This section of the regulation contains a points and fees coverage test for use in calculating whether a transaction is a high-cost mortgage. Where applicable, the total points and fees thresholds are as follows:

  • 5% of the total loan amount for loans greater than or equal to $26,968 (32(a)(1)(ii)(A)).
  • 8% of the total loan amount or $1,348, whichever is less, for loans less than $26,968 (32(a)(1)(ii)(B)).

Ability to Repay/Qualified Mortgages (ATR/QM): This section of the regulation contains standards for determining whether a transaction is a QM. In part, a transaction is not a QM if the transactionā€™s total points and fees exceed certain thresholds, based on specific loan amounts (1026.43(e)(3)(i)). Where applicable, the thresholds are as follows:

  • Points and fees may not exceed 3% for a loan amount greater than or equal to $134,841
  • Points and fees may not exceed $4,045 – for loans greater than or equal to $80,905 but less than $134,841
  • Points and fees may not exceed 5% for loans greater than or equal to $26,968 but less than $80,905
  • Points and fees may not exceed $1,348 for loans greater than or equal to $16,855 but less than $26,968
  • Points and fees may not exceed 8% – for loans less than $16,855.

As always, if you have questions about updates to thresholds or any other compliance concerns, please reach out to the Compliance Hub Hotline via chat, phone, or email.

Are the Algorithms Playing Fair? CFPB Looks at AI in Credit Scoring

In its January 17, 2025 special edition of Supervisory Highlights, the Consumer Financial Protection Bureau reiterated its concerns regarding fair lending risks related to new technologies. In this case, the focus is on complex automated underwriting systems, particularly those that use artificial intelligence (AI) or machine learning (ML).

The guidance highlights two major concerns. The first is whether lenders are testing their credit scoring models to determine if the models use prohibited factors, or proxies for prohibited factors, in their decisioning or if the scoring model is resulting in disparate impacts. The concern about prohibited factors is fairly clear; lending decisions should not be based on prohibited factors. That should, at least in theory, be something that AI/ML systems can be simply programmed not to do.

Disparate impact is a bit more complicated. As stated in regulatory guidance on fair lending, disparate impacts are not absolutely prohibited as prohibited factors are, but where a credit decisioning method results in disparate impacts, a lender should be able to demonstrate a business justification for the method and also that the lender is employing a model that minimizes disparate impacts while meeting legitimate business needs.

The guidance therefore indicates that lenders should be testing for fair lending concerns when using AI/ML models and, where disparate impacts are found, documenting the process for selecting the model chosen. This should include documenting the business needs the model serves, including specific standards that are used to evaluate whether the model meets those needs. In addition, the lender should be able to demonstrate that it has reviewed other models to determine whether the identified business needs can be met by another model with less discriminatory effects. Without documenting the model selection process in which models are compared based on their ability to further business needs and reduce disparate impacts, a bank may not be able to demonstrate that it has adhered to fair lending requirements.

The second concern discussed in the guidance is the use of ā€œalternative data,ā€ which may include hundreds of different variables, used in more complex credit scoring models. Variables that are not clearly related to consumersā€™ finances may invite scrutiny from a fair lending perspective, both in terms of whether the variable is truly related to creditworthiness and also whether it may be proxy for a prohibited factor. Lenders should therefore be able to demonstrate the relevance of variables that are entered into automated underwriting systems, as the agencies have discussed more specifically in other guidance.

Additionally, the guidance notes that the use of complex AI/ML credit scoring models with dozens or hundreds of variables does not alleviate a lenderā€™s responsibility to identify the specific reasons for adverse actions. Banks must therefore be able to demonstrate that the reasons stated on adverse action forms are, in fact, the primary reasons that an adverse action was taken. The guidance states that lenders should be able to demonstrate through testing that the methods used to identify the primary reasons are reliable and accurate.