Stat oil

VISITING the homes of poor Africans who want to borrow money helps Finca International, an American microfinance firm, weed out likely deadbeats. If an applicant has an indoor toilet, or household gifts from a relative working abroad, that is a good sign. Interviewing neighbours also helps, says Mike Gama-Lobo, who looks after Finca’s operations in Congo, Malawi, Tanzania, Zambia and Uganda. Such visits work so well that only 1.5% of loans default each year, but they come at a cost: Finca employs more than 1,200 travelling loan officers in these countries.

Hence Mr Gama-Lobo’s interest in using other data sources to calculate creditworthiness. Nine out of ten loan applicants use a mobile phone. With permission from potential borrowers, analysing usage patterns can help reveal those most likely to default. Frequent calls to or from a rich country are a good sign. So are weekday calls to a nearby market town: that suggests commercial activity.

Grabbing whatever data you can makes obvious sense in emerging markets where credit bureaus are underdeveloped. But it works in the rich world, too, where younger people and immigrants often have no credit histories. Bureaus themselves are now using everything from court records and rent payments to utility and phone bills. And a range of start-ups are also busily exploring alternative data.

Some firms piece together scores by analysing applicants’ online social networks. Professional contacts on LinkedIn are especially revealing of an applicant’s “character and capacity” to repay, says Navin Bathija, the founder of Neo, a start-up that assesses the creditworthiness of car-loan applicants. Neo’s software helps determine if applicants’ claimed jobs are real by looking, with permission, at the number and nature of LinkedIn connections to co-workers. It also estimates how quickly laid-off employees will land a new job by rating their contacts at other employers.

As statistics accumulate, algorithms get better at spotting correlations in the data. Applicants who type only in lower-case letters, or entirely in upper case, are less likely to repay loans, other factors being equal, says Douglas Merrill, founder of ZestFinance, an American online lender whose default rate is roughly 40% lower than that of a typical payday lender. Neo’s efforts to improve accuracy include recording borrowers’ Facebook data: Mr Bathija reckons that within a year there will be enough evidence to determine if making racist comments on Facebook is correlated with a lack of creditworthiness.

Facebook data already inform lending decisions at Kreditech, a Hamburg-based start-up that makes small online loans in Germany, Poland and Spain. Applicants are asked to provide access for a limited time to their account on Facebook or another social network. Much is revealed by your friends, says Alexander Graubner-Müller, one of the firm’s founders. An applicant whose friends appear to have well-paid jobs and live in nice neighbourhoods is more likely to secure a loan. An applicant with a friend who has defaulted on a Kreditech loan is more likely to be rejected.

An online bank that opens in America this month will use Facebook data to adjust account holders’ credit-card interest rates. Based in New York, Movenbank will monitor messages on Facebook and cut interest rates for those who talk up the bank to friends. If any join, the referrer’s interest rate will drop further. Rates and fees will also drop if account holders spend prudently. Efforts to define customers “in a richer, deeper fashion” might eventually include raising rates for heavy gamblers, says Brett King, Movenbank’s founder.

Perhaps no company has gone as far as Lenddo, a Hong Kong start-up that owns online lenders in Colombia and the Philippines. Loan-seekers ask Facebook friends to vouch for them. To determine if those who say “yes” are real friends rather than mere Facebook contacts, Lenddo’s software checks messages for shared slang or wording that suggests affinity. What’s more, the credit scores of those who have vouched for a borrower are damaged if he or she fails to repay. Put the word out about this “social-enforcement mechanism” and “boom, the money shows up,” says Jeff Stewart, Lenddo’s boss.

Tweet and sour

Many “digital natives” who came of age with the web will care less about such loan terms than older folk. Andrea Higuera, a 21-year-old, three-time Lenddo borrower from Colombia, says that merging her Facebook world and financial life “really doesn’t bother me at all”. A recent graduate and now an office worker in Bogotá, she was more bothered by the bureaucratic rigmarole that preceded a normal bank branch’s refusals of a credit card and approval of a very small loan.

Big banks will tread carefully, nonetheless. Although they monitor social media for marketing, using the data to assess loan applicants is “a dangerous game” that big banks are ducking for now, says Frank Eliason, head of social media for Citibank. He notes that in the past six months some people have dropped Facebook for fee-based networks such as App.net that offer greater privacy. Schufa, a German credit bureau, abandoned plans to mine Facebook, Twitter and LinkedIn after a public backlash last year.

Employees at small banks often search social media or the web for the names of loan applicants, says Jack Vonder Heide of Technology Briefing Centers, a consultancy. A loan officer might deny a loan upon learning that, say, an applicant is getting divorced. But if that process was automated and industrialised, it could turn a big bank into “very juicy fodder” for the press. A bank might run afoul of privacy laws as a result, Mr Vonder Heide says. And it would certainly be vilified on the very social-media sites it was using to make decisions.