February 2025
In January 2023, the Office of the Comptroller of the Currency (OCC) released a revised version of the Comptroller’s Handbook on Fair Lending,” including guidelines and processes for examiners to evaluate the risk of fair lending and assess compliance with fair lending regulations. It preserves the essence of the current procedures for fair lending examinations from the January 2010 version and adds important details on examination scenarios, risk factors, and sound risk management practices.
Given the updated guidelines on bank risk management processes, Promontory expects the OCC will take an aggressive approach to fair lending risk in the context of models and model risk management (MRM). We expect firms may face certain challenges in implementing effective fair lending data analytics and MRM frameworks.
In our experience, many banks have sufficient processes to evaluate individual controls; however, they often struggle to assess the effectiveness of the overall system of fair lending risk controls.
The handbook highlights statistical analyses for high-volume focal points,6 especially underwriting, pricing, and redlining assessments. When activity volume is significant, data analytics help drive fair lending assessments by using statistical analysis to identify potential risk areas and direct investigative efforts. However, in practice, we see banks fail to identify and mitigate significant data issues and modeling limitations, thus failing to realize the full benefits of statistical analyses. Additionally, firms struggle to incorporate fair lending models in an MRM framework due to differences in construction and usage from risk models. It’s important to address data, modeling, and post-analysis file review for an effective fair lending framework.
In performing statistical analyses and comparative file reviews, banks should first evaluate the sample size and create homogeneous segments. Comparative file reviews, which evaluate differences in outcomes for similar applicants, complement statistical analysis in fair lending assessments, providing the additional benefit of identifying data gaps and overlooked factors in models.
When considering data accuracy in fair lending reviews, banks should focus on the fields used in statistical analyses. In the case of mortgage lending, this requires going beyond the reportable fields in the Home Mortgage Disclosure Act reportable fields (HMDA). Banks should also use demographic proxy methodologies, e.g., Bayesian Improved First Name Surname and Geocoding (BIFSG), to analyze material segments for which demographic information is not collected, e.g., credit cards.
While regulators have issued detailed documentation, development, and validation standards for modeling credit, market, and other risks, there is less guidance for fair lending models and related MRM. We see several key challenges for fair lending analyses and MRM:
The handbook acknowledges that automated and validated credit scoring models present lower fair lending risk. But, complex scoring systems (i.e., those relying on machine learning or data seldom used for credit decisions) carry the greatest fair lending risk, and it instructs examiners to assess the MRM model risk management practices for machine learning.
Risks increase with machine learning or alternative data, as the modeled relationships are no longer explicit and the data is more likely to hide proxies of the protected classes. Firms and their vendors should also search for less discriminatory alternatives (LDA). In the case of machine learning, it is common to find, often aided by automated tools, alternative models with similar performance but less bias.
The file review process should be statistically driven. We observe that the overall process is often poorly rationalized, and statistical analysis is not used sufficiently to determine sample sizes and files to be reviewed. We advise our clients to tie the process to the firm’s risk appetite, which also helps interpret the results and size the response when issues are found.
We also observe that fair lending analysis and file reviews are often not proactive and are performed with a substantial delay. Many institutions perform their analyses annually and skip ongoing monitoring. On-going monitoring could yield significant benefits, including early detection of risks ahead of the annual assessment, even if using simple techniques like data raw analysis along key business-specific dimensions.
Read a version of this insight on IBM Promontory Blog.
Read other insights: