Utilities, telecommunications and rent payment data ought to be used in mortgage underwriting, especially since federal regulators say they are focused on improving racial equity and may update their credit scoring models.

So argued a report released Tuesday by Michael Stegman, a non-resident fellow at the Urban Institute and former senior policy advisor on housing in the Obama White House, and Kelly Thompson Cochran, deputy director of FinRegLab. But, the authors acknowledge that obstacles abound in changing an underwriting process dictated in part by Fannie Mae and Freddie Mac.

Including rental data is a crucial first step from a racial equity standpoint, Stegman and Cochran declare, because renters are disproportionately people of color and it is the payment most analogous to a mortgage. However: collecting rent payment data is hard since there are 10 million smaller landlords, which make up 44% of the U.S. rental market, the report notes.

“It’s much more complicated to gather that data, rather than utility or telecommunications [data], because both of those markets are more consolidated,” said Thompson Cochran. “Thinking through how to build and subsidize that infrastructure, to potentially reach more consumers, is really important.”

Also, standardizing rental payment data means cooperation from numerous stakeholders, the authors state.

---

Sponsored Content

---

Stegman said that federal agencies such as the Federal Housing Finance Agency (FHFA) could facilitate the transition by providing funding or clarity on regulations. That’s not quite happening right now, though Fannie Mae has taken an initial step to include rent payment history.

“There is no grand strategy and there is no grand plan,” said Stegman. “But certainly there needs to be coordination across a range of regulatory and executive agencies.”

Rental history was a centerpiece of mortgage underwriting, before automation took over in the 1990s. It was then that the government-sponsored enterprises created the architecture for their now-ubiquitous automated underwriting systems — and those credit models didn’t include rental payments.

It is those automated underwriting systems that undergird credit decisions across the GSEs’ whopping $7.2 trillion mortgage portfolio. They’re baked into the mortgage securitization process, too. The secondary market prizes consistent data across lenders and portfolios.

So, even if FHFA approved alternative credit scoring models, the report’s authors expect the switch would take a long time to implement. The FHFA, which oversees Fannie Mae and Freddie Mac, launched an effort in 2017 to study the impact of adopting an alternative credit scoring model. Four years later and the study is still ongoing, though FHFA expects to complete it by early 2022.

The FHFA has taken more action toward alternative credit scoring than the other federal agency that supports the mortgage market, the Federal Housing Administration.

Unlike the FHFA, the FHA, which provides crucial support for first-time homebuyers and borrowers of color, hasn’t yet announced its intention to use a different credit scoring model, although Congress has sought to push the FHA in that direction for more than a decade.

In 2009, Congress tasked the FHA with creating a pilot program to test out using alternative data such as rental history. But the authorization expired in 2013 without the FHA taking any action. Legislation proposed in 2019 to restart the process failed to advance from the House of Representatives.

Earlier this year, advocates asked the Biden administration to direct the Department of Housing and Urban Development to use existing programs — many of which serve low-income borrowers and borrowers of color — and its relationships with subsidized rental housing providers to report rental data. But the path forward for those initiatives is unclear.

“There are 2 million families that receive rental assistance through vouchers, where they pay part of the rent and HUD delivers the rest, and that happens seamlessly,” Stegman said. “Extending rent-reporting to on-time payments to landlords that are a part of the voucher program is certainly something that should be up for discussion, along with a number of other rental assistance and self-sufficiency programs that HUD runs.”

There are steps by federal regulators that don’t need Congress’s go-ahead, and some are underway. Fannie Mae earlier this year said it would include positive rental payment history in its underwriting process.

But Fannie Mae’s plan depends on additional cooperation from both the lender and the borrower. For loans that Fannie Mae’s automated underwriting system rejects, the GSE now checks to see whether 12 months of positive rental payment history would make the loan eligible. If so, Fannie Mae alerts the lender. Then, the lender can, if they wish, ask the prospective borrower for permission to share their bank account information with the GSE.

Freddie Mac has taken a different approach. It hopes to encourage landlords to provide access to rental payment data, by recouping a portion of closing costs for properties it finances. In exchange, the landlord would use a platform that reports on-time rent payments to the credit bureaus. But sharing the data also hinges on consent from the prospective borrower.

The report’s authors say that they hope to take advantage of a new sense of urgency to close the racial homeownership gap.

“It’s not going to happen overnight,” said Stegman. “But we hope to bring enough information from these different sources into one place to begin a broader conversation about the importance of using alternative data in credit underwriting.”