As the year 2021 winds down, it is time to start thinking about ensuring that your HMDA submission is ready to be filed by March 1, 2022. Before the end of this year Compliance Officers, and those ultimately responsible for the HMDA submission signoff, should use this time to make sure that the HMDA data not only pass the CFPB’s edits checks but also pass lending industry-based sanity checks.
Here are three reasons why it is important to go beyond the HMDA Edit Checks before submitting your data to CFPB:
- Validity is not verification. Passing the edit checks do not mean your data is correct; it only means your data is “valid” in terms of the required HMDA format. Therefore, it is incumbent upon the lender to engage in data verification or HMDA quality assurance using a tool like HMDA Ready™ and LendingPatterns™. This quality assurance effort should seek to answer questions like the following:
- Alignment with industry benchmarks. For this you would first go back to the previous year and ask yourself some questions. Does my HMDA data look like my industry segment’s data in terms of reporting? For example, is your reporting distribution of Action Taken categories, the proportion of applications with race/ethnicity, gender and age not provided; the distribution of denial reasons; the reporting of applicant income; to name a few, in line with the industry? If there are material reporting misalignments, you should investigate further.
- Justify the misalignment. Just because your HMDA reporting profile is materially out of line with other lenders does not necessarily mean your institution’s data is wrong. However, you should document the reasons why your profile is different. Some possible reasons may be because of how you define a “HMDA application,” your unique business model, the type of origination or loan sourcing strategies employed, your automated underwriting system, or whether and how your personnel (or system) is trained (programmed) to request the HMDA government monitoring information.
- Need to Assess Changes in PPP impact regularly. Changes in origination policy, practices, and procedures can reveal themselves in the HMDA data. You will want to a) keep a PPP Change Journal going forward, and b) look back over the past year for PPP changes that could have had an impact on your soon-to-be-filed HMDA data. Even if HMDA data are not affected, your document will be extremely useful for your fair lending monitoring as “Plus data” to control for factors not available in HMDA.
- System of record for fair lending. A file that perfectly passes the CFPB edit checks can be seriously flawed for fair lending analysis. There are many HMDA data issues that seem innocuous on the surface that can have profound results on fair lending analyses. For example, the strictest definition of an application for TRID may seem appropriate for most mortgage lending activities. Most lenders use the same definition for HMDA. However, this can potentially lead to underreporting of HMDA reportable credit decisions and a distortion of prohibited basis disparities. Lenders in the broader consumer market should be on alert to HMDA data that reveal extremely low denial rates. This could be the result of the application definition excluding HMDA reportable applications. Such an exclusion can make the lender appear to be unfair when they are not.