Case Study: Fair Lending Analysis and Risk Mitigation

Premier Insights, Inc. conducted a fair lending assessment and analysis for “Pine Haven Bank” fictional name, focusing on mortgage and consumer loan underwriting. The assessment involved a risk assessment of Pine Haven Bank's lending function and a review of HMDA data. Statistical analysis and a file review were conducted to identify potential disparate treatment and impact. The study found no evidence of discrimination but highlighted areas for process improvement to mitigate fair lending risk and address denial rate disparities. Recommendations were provided to enhance Pine Haven Bank's fair lending program.

This case study is intended to illustrate Premier Insights, Inc.'s analytical capabilities. The names of the financial institution and market areas have been changed for confidentiality purposes.

Introduction

Premier Insights, Inc. was retained by Pine Haven Bank, a community bank serving the fictional markets of Riverton, Meadowbrook, and Hillside, to address fair lending concerns identified in a recent regulatory examination. This case study highlights the approach, findings, and recommendations of Premier Insights in assisting Pine Haven Bank to enhance its fair lending program. The focus was on analyzing lending practices, identifying areas of risk, and providing specific recommendations to strengthen the bank’s compliance efforts. The analysis focused on underwriting and HMDA lending, as this was the focus of the regulatory examination.

Scope of Review

The project involved two main components:

• A fair lending risk assessment of the bank's lending function to identify areas of potential risk.
• A fair lending underwriting review of HMDA reportable mortgage loan applications for the calendar year that was analyzed, including a physical file review and statistical analysis.

While all lending areas were considered, the review emphasized mortgage and consumer credits. The goal was to assess for potential disparate treatment and ensure adherence to bank policy.

Fair Lending Risk Assessment

The risk assessment examined several key areas within the bank's lending function, including:

• Underwriting Process: The bank used a software application to process loan applications, which generated a recommended decision. However, the system only considered credit score, DTI, and LTV. The underwriting process lacked consideration of other critical factors like employment history and derogatory credit issues, which could lead to inconsistencies. Recommendations included expanding the underwriting process, better defining mitigating factors, and clarifying when these factors apply.
• Steering: The bank offered various consumer loan products, including fixed-rate and variable-rate options, which could lead to steering risk. There was a potential for overlap between consumer and commercial products. Recommendations included clarifying distinctions between overlapping products and clearly defining which products are appropriate for certain situations.
• Monitoring within the Lending Function: While the software system allowed for overrides, exceptions needed to involve senior management and be documented to ensure consistency and fair lending implications.
• CMS Monitoring: The bank's compliance staff did not conduct reviews related directly to fair lending. A recommendation was made to implement a fair lending monitoring program.
• Second Review Process: While the bank had a second review process for denied applications, this could be strengthened by having senior or executive management conduct the reviews using a checklist to ensure all mitigating factors were reviewed equally.
• Denial Rate Disparities: Analysis showed a significant disparity in denial rates between black and white applicants, with black applicants being denied at a much higher rate (40.79%) compared to white applicants (5.17%). However, once underwriting criteria were considered, these disparities were not statistically significant, although such a large difference in denial rates still elevates risk. The difference appeared to be largely driven by the low denial rate for white, non-Hispanic applicants. The recommendation was that the Bank should be aware of denial rate differences and understand the risk associated with such disparities.
• Data Integrity: Some inconsistencies were found between the data in the electronic file and the imaged files, especially with DTI and LTV calculations. The recommendation was that the Bank address these problems with their data management.
• Marketing: The bank advertised in print, radio, billboards, and social media, but most of the advertising was for brand awareness rather than specific products. The Bank operated in predominantly minority counties. The recommendation was that the Bank should continue to ensure that marketing efforts are general to all of the Bank's markets and that minority communities are being reached.

Statistical Analysis and File Review

A statistical analysis of loan data was performed, focusing on black versus white and male versus female applicants. The analysis used data from the bank’s loan origination system and HMDA LAR. The data was limited to owner-occupied, first-lien applications.

• Initial analysis showed a statistically significant disparity in denial rates for both black and female applicants.
• After controlling for credit score, DTI, and LTV, the disparity for female applicants became statistically insignificant, while a disparity for black applicants remained.
• However, once low-side/high-side factors were added into the regression, all conditioned disparities became statistically insignificant, indicating that differences were accounted for by policy criteria.

A physical file review of a sample of mortgage loan applications was also conducted. The review included approved loans with deficiencies and denied loans with higher credit scores to evaluate if policy was consistently applied. The review indicated that the decisions were generally based on bank policy and creditworthiness. However, inconsistencies, documentation issues, and errors in DTI and LTV calculations were noted. Importantly, the decisions were based on the data shown in the files, meaning that while there were calculation errors, they did not suggest discrimination.

Conclusion

The analysis conducted by Premier Insights, Inc. provided Pine Haven Bank with valuable insights into their lending practices. While no findings of discrimination were identified, the bank would benefit from enhancing its processes with regard to fair lending. The recommendations offered aimed to improve the bank’s evaluation and mitigation of fair lending risk. The findings of the analysis should help the bank make more consistent and fair decisions while improving regulatory compliance and fair lending practices.

Premier Insights, Inc. Expertise

Premier Insights, Inc. is a research-consulting firm specializing in fair lending and Community Reinvestment Act (CRA) services. The firm, founded by Brandon A. Roberts, Ph.D., brings extensive experience in banking, economics, and policy. Roberts is a published author and has consulted with numerous financial institutions. His academic background and industry expertise make Premier Insights, Inc. uniquely positioned to assist financial institutions with complex regulatory and compliance matters.