Maybe it’s a sign that our industry is finally coming to grips with the TILA-RESPA Integrated Disclosure (TRID) or “Know Before you Owe” rule because the Mortgage Bankers Association National Technology in Mortgage Banking Conference & Expo in Los Angeles this week wasn’t all TRID all of the time. A significant number of the sessions and hallway discussions focused on the next shoe to drop: the coming Home Mortgage Disclosure Act (HMDA) reporting rules.
Leading legal and compliance experts at several different HMDA sessions warned attendees that the amount of data and the unprecedented level of transparency that it will give regulators pose heightened compliance risks for banks and mortgage lenders. Even though the effective date of the new rules is still more than a year and a half away, the expert consensus at the conference was that it’s not too early to start developing comprehensive HMDA compliance and reporting solutions.
What’s Changing and What’s at Stake
On Tuesday I participated on a panel entitled: “Preparing for Fair Lending & CRA Examinations.” It was moderated by Ken Markison, the MBA’s Regulatory Counsel, and included senior attorneys from three leading law firms, as well as Joshua Weinberg from First Choice Loan Services.
In that session, legal experts discussed how regulators and enforcement agencies currently use HMDA to identify Fair Lending violations and Community Reinvestment Act issues. Melanie Brody, a partner at Mayer Brown noted that today Fair Lending actions are often based on a comparison of a financial institution’s lending practices versus their competitors or peers in a specific market. Recent Fair Lending settlements, she said, “tend to turn on statistics.”
Brody then went on to discuss how this kind of analysis was the basis of several recent high-visibility settlements. In each of the cases, the smoking gun was the level of lending to, or the number of loan applications from, specific minority groups versus the bank’s peers. But the analysis didn’t really drill down into underwriting and pricing decisions and the scope of the review was usually at the census track or MSA level, not the property or zip code level.
Kathleen Ryan from Buckley Sandler explained how the new data requirements will give regulators the ability to pursue redlining and CRA examinations on a more granular level. She began her presentation by quoting from the FFIEC’s press release that accompanied last year’s HMDA report. It stated: “The HMDA data alone cannot be used to determine whether a lender is complying with fair lending laws. They do not include many potential determinants of loan application and pricing decisions, such as the applicant’s credit history, the debt-to-income ratio, the loan-to-value ratio, and others.”
But that will change under the new rules. Lenders will soon have to provide all of those missing elements in the new HMDA data set as well as interest rate and spread, discount points, and origination fees and lender credits.
Similarly, the LAR filings will now include property addresses, not just census track information, as well as the loan officer’s NMLS ID number. These expanded data reporting requirements mean that lenders need to be connecting the dots between what information is being submitted and the possibility of a Fair Lending review.
Today, she said, HMDA and CRA examinations focus on the census track data; tomorrow they will be able to make property-by-property comparisons.
Similarly, examinations currently tend to look at retail branch and storefront operations. But in the future, they will be able to compare wholesale to retail channels and even track individual LO performance.
From a CRA perspective, regulators will be able to see whether banks are buying loans to boost their community reinvestment ratings.
They will also be able examine what type of lending is taking place within the bank’s assessment area, and at what price. In the past, examiners were investigating to see whether a bank was “lending or not” to a specific group or within a certain market. Now they will be able to see “how you are lending and pricing,” Ryan said. And it won’t require an examination to get this information: it will all be on the regulators’ desktops.
The stakes are high for Fair Lending violations. They can include costly settlements, class-action lawsuits, and reputational damage. An important point to remember is that Fair Lending is both an art and a science: interpreting the data, including building the statistical model, is the art, and preparing the data to build the model is the science.
An inadequate CRA rating can put banks into the penalty box in terms of mergers and acquisitions as well as other applications, according to Joseph Lynyak, a partner at Dorsey & Whitney.
Brody added that the Consumer Financial Protection Bureau has also been steadily sending the message that they are concerned about data integrity. The Bureau has lowered the threshold of errors that it will tolerate before it requires a lender to resubmit its LARs and levies civil penalties.
My fellow panelists agreed that having accurate, easily accessible data will make the LAR reporting more efficient (read: less painful and expensive) and will support better outcomes in the event of exams. “Data drives effective response to bank regulator or CFPB CRA and Fair Lending examinations,” said Lynyak.
Self-Assessments and Technology
Several panelists, including myself, recommended that lenders re-think their approach to HMDA reporting process, making it an ongoing activity, rather than a once-a-year event.
In my presentation, I compared the risks of trying to comply with the new HMDA rules without technology and a new approach to data collecting and review to a high-stakes game of “Heads Up!” In that game, a player holds his or her cellphone up to their head with the screen facing the audience, so the audience can see what the contestant can’t.
Lenders submitting LARs without the time and/or the tools to do a self-assessment of the HMDA data first can’t see what they are delivering… but regulators and the plaintiffs’ bar can.
Self-assessment on a monthly or at least quarterly basis will help lenders know where they stand on fair lending compliance and identify issues as they’re developing, not after the fact when it’s too late to act. This ongoing monitoring can also help catch marketing programs and LO behavior that might later be deemed questionable, or what’s worse, evidence of disparate impact.
If a lender does uncover issues, being proactive in bringing them to regulators’ attention may mitigate the consequences or at least improve their negotiating position.
Technology will be critical in enabling lenders to first capture all of the new data elements that will be required for LARs and then to analyze them. Keep in mind that the CFPB wants data on every loan and application—not just those that close, were denied, canceled or withdrawn (called orphaned loans).
Tracking “orphaned” applications months down the road can be challenging, particularly if they are coming through third-party originator channels.
Weinberg, the final panelist, noted that his mortgage company, which is affiliated with a bank, is regulated by CFPB, FDIC, and various state regulators. He then went on to explain how the company has developed a comprehensive compliance management / internal and external auditing program to monitor Fair Lending risk.
His advice to lenders: “If you can’t explain your data or show your work, only a miracle will save you” in the event of a Fair Lending exam. My advice? Know (the implications of your data) before you submit.