In November 2021, CFPB issued a request for information inviting the public to comment on the effectiveness of HMDA.
HMDA, the Home Mortgage Disclosure Act, requires most financial institutions to report certain data points about their mortgage transactions. This results in a publicly disclosed annual data set of over 20 million transactions. This data is unique because it serves three macro use cases at once:
- Fair access to mortgage financing
- Better business operations
- Analysis for policy decisions
Given the size of the data, however, an annual HMDA data set cannot be analyzed in its entirety with desktop software like Excel. Polygon Research solves this last mile problem by taking the annual HMDA data – 5 years of it, in fact – and unlocks it in a powerful, cloud-based analytical platform. Our product, HMDAVision, has been on the market since 2018. Over the past 4 years, we’ve gotten to know the HMDA data very well, which put us in a good position to respond to the CFPB's RFI.
So what did we have to say to the CFPB?
We’ll provide one example here. See the link at the bottom of this post for our complete response.
HMDA is an ecosystem of lenders, the CFPB, analytics providers like Polygon Research, and a myriad of other stakeholders. We feel the CFPB has done a commendable job with HDMA to date. Challenges remain, however. A core example is its responsibility for implementation decisions on institutional coverage and data scope. Another is improving and ensuring data quality. While the former is solely in the CFPB's domain, the latter falls in a more collaborative space. For instance, in a HMDA tech sprint hosted by the CFPB last year, Polygon Research formed one of 17 teams focused on improving data submission or publication. Since HMDAVision was already our answer to data publication, we rolled up our sleeves to see how we could drive innovation on HDMA data submission. Critically, this is an area where costs pertain to lenders, and a chief nexus of the question about expanding or contracting HDMA.
The improvement we offered to CFPB in our RFI response was simple, but important: the technology entered into its tech sprints be open sourced – e.g. placed under an open source license and posted to GitHub – or, if the results fall short of working code, described in detail in a public post. This would keep participants from bringing pre-baked solutions they’re not willing to share, and would at the same time serve as a multiplier effect on the type of innovation that stands to tame HMDA reporting costs. We feel the CFPB will be most effective by solving problems only it can solve, and by fostering collaboration and innovation for the rest.
Polygon Research's Response to CFPB's RFI: Regulations.gov