The act of borrowing and lending traces its roots back to ancient Mesopotamia, but Upstart Holdings Inc. says it’s bringing a new approach to the practice through artificial intelligence—and seeing growing reception in the process.
The financial-technology company uses artificial intelligence to inform lending decisions as it crunches more than 1,000 data points through machine-learning models. In Upstart’s
view, the tech-focused strategy helps expand credit access at better rates without bringing on additional risk, because it can account for a broader set of factors beyond what’s captured by traditional credit scores.
While the company doesn’t often explicitly talk about artificial intelligence when appealing to consumers, it does try to emphasize that people are “more than just [their] credit scores,” Chief Executive Dave Girouard told MarketWatch. “Even people with high credit scores think being represented with a three-digit number doesn’t feel right.”
When dealing with banks and regulators, Girouard said Upstart is more deliberate in calling out that artificial intelligence underlies its risk models: “They care about how we’re doing what we’re doing.”
The larger financial community may be gradually coming around to new ways of thinking about credit. Upstart disclosed on its latest earnings call that seven lenders on its platform no longer require minimum FICO scores, a count that’s been rising in recent months.
Upstart continues to build traction in personal lending, helping propel the company to 252% revenue growth in its most recent quarter, and Girouard sees opportunity in other areas of the credit landscape as well. The company is in the beginning stages of building out its auto-lending business and now estimates that it could see about $1.5 billion in auto transaction volume during 2022, according to the forecast it delivered late Tuesday.
Upstart shares gained 36% Wednesday, following that last report, though they’re off about 5% in Thursday’s session.
Auto lending is a more complicated area than personal lending, according to Girouard, but because of that, the company faces less competition. Girouard also sees a significantly larger market opportunity in auto lending than in personal loans.
The company is involved in auto refinancing and expects that it can be successful by putting a digital spin on parts of refinancing that may have otherwise stopped people from going through with the process altogether. These include pain points such as seeking to get a title adjusted, which can require a trip to the Department of Motor Vehicles.
“We’ve built a program that automates all of that,” he said.
Beyond personal loans and auto loans, Upstart has ambitions in areas like small-business lending and mortgages. It’s aiming to be in the small-business lending market later this year and to enter the mortgage-lending market by 2023.
Artificial intelligence “is one of the most important technologies emerging in history,” Girouard told MarketWatch, and Upstart is “at the forefront of applying it to this 5,000-year-old problem of lending.”
That upbeat view on the potential for AI-driven lending is one reason Upstart announced earlier this week that it has received authorization from its board to buy back up to $400 million of its stock, in a move viewed as somewhat surprising for a rapidly growing fintech name. The company cited the recent volatility in its share price, which has come down 64% from its October highs, despite what it views as an attractive long-term opportunity.
“Because we’re quite profitable, we have the luxury of deciding the best use of capital at any given time,” Girouard told MarketWatch. Though the authorization isn’t a guarantee that Upstart will end up repurchasing $400 million in stock, he said that “at the right price it could make sense” for the company to conduct share buybacks “if the market doesn’t appreciate what we’re doing.”
The move also “flexes that we’re a profitable business” when other fintech companies, in his view, are “burning through cash” trying to reach a breakeven point.
Upstart logged net income of $59 million in its most recent quarter, up from $1 million a year earlier.
Though Upstart is seeing strong growth, its shares have come under pressure recently, partly due to a selloff in fast-growing names. There are also some concerns about future delinquency trends, though Upstart doesn’t expect to see a profound negative impact on its business.
Chief Financial Officer Sanjay Datta noted on Upstart’s earnings call that while defaults were “at an unnaturally suppressed level for more than a year,” trends are starting to normalize due to the waning of pandemic-driven stimulus benefits.
The company and its partners have been “anticipating this shift and as the loans on our platform have been priced accordingly, we are not expecting any meaningful adverse impact from rising defaults on our volumes or economics,” he continued.
The commentary sparked mixed reactions on Wall Street.
Upstart’s latest quarter “defied normalization worries,” wrote Citi Research analyst Peter Christiansen, who cited the company’s strong fourth-quarter earnings beat and its “confident” outlook for the full year that easily exceeded his own expectations.
But while Wedbush’s David Chiaverini acknowledged that rising delinquency rates haven’t stood in the way of Upstart’s growth yet, he wrote that he was still “concerned about the company’s funding profile if the deteriorating trajectory of its delinquency rates don’t normalize in coming months.”
Upstart’s Girouard is optimistic that his company’s AI approach will enable it to hold up better than the competition in times of economic stress.
“If suddenly there’s a huge rise in unemployment, you should expect loss rates to go up in any system,” he said. “But AI can handle it much more adeptly” he said, noting that the AI system is “responsive.”