The digital footprint any mobile user leaves can equal or exceed the information content of credit bureau scores, a team of researchers claims. In other words, it is possible to determine creditworthiness.
The issue is the extent to which mobile service providers or their partners have in their own domain “mobile footprint” markers that might be used to create creditworthiness proxies or other predictive measures that are of value to retailers, advertisers and marketers as well as financial institutions.
That likely is most useful for mobile service providers who have business interests in mobile advertising, banking or marketing support services sold to third parties.
Lending companies, for example, have created apps which analyze information stored on mobile phones including text messages, emails and frequent phone calls.
Micromoney has built a service that builds credit scores for people without bank accounts, for example, using mobile phone behavior, in part. “In an emerging market, a smartphone can tell us everything we need about its owner, so we can estimate his or her credit worthiness,” the company says.
Branch, Lenddo, InVenture, and AvantCredit are other firms that mine phone and mobile internet behavior to make predictions about creditworthiness.
Phone behavior captures many behaviors that have some intuitive link to loan repayment. One study found patterns in how expenses are managed, such as variation (is usage erratic?), slope (is usage growing or shrinking over time?), and periodicity (what are the temporal patterns of usage?). Demographic features (gender and age) have very low correlation with repayment, on the other hand, the study found.
Besides, some of the indicators are social media updates, prepaid top ups, mobile bill payments, location services to track a location of client and driving habits. The question is why mobile firms have not themselves considered ways to create business intelligence that generates revenue from mobile-specific data stores, the way Facebook monetized “knowing who your friends are” or Google monetizes your location or search history.
Or, perhaps, more relevantly, we might ask whether it is possible to create services with predictive value for potential partners relying first on the mobile footprint, and then combining such predictions with other available data stores to create business-relevant predictors of behavior that can be sold as a service to third parties.
Orders from mobile phones (default rate 2.14 percent) are three times as likely to default as orders from desktops (default rate 0.74 percent) and two-and-a-half times as likely to default as orders from tablets (default rate 0.91 percent). Orders from the Android operating systems (default rate 1.79 percent) are almost twice as likely to default as orders from iOS systems (1.07 percent).
One obvious reason for that difference in behavior is that consumers purchasing an iPhone are usually more affluent than consumers purchasing other smartphones.
“For example, the difference in default rates between customers using iOS (Apple) and Android (for example, Samsung) is equivalent to the difference in default rates between a median credit score and the 80th percentile of the credit bureau score,” say researchers Tobias Berg, Frankfurt School of Finance & Management; Valentin Burg, Humboldt University Berlin; Ana Gombović, Frankfurt School of Finance & Management and Manju Puri, Duke University, FDIC, and NBER.
The finding that mobile customer behavior can be a proxy for creditworthiness is among the most-significant new ways mobile user data can be mined for business value. The possible issue is what else might be done.
