Could Your Facebook Profile Get You a Mortgage?
Monday, April 6, 2015
Across the globe, there are an estimated 2.5 billion people who are “unbanked” or “underbanked.” This means that almost a third of the world’s population either doesn’t have a bank account or doesn’t have access to bank lenders or credit agencies. This makes it nearly impossible to buy a moped, invest in a business, or pay medical bills or tuition—and, ultimately, to rise out of poverty. Access to credit is as correlated to middle class status as perhaps any indicator.
Which is why microfinance, the banking practice of lending small amounts of money, has become an $81.5 billion industry. But as anyone who has watched the microfinance industry can tell you, it’s far from a perfect solution. Microloans come with high lending rates—on average 35%. One reason rates are so high is because it’s expensive for banks to evaluate lending candidates with thin a credit file or no file at all.
When I first began working with outsourced teams in the Philippines 10 years ago, my young and skilled staff often asked for loans. Although my employees and hundreds of thousands of other Philippine professionals earned middle-class wages, they lacked access to credit services that would allow them to invest in themselves and their families.
But while people may not have traditional assets, like houses or cars, they do have social assets in the form of friends and family connected by social networks and the Internet. My business partner and I saw the proliferation of smartphones and digital social networks as an opportunity—and in early 2011, we founded an online platform called Lenddo to turn this “connectedness” into proof of identity and creditworthiness. As of January 2015, we have hundreds of thousands members worldwide.
The idea that social networks can serve as underwriters relies on the theory that by understanding an individual’s community interactions, one can understand his or her trustworthiness. If this rings true, the next question becomes: How do you make vouchsafing digital? We designed an algorithm that would analyze a person’s social networking behavior—whom they connected with, how often, via what channels—and create a credit score that could predict a person’s repayment behavior. Applying these scores, we were able to screen out those likely to default and to connect trustworthy borrowers to the right amount of capital at the appropriate interest rate for their risk profile, through lending companies that, at first, we owned.