Did you know that some companies are currently considering using data such as who you’re friends with on Facebook to decide whether you’re a good credit risk? With the rise of the financial technology sector, savvy companies are gathering information about borrowers and they’re able to analyze these large amounts of data using algorithms which can determine which unconventional factors might make you a good credit risk.
It’s all part of the trend across all industries towards using big quantities of data to help inform and improve how businesses work. Scientists, for example are using big data to help better understand and predict things like the weather, and baseball franchises use big data to determine which players might be undervalued – as seen in the film Moneyball.
This article will look at the impact that big data could have on the financial industry and particularly on your credit score. To help paint a picture of that data driven future, I interviewed Kevin Sandhu, the CEO of an online lending and financial technology company, Grow (formerly Grouplend).
Also read: How to Read Your Credit Bureau Report>
The Brave New Credit World
While it might sound off-putting to hear that companies are gathering data on you, it’s nothing new. “In many ways, credit scores are the original big data,” said Sandhu.
The problem with credit scores these days is that they’re not as reliable at predicting someone’s creditworthiness as they might have been in the past. “Take, for example, the 30-year old living in Toronto or Vancouver where housing prices are exorbitant,” Sandhu explains. “He or she is financially responsible and has never missed a rental payment. That person’s credit score does not reflect this positive behaviour however, because rental payments — unlike mortgage payments — aren’t reported back to the credit bureaus.”
Also read: Is Your Credit Score Mortgage Ready?>
Big data has the potential to include other information like a person’s cash flow or rental payments that could provide a different view of their financial reality. By supplementing credit scores with this information, this could result in a lower interest payment for those with low credit scores.
What Could Facebook Friends Say About a Borrower?
Using Facebook friends as a way to determine credit worthiness isn’t something that will be happening any time soon. There’s still a lot of data to collect to determine what impact something like that might have on your credit worthiness – if there is an impact at all.
Sean O’Connor, Grow’s VP of Partnerships, believes, however, that they could find a correlation between a person’s number of Facebook friends and their creditworthiness.
“He thinks that constant exposure to a wide network of people’s self-curated moments of happiness, which are often related to consumption of consumer goods,” Sandhu said, “will result in a higher level of social envy, and, in turn, lifestyle inflation.”
But Sandhu disagrees with O’Connor on the potential impact of one’s Facebook friends.
“To me, the more likely predictor of creditworthiness from your collection of Facebook friends is not how many; rather it’s who. If you are friends with people who tend to be more creditworthy,” he said, “the hypothesis is that you yourself will be more creditworthy, because your friends are a primary point of influence.”
What Other Factors Could Predict Credit Worthiness?
One of the reasons why big data is suddenly able to transform industries is because technology has finally gotten to the point where computers can learn and find patterns without a programmer telling them to specifically look for them. Humans are often not asking the right questions.
But machine learning algorimths are able to detect differences and patterns that a person might not have thought to consider.
“This is the best part of employing machine learning algorithms and other big data techniques to solve problems that relate to credit,” Sandhu said. “While it’s fun to guess at the conclusions the data will form, the biggest drivers will likely come from answers to hypotheses that we haven’t thought of yet.”
“Our job today is to collect the data that will eventually form these answers, even if we don’t know quite yet what the questions are that we should be asking.”
When Will Big Data Be Used To Determine Creditworthiness?
It’s already doing so in some initial important ways. Fintech companies are using additional information to decide who to lend to and what to charge them in interest.
“At Grow,” Sandhu said, “we look at myriad data points beyond traditional credit metrics that determine whether or not we will lend to each individual applicant. We look at demographic, psychographic, and behavioural data; we pull in data from social media; we look at a person’s cash flow.”
For companies that operate primarily online, this information is both to prevent fraud and to determine credit risk.
“While the number of friends you have on Facebook doesn’t necessarily affect your interest rate today, that’s not to say that it won’t in the future. One of our favourite quotes at Grow is, ‘all data is credit data — we just don’t know how to use it yet.’”
As companies like Grow continue to collect data from their customers and start to see statistically relevant patterns, expect this information to gradually start changing the way that they lend.