By James Rufus Koren
Los Angeles Times.
All it takes is a keystroke, maybe two, to hurt your chances of borrowing money from online lender Basix.
Like a growing number of personal and small-business lenders, Basix looks at much more than your financial history when determining whether you’re likely to repay a loan.
Among thousands of factors is whether you type your name with proper capitalization or in all capital letters.
“If you fill in your name in all caps, you’re a much higher risk,” said Douglas Merrill, founder and chief executive of ZestFinance, the Hollywood parent company of Basix.
If that sounds absurd, consider the motto in big block letters on ZestFinance’s website: “All data is credit data.”
It’s a philosophy that a growing number of lenders and credit-scoring firms are living by as they use more and more data, much of it unrelated to money, to augment traditional underwriting practices.
Alternative types of credit scoring have been around for years, taking into account factors not always captured by traditional scores, such as whether borrowers pay their cellphone bills promptly. They’ve been used to assess the creditworthiness of consumers with little or no traditional credit history, including those in developing countries.
Now, a new generation of startups has developed scoring models that look at such things as what a borrower studied in college and how a restaurant rates on Yelp, and they’re using them more widely with businesses and individual borrowers with middling or even good credit.
The abundance of digital information, and the rapidly growing storage and computing power able to comb through it all, is changing industries including agriculture and health care. It should come as no surprise, then, that it’s shaking up consumer finance.
Proponents say these new methods should help borrowers get credit, just as it has helped farmers increase their yields.
“In banking, it’s inconceivable that in the future we’ll be making financial decisions in the way we do today. We’re making decisions about people based on less than 5 percent of the information about them,” said Asim Khwaja, a professor of international finance and development at the Harvard Kennedy School who has studied alternative credit scoring in the developing world. “There’s a lot of excitement in this field.”
But there’s also plenty of skepticism.
Chi Chi Wu, an attorney with the National Consumer Law Center, said traditional credit scores are a known quantity, having been used for decades. What’s more, traditional credit-scoring firms offer tips on how consumers can improve their credit scores.
New underwriting models, though, have little to no track record, she said.
“How have they been tested to show they’re predictive and relevant?” Wu said. “What if a data point is your astrological sign? Does that mean anything about creditworthiness?”
And then there’s the question of whether it’s ethical to use certain kinds of information, such as information pulled from a Facebook account.
That’s a question that blew up last summer after the social network secured a patent for a program that would assess a borrower’s creditworthiness based on the credit scores of Facebook friends. Facebook declined to comment.
ZestFinance doesn’t use social media data, not because it wouldn’t be useful but because Merrill is uncomfortable with it. And he thinks customers would feel the same way.
“We take a lot of time to make sure there’s nothing being used that we find personally creepy,” he said.
Still, these new companies, founded by a mix of tech and finance industry veterans and backed by venture capital, say their models work, and that they should provide not only more access to credit, but also better terms.
Their systems crunch numbers in different ways, but they all have the same goal: “We try to re-create the holistic view of the borrower,” said Merrill, a former Google executive and Rand Corp. researcher.
Another way to think about it: Lenders are trying to use vast troves of data, financial and otherwise, as a modern replacement for the gut feeling a banker might have had about a potential borrower a generation or two ago.
ZestFinance collects thousands of pieces of consumer information, some submitted in an online application, some obtained from data brokers, and runs them through algorithms that judge how likely it is a borrower will repay.
Merrill acknowledges that in many cases, there’s no explanation for why a particular data point helps or hurts a credit score.
For example, borrowers who write in all-caps are riskier, the firm’s credit scoring system discovered after underwriting thousands of loans.
“We don’t know why. It just is,” Merrill said.
At other firms, the connections between nonfinancial information and creditworthiness are a bit clearer.
Social Finance, which goes by SoFi, was founded in 2011 by four entrepreneurs with backgrounds in finance, software and business consulting.
SoFi looks at how long prospective borrowers have been working in their professions and what they studied in college.
That information points to how likely it is borrowers will remain employed, or find a new job if they lose their current one, said Teresa Jackson, SoFi’s vice president of credit.
“I would venture to say someone who is a surgeon is more likely to become re-employed than someone who got a degree in art,” she said.
SoFi borrower Alfonso Brigham earned a bachelor’s degree in business administration from the University of Southern California in 2005. Those credentials and his job at a major financial institution helped him quality for a $711,000 mortgage on a one-bedroom condo in San Francisco’s Nob Hill neighborhood.
Brigham said he went to several banks but was told he’d need to come up with a 20 percent down payment, which would have amounted to nearly $160,000.
SoFi let him put just 10 percent down without having to pay for pricey mortgage insurance. He closed escrow Dec. 1.
Brigham, who underwrites loans for a living, said he can see how the nonfinancial information may have burnished his creditworthiness.
“I’m working in the financial district, which is a big part of the economy here. If I were going to look at a customer [like me], at the end of the day, I’d say, ‘This guy is employable,'” he said. “If someone’s majoring in Russian literature, what can they really do with that?”
Businesses also get judged on factors beyond their finances.
Montreal online business lender IOU Financial gleans tidbits from social media sites.
Phil Marleau, the publicly traded company’s chief executive, said terrible online reviews could be a sign of trouble, even if a business has a healthy profit-and-loss statement.
“At the end of the day, we’re really trying to judge the reputation of the business … by looking at what the crowd is saying,” he said.
Even old-line credit bureaus are considering tapping the new sources of information.
Experian is examining social media data to provide credit scores for businesses too young or small to have a credit history, said Eric Haller, executive vice president of research unit Experian Data Labs in San Diego.
The firm’s data scientists took business credit information and combined it with information from Twitter, Facebook, Yelp and others. Based on that analysis, the firm is working on a credit-scoring system that could be based solely on social media information.
“This is kind of the earliest picture you can get of somebody’s success,” Haller said. “If you’re a consumer-facing business, you’re going to have some type of digital footprint. You might not have anything else yet.”