Using Mobile Phone Payments To Get People With No Credit History Some Credit

For people who’ve never had a credit card, it’s difficult to access the financial system. That’s why one startup is mining mobile phone payment records to show loan-worthiness.

If you’ve ever been refused store credit even though you have a decent job and plenty of money in the bank, you’ll understand the biggest barrier facing the financially excluded in the developing world. Often what stops people from getting loans to build their businesses or buy assets isn’t their profile as customers. They may be ideal applicants, with regular income and the willingness to repay. Their problem is proving it.


If you’re on the outside the financial system, you don’t have a credit history that a loan officer can refer to; you may not have much documentation at all. In effect, you face a Catch 22. You’re excluded from the financial system because you’re excluded from the financial system.

It’s this problem that Nicole Stubbs has been trying to address with First Access, a New York startup that mines alternative financial data, including mobile phone payment histories.

“Poorer people are not necessarily higher credit risks but, because they can’t prove it, they end up paying much higher interest rates. There’s a price distortion from there not being enough information,” she says. “The big opportunity is that over a billion people now have financial records. They just don’t know it yet and they don’t have access to those records. They’re in the form of mobile phone records.”

Certain types of behavior may offer clues to someone’s credit worthiness. For example, someone topping up their phone regularly at a certain time of the week may be more stable and dependable, even if it’s only a small amount each time. Someone topping up with one large amount, but leaving their account empty for three weeks, may be living a more volatile existence and represent a higher risk. Similarly, if someone is making or receiving international calls, it may indicate that they have friends or family in the diaspora, and are more likely to have access to remittance payments.

“We can see all the behavioral patterns for all the people who always repay on time and people who never do,” Stubbs explains. “Using that, we build models that predict loan performance and give a loan size recommendation for someone who’s never had access to credit before. We benchmark that new person against all the other people who’ve taken out loans.”

First Access helps banks and micro-finance lenders make decisions about applicants within about 90 seconds. It works as follows. Using text, the loan officer makes a request for information. First Access sends a message to the applicant asking for permission to access their data, the applicant approves, and the loan officer gets a reply with a recommended amount. Either the loan is approved immediately, the officer is prompted to get more information, or the loan is refused.


See Stubbs talk more about the concept in this PopTech talk:

First Access started with a pilot in Tanzania, working with Vodacom, a local mobile carrier. It’s since added another carrier, Airtel, as well as several lenders. And it’s also in Kenya. According to Stubbs, the company has helped approve more than 400,000 loans so far.

The service reduces lending costs through better segmentation of customers. “In developing countries, when every customer applies the same way, and with the same evaluation process, it results in higher interest rates for everyone,” she says.

Lenders can approve the least risky customers quickly, minimizing costs for that group. “It means you have risk-adjusted pricing across the whole spectrum of borrowers, so very low risk people are not taking up a ton or your time, and you’re spending more time on medium and higher risk people,” she adds.

First Access wants to expand further in Africa and South America over the next two years. There should be plenty of demand for the idea, though there are other a number of other alternative credit-scoring startups. Omidyar-backed Cignifi does something similar, though its service is in the form of leads for lenders to follow up, not a real-time system for approval.


“We’ll find that the top 50% of borrowers by score will represent under 1% of losses for a bank,” Stubbs says. “That’s key because up to half the people applying for loans could be expedited dispersal or even instant approval and they would have little impact on your losses.”


About the author

Ben Schiller is a New York staff writer for Fast Company. Previously, he edited a European management magazine and was a reporter in San Francisco, Prague, and Brussels.