Dave Parker: CFPB’s New Attitude Toward HMDA Fuels Need for Better Tools
Dave Parker is Chief Product Officer for LoanLogics, a Jacksonville, Fla.-based provider of loan quality technology for mortgage manufacturing and loan acquisition. He is responsible for defining and executing the vision and direction of the company’s product portfolio and designing new solutions for the industry’s current challenges. He has more than 30 years of experience in mortgages and technology. Inquiries can be sent to LoanLogicsInfo@LoanLogics.com.
As it typically happens with a new presidential administration, there’s a new attitude in Washington toward the housing market. And one of the most significant changes has been the CFPB’s recent decision to roll back flexibility when reporting Home Mortgage Disclosure Act data. But are lenders ready?
CFPB Acting Director Dave Uejio seems to think so, saying one reason for the change is that “many financial institutions have developed more robust remote capabilities and demonstrated improved operations.” While this may be true, however, many lenders have not developed great HMDA reporting capabilities—nor are they maximizing the power of automation for HMDA audits and LAR report submissions. But they certainly can.
How Automation is Streamlining HMDA Reporting
In March 2020, at the beginning of the COVID-19 pandemic and the mammoth shift within the mortgage industry toward remote work, the CFPB announced certain lenders were no longer required to report quarterly information under HMDA so they could focus on serving borrowers. The move was never meant to be permanent, however, and now the agency will require all financial institutions to submit their 2021 first quarter data by May 31. With that, the CPFB plans to exercise the full scope of its supervisory and enforcement authority provided under the Dodd-Frank Act.
Regardless of how thorough human staff are, manual audits are incapable of providing the confidence to either lenders or regulators that HMDA data is accurate. It’s important to remember that the Federal Financial Institutions Examination Council (FFIEC) has the power to test whatever loan samples they choose, including a single sample from a lender’s entire HMDA LAR. If a lender’s LAR data contains just a few errors on sample loans, the CFPB would require lenders to resubmit it.
However, rules-based, data-driven digital technology can ease the burden of HMDA reporting by helping lenders shift from what is a traditionally cumbersome manual process to a highly automated one. The key is combining automated document processing technologies that can cull through thousands of data points across hundreds of pages in loan file documents with automated HMDA auditing tools to quickly “pass” the bulk of “clean” loan files and identify only a subset of files that require a second look.
For HMDA reporting, automation helps replace manual processes in three specific areas—data validation, audits, and data transformation. While beneficial to any lender, this is extremely helpful for originators with a large amount of reportable applications/loans. In this case it becomes impractical to export rows and rows of data into an Excel file, manually review data for existence and accuracy, and create macros to do the data transformation for this final important step in the reporting process.
Data validation involves the first set of checks done on HMDA data, which are often called edit checks. These checks are made to determine whether the lender has all the data they need for HMDA reports and whether there is any missing, inconsistent or conflicting data in their files. For example, they can confirm the existence of reasons for denied loans or that the borrower’s credit score has the correct number of digits.
For audit testing, automated rules can be used to perform quality checks or “data scrubs” for defects before reports are submitted to regulators. You might think that these would be covered in your pre- and post-close quality reviews, but the problem is these reviews do not include loan applications that never make it through the origination process—plus they are typically done through sampling, not on every loan file.
Automated audit rules can quickly and efficiently check data for all loans to be included in your LAR submission. Tests can check for things like the accuracy of the rate spread or geocode data from a lender’s LOS, or if a credit score in the LOS differs from source documents—which is more common than you’d think. These discrepancies are identified within seconds and auditors can quickly validate the truth using all data and document resources.
Data transformation is the final process that can be automated. The CFPB’s specific formatting requirements for LAR submission can be codified into rules and applied across each loan file. For example, if the “Action Taken” on a particular loan was “Withdrawn,” the LAR requires the credit score value be transformed to reflect “8888” for “Not Applicable.” This can all be automated as a final preparation step for LAR generation, enabling lenders to quickly change data fields and formats before generating their final submission file.
Some lenders that already have high confidence in their LOS system data can also use these same HMDA auditing tools separately from document processing technologies, with the same advantages of significant automation and minimal human operator involvement. Another alternative is to incorporate documents in a subset of loan file reviews as a quality check. This could be done similar to performing QC on a percentage of the loans or could be done randomly throughout a year.
But regardless of whether you want a comprehensive view of your HMDA data, or rely on a “trusted source” for that data, by using automated rules to validate and transform HMDA data to the CFPB’s requirements, lenders can replace manual processes in these three specific areas of data validation, audit, and data transformation. Comprehensive discrepancy reports enable lenders to see what the rules identified as errors and provide an export to update the system of record—often the loan origination system—to prevent ongoing systemic data issues.
The Value of Complete Transparency
Ensuring complete accuracy for HMDA reporting requires a 360-degree view of the lender’s loan file data, comparing all documents and data in the lender’s LOS and other system data. The only way to efficiently do that is through automation. Technology capabilities make it easier to automatically clear the vast majority of loans that have no errors and generate a focused view of loans with flagged defects, which makes the auditor’s job easier and more efficient.
The benefits of these technologies are tremendous. In our own experience, we’ve seen lenders automate more than 90% of their loan auditing tasks and, when required, perform a complete file review in less than five minutes. Not only are they lowering costs by focusing their teams only on tasks that truly need human review, but they are gaining greater confidence in their HMDA data and reports as well. This verified, validated data can also be used to make business process improvements by leveraging it for use in other applications, generating valuable insights into a lender’s denial patterns, approval rate trends, pricing strategies, loan programs, and more.
For cost reasons alone—including costs associated with greater scrutiny and possible enforcement from the CFPB—ultimately, lenders will have no choice but to use automation for HMDA reporting. But it’s not just about saving money. It’s about knowing you’re doing business the right way, gaining better business insights, and creating happier team members, who are now able to work more efficiently. In other words, in an age of increasing standards for compliance, it’s all the things that make good lenders great.
(Views expressed in this article do not necessarily reflect policy of the Mortgage Bankers Association, nor do they connote an MBA endorsement of a specific company, product or service. MBA NewsLink welcomes your submissions. Inquiries can be sent to Mike Sorohan, editor, at firstname.lastname@example.org; or Michael Tucker, editorial manager, at email@example.com.)