With a Little Help: Friends Vouch for Borrowers in New Loan Model

Imagine offering a borrower a better rate or a bigger loan because a handful of his buddies say he is an upstanding guy who they are so confident will pay back his loan that they are willing to cover a portion of it if he doesn't.

It is a concept at least as old as the printing press, but it is finding new life as the business model for a score of fintech startups looking to solve one of the financial industry's persistent issues — how to properly lend to those often left out of the system because of their income, their location or their age. The firms are turning to social relationships as a matter of determining trustworthiness.

"Lending circles, and the idea of people getting together in networks and supporting each other [financially], is centuries old," said Yee Lee, chief executive of Vouch, a lender that relies on a model like that described above. "Typically in these cases there are incredibly high repayment rates, because of the support network."

Vouch offers loans between $500 and $15,000 to individuals and it scores their creditworthiness initially through traditional methods, such as income and FICO score. However, the borrowers then give the company the contact information for as many as five people who will "vouch" for them and that can bring down the rate or increase the amount. The key question for the network is "Do you trust this person with money and, if so, how much?" The company then asks the respondents to put their money where their mouth is by asking them to cover that amount if the borrower doesn't pay. Lee said the average vouched amount is $110.

The company launched in 2014 by lending from its own balance sheet, but late last year it closed its first debt securitization deal with an asset management firm and entered into a partnership with a large hedge fund focused on consumer lending. (Lee declined to name the firms.)

The financial industry has been debating the virtues of using alternative methods for credit scoring for a few years now. Some say it holds promise for the underbanked, while others say the idea of having creditworthiness tied to social networks poses its own risks.

"We see promise in not only in reaching the unbanked, but the more data you have, the more confidence there is in underwriting," said John Thompson, a senior vice president at the Center for Financial Services Innovation. "You also have to be careful about unintended consequences, so we hope these models continue to be tested and refined."

Still, it appears that value in using social and nontraditional data in underwriting is beginning to catch the attention of banks, said Juan Mazzini, a senior analyst with Celent.

"For years they said, 'This is not going to work,' " Mazzini said. "Now some banks are starting to look at partnering with some of these fintech companies to offer an alternative financing to customers."

These partnerships are not likely to be a flash in the pan, in part because it means bringing in new, previously overlooked customers, but also it is likely to please regulators, he added.

"It's difficult sometimes for banks to reach the unbanked and underserved markets with credit products," he said. "And a focus on financial inclusion is also part of the agenda for many governments around the world."

Banks in the United States are behind other countries in partnering with social lending startups largely because of regulatory issues, said Patricia Hines, also an analyst with Celent.

"In the U.S., the OCC oversees the use of credit scoring models by banks," she said. "If a bank wants to revise its credit models they need to undertake a rigorous analytical testing process, comparing the performance of existing models against proposed models over time."

But some U.S. banks, specifically those with a mission of serving the underserved, are exploring alternative lending models. For instance, Spring Bank in the Bronx, N.Y., is working with a firm called Happy Mango to offer loans to low- to moderate-income consumers using alternative credit scoring. Social data is a small part of the overall mix in how Happy Mango judges creditworthiness; it also analyzes spending and saving habits as opposed to just looking at prior loan payment history. This is an effective model to extend credit those that may not have built up a traditional credit history, said Melanie Stern, the $130 million-asset bank's director of consumer lending.

"We were looking to offer a series of loan products to lower-income consumers, and wanted to use a way to evaluate them for a loan without using traditional credit scores," she said.

The partnership with Happy Mango is more than just credit scoring, though. It offers a "self-assessment" tool, where consumers can input information about their financial habits and receive feedback to help them manage their money better.

"So even if they don't get a loan, it still provides them with a sense of where they are financially," Stern said.

Kate Hao, Happy Mango's CEO, said this makes the service different from just a formal loan because it offers individuals "ongoing analysis and insight into what kind of loan product might be right for them."

Using social data also has benefits in small-business lending, said Chris Hale, CEO of Kountable, a San Francisco-based company focused on trade finance by offering loans to business that are young but past the startup phase and are looking to expand their operations.

The company uses an algorithm that creates a "k-score" based on the business owner's social networks. Like Vouch, the social data is not used as the primary factor in extending a loan, but as one step in determining if the person should be seriously considered for a loan, Hale said.

The score is based on an idea called "cave theory," Hale said, which states that a nonentrepreneur has a smaller set of tightknit connections, while entrepreneurs tend to be "cave-spanning connection makers."

"Entrepreneurs usually bridge from one social cluster to another, and have a wide variety of connections," he said. "We look at things like geographic reach, and many other subtle markets that make up the k-score. We use that to determine if this is someone we now want to spend serious time on in evaluating for a loan."

Kountable began offerings loans to business owners in rural Rwanda that were not part of the formal banking system, and Hale said the company is looking to enter underserved markets in the U.S., citing Detroit and St. Louis as two examples. He said Kountable also has "pending relationships" with some banks that could not yet be announced since they are not finalized.

Like Vouch's Lee, Hale said the basis of what Kountable is doing is merely "taking new technology and combining it with some very old finance concepts."

For reprint and licensing requests for this article, click here.
Bank technology Consumer banking Marketplace lending Analytics
MORE FROM AMERICAN BANKER