The Rise of Machine Learning in Lending

Mortgage and Lending with Olympus Labs

To a banker, the most important things about any business are that it has existed longer than two years, it has an excellent credit rating and it has enough working capital left over after meeting expenses to manage emergencies as well as its basic needs. Bankers also care about the personal credit history of the business owners, especially as it relates to their credit score and whether they have filed bankruptcy recently enough to still appear on their credit bureau report. This approach is quickly becoming outdated .

Problems with the Traditional Approach to Business Credit Approval

Some players in the financial industry have concluded that the old way of evaluating applicants for credit may not be the best way. For example, a business owner may be at his or her credit limit on a personal card due to spending the money to launch the business. Bankers are also increasingly realizing that little to no competition exists for the FICO credit scoring model for business applicants. The problem is that increasing numbers of entrepreneurs and the rapidly changing definition of what it means to be in business today don’t necessarily fit into this model.

Credit Review via Machine Learning and Deep Learning

It turns out that artificial intelligence may be better at predicting who will become delinquent on a business debt and who will pay it on time. In April 2018, a paper published by three Brazilian financial researchers explored data from over 711,000 personal credit card holders and determined that just under one percent of them are currently behind on their payments.

What each applicant had in common was that the credit card issuer used deep neural networks rather than standard credit analysis methods when making a lending decision. Although the data was unorganized and complex, it detected high-risk credit applicants far better than human bankers could do.

Back in the United States, bankers are increasingly relying on machine learning and deep learning to save time when processing credit applications. They also depend on these forms of artificial intelligence to help them make the most appropriate decision based on legitimate risk instead of an established set of criteria that may not make sense to apply to everyone.

What Does the Head of Global Sales at Square Think?

Square changed the financial industry by creating new solutions for payment and payroll issues. Michael Coscetta , Head of Global Sales at Square, said this about the relationship between artificial intelligence and business lending:

“What we’re using machine learning and data for are to make new avenues out there, to create new channels that didn’t exist, to offer loans at scale. One of the beauties of what machine learning allows is we have millions of Square merchants. If we were doing this by eyesight and by human beings, we would be able to help only a fraction of a percent of the merchants we currently do. So to me, this is a great example of how you can seek new ceilings of what a business is capable of by applying data in machine learning.”

With companies like Square leading the way, it may not be long until all business lenders adopt the same philosophy.



This entry hasn't been re-blogged:

Re-Blogged By Re-Blogged At
Real Estate Technology & Tools

Post a Comment
Spam prevention
Spam prevention
Post a Comment
Spam prevention

What's the reason you're reporting this blog entry?

Are you sure you want to report this blog entry as spam?


David Jackson, MBA

Financial lending analyst
Ask me a question
Spam prevention