Using data science to make a difference
Data Science Senior Manager Carlos Plá feels a sense of duty to help reshape a housing system that historically has left many underserved.
"Institutional racism in housing has a long history," said Carlos, who leads the Single-Family Analytics team. "Addressing that is a special responsibility that all of us who work here must share. I see helping to increase equitable access to credit as only one part of a broad mosaic, but it’s one part that I can influence, and I’m motivated to make a difference wherever I can."
Making a difference is what his team set out to do. Carlos’ team rose to the challenge of finding a way to help lenders qualify renters for a mortgage using their rental payment history, and a way to help borrowers without a credit score qualify for a mortgage based on their positive cash management habits.
Using data to identify possible first-time homebuyers
Simran Batra, a Data Science Senior Associate, spent months combing through and analyzing bank statement data to determine how to identify consistent rent payment transactions. She used sampling, text analysis, natural language processing, time series analysis, statistics and machine learning to make sense of this novel data source.
After many iterations, she and her teammates finally landed on a predictive model for identifying rent payments, the key enabler to help lenders qualify borrowers based on their history of on-time rent payments.
“It was exciting that we were doing something to help first-time homebuyers with thin credit,” Simran said. “But there was still that gap of people who don’t have a credit score. We wanted to address that gap.”
Unlocking opportunities for individuals without credit
Next, Simran and her teammates used this work as a springboard to create a model to analyze a renter’s overall money management activity if they do not have a credit score.
"For years, borrowers without credit scores had few alternatives when applying for a mortgage due to certain credit-based restrictions," Carlos said. "Now, because of the innovation, coupled with other updates to our risk assessment, Fannie Mae is responsibly lifting these restrictions to serve more borrowers."
One piece of the housing equity puzzle
In the U.S., some 26 million adults are “credit invisible,” meaning they have no credit history with any credit reporting companies, and 19 million consumers have credit histories too limited to score*. What’s more, 15% of Black and Latino consumers are credit invisible, while just 9% of White and Asian Americans fall into that category*.
“None of us can fix the whole system, but we can each bring an equity focus to the pieces that we touch,” Carlos said.
Carlos and his team plan to build off these innovations so Fannie Mae can responsibly serve more borrowers and continue improving equitable access to affordable housing.
“I know people are going from not qualifying for a mortgage to becoming homeowners and that’s incredibly fulfilling,” Simran said. “That’s one of the main reasons I joined Fannie Mae – to make an impact on people’s lives.”
At Fannie Mae, we all make an impact. Join our team to help make a difference in equitable access to housing.
*Source: Consumer Financial Protection Bureau