[Editor’s note: This is a guest post from Avant. Avant is a platinum sponsor and will be in attendance at LendIt USA 2015 on April 13-15. In this post, they talk about financial technology and the middle class.]
In 2008, the financial crisis deflated the average American’s options for credit. The mid-range consumer, the 40 percent of Americans in the middle, were hit the hardest as banks and other lenders started to only serve the sub-prime and super-prime markets. A barbell shaped model emerged.
Little was understood about the customers left in the middle who were making $35,000-$75,000 each year. They were a difficult segment of the market to serve because FICO® Scores didn’t accurately predict their creditworthiness or risk of default. The middle class’s need for credit was unaddressed and represented a huge opportunity within the market that was ready for innovation.
Lenders are already starting to better serve this group. According to The Wall Street Journal, personal loans to middle-class borrowers rose to $26.5 Billion, numbers that are still below the pre-2008 numbers, but are on the rise. Data and technology are leading the way to lowering the cost and barriers of borrowing for these customers.
The widely used underwriting models, like the FICO® Score, are updated every 3-4 years. Innovative, newer underwriting models are coupled with technology that provides easier deployment and can be updated every week. Lenders like Avant started to innovate after seeing the weakness of the older models, creating technology and systems that better predicted the repayment and the default behaviors of this middle class group.
At Avant, we’re using models that are only starting to see the light of day: machine learning algorithms.
Avant’s lead Data Scientist, Tong Lu said, “Our business model ties really closely to machine learning algorithms. It requires a lot of ingenuity and careful implementation, but our focus makes it doable.”
We’re not the only lenders moving to a more aggregated data model. According to a report by McKinsey and Company, emerging lenders have access to more data to make more accurate credit decisions. The report said that lenders are “using increased computing power and new sources of information and data (including mobile-phone usage patterns, utility-bill payment history, and others) to build better risk models.”
But the underwriting model isn’t the only place lenders are focusing innovation efforts for the middle-class. All customers applying for personal loans have to prove that they have a source of income, and the standard method of verification involves faxing paystubs or bank statements. We already know that our customers have jobs and lives that they have to attend to, so the loan process should be respectful of that. Avant is deploying new technologies where no faxing of any sort is required.
Another innovation that underpins these trends is bringing the entire application process online. Removing the traditional “visit to the loan officer” presents its challenges for identity verification, but it’s also is ripe with opportunities. Companies that can excel at verification online can remove the risk of bias that the human interaction between a bank officer and an applicant presents. By removing human bias, the models can identify the consumers true likelihood of repayment based on nothing more than his/her past credit history and financial information.
There’s still a lot of work to be done to serve the middle class, but that’s what makes this such an exciting time to be in the lending industry.
