Big Data Credit Scores – The Future

Observers have talked about “big data” – the analysis of various metrics to better address consumer needs with various products and services. This task has been made much more feasible in recent years as more and more transactions are made via the internet, where data is collected and stored with relative ease. The abundance of digital information, and the rapidly growing storage and computing power available to comb through it all, is changing industries from agriculture to healthcare. It should come as no surprise, then, that it's shaking up finance as well. 

In the financial services world, this analysis is primarily directed toward credit scoring. At my company RealtyShares, for example, we are looking at ways to move beyond traditional FICO scores to better assess risk and a borrower’s ability to repay. That is the case even as the FICO scoring system itself has begun to make adjustments to its underlying data metrics. But some firms have already gone much further. Over the last few years, for example, many companies have been utilizing Facebook for its huge repository of data, which contains strong indicators of users’ socioeconomic status. The schools you attended, where you work, who your friends are – all of these are understandable, though perhaps initially discomforting to the consumer -- indicators of credit risk.

All of this effort is aimed at developing more efficient markets. One-on-one meetings with borrowers or tenants have an intuitive appeal to those of us who think we can trust our “gut” as to the persons’ credit-worthiness -- but time has shown that the industry prefers more objective data. The FICO scoring system is now widely used for consumer credit and other purposes. New efforts are aimed at further improvements on this approach.

What’s Being Discovered

Some of the early results are surprising. Among thousands of possible factors is whether you type your name with proper capitalization or in all capital letters. “If you fill in your name in all caps, you're a much higher risk,” said Douglas Merrill, founder and chief executive of ZestFinance (which owns Basix). Social networks and cues are also important indicators. “The data we have on customers via social networks says more about them than their FICO,” said Alex Sion, president of New York-based Moven. “You can make credit decisions based not on a faceless score, but on who you know.”

Atlanta-based Kabbage Inc., for example, asks that small businesses seeking loans grant Kabbage access to Amazon, eBay, Xero and other e-commerce or accounting sites to assess those businesses’ creditworthiness. Customers must link at least one such account for underwriting decisions. The company, which has extended more than $1 Billion in loans since launching in May 2011, also may take Facebook, Twitter and other social accounts into consideration when determining whether to increase a loan. In reviewing these social accounts, Kabbage looks at what customers are saying about the borrower's business and the quality of its customer service.

San Francisco lender Social Finance (which does business as SoFi) looks at how long prospective borrowers have been working in their profession and what they studied in college. That information points to how likely it is borrowers will remain employed, or find a new job if they lose their current one. “I would venture to say someone who is a surgeon is more likely to become reemployed than someone who got a degree in art,” said Teresa Jackson, SoFi’s vice president of credit.

The German company Kreditech says that it uses up to 20,000 data points when assessing an application for a loan.

The Times They Are A-Changin’

Businesses who utilize this information for only their own purposes and are not providing it to third parties aren’t under the same regulatory environment as larger credit agencies like Experian or Equifax. “They don't provide it to third parties like a reporting agency does,” said Maneesha Mithal, the associate director of the FTC's division of privacy and identity protection.

And some remain unimpressed with the field. “The accuracy and fairness of big data credit technology is unproven,” said Aaron Ricke, a former lawyer for the Federal Trade Commission.

Yet big data credit scores show promise for segments of the population that are off the radar of credit card companies and the usual providers of credit data. Most current credit scores, for example, don’t take into account ordinary payments to utilities, landlords, and telecom providers -- even though research shows that such payments are helpful in scoring people’s creditworthiness.

If the field develops as hoped, lenders may be able to offer borrowers lower rates – and as rates decrease, more borrowers will come to that lender, driving more data into its proprietary model. It looks likely, then, that more and more companies will incorporate proprietary data to improve upon FICO scores to indicate repayment prospects.

Nav Athwal is the Founder and CEO of RealtyShares.com, an online crowdfunding for real estate marketplace connecting individual and institutional investors to private real estate opportunities nationwide.  Nav started his career as an electrical engineer, before transitioning into real estate law where he worked as a Real Estate and Land Use Attorney at Farella Braun & Martel, LLP. As an attorney, he led some of the largest mixed-use residential, commercial and renewable energy real estate projects in California on behalf of National and International clients including public REIT’s real estate developers, property owners, investors, non-profit housing providers and government agencies.  Nav was named to the LinkedIn Next Wave List a list of the top professional under 35.  He is a frequent contributor to Forbes and has appeared on CNBC, Fox Business and Bloomberg Radio.