BID® Daily Newsletter
Oct 28, 2019

BID® Daily Newsletter

Oct 28, 2019

Using Alternative Data For Creditworthiness

Summary: When it comes to using alternative data, some financial institutions don't trust it very much, while others are tinkering with it. What you should know.

According to researchers, the average person knows about 600 people. That said, people say they only know 10 to 25 people well enough to actually trust them.
When it comes to using alternative data, some financial institutions don't trust it very much, while others are tinkering around some. For those seeking to serve more customers--especially Gen Zers without much credit history--this can offer an option perhaps.
There are companies whose lending-as-a-service platforms use alternative data, such as education and employment history, which can show borrowers can indeed meet their obligations and be good customers. Machine learning is also used to make credit underwriting and pricing decisions. These platforms' sophisticated algorithms may enable you to experience loss rates below others in the industry and that is the attraction.
One such technology company worked with the CFPB to compare the outcomes of its alternative underwriting model with a more traditional underwriting model, based on information from credit bureaus, without the aid of machine learning.
The tests found that the alternative model approved 27% more applicants for consumer installment loans than the traditional model, and yielded 16% lower average APRs for approved loans. Moreover, credit access was expanded across all of the races, ethnicities and genders that were tested, increasing acceptance rates by 23-29% and decreasing average APRs by 15-17%.
The alternative model was particularly useful in expanding credit to Gen Zers, as the tests showed that applicants under 25Ys of age are 32% more likely to be approved with these models than other ways of underwriting.
Indeed, a number of community banks have begun using a lending-as-a-service platform for that very reason: to attract more "digitally savvy" college graduates just starting out. The platform's algorithms take into account their educational and work history to determine if they have the discipline to pay back loans, as they embark on their careers.
The platform can also be used for things such as determining the creditworthiness for anyone whose credit file may not be perfect or where their work history might be less normal. Auto loans, home equity loans and credit cards can all in theory be underwritten using such an approach.
The opportunity is certainly there given about 26mm Americans are "credit invisible" according to the CFPB. This means they have no credit history with a nationwide consumer reporting agency. Another 19mm or so have a credit history that has gone stale, or is insufficient to produce a credit score under most scoring models.
If your financial institution wants to expand the number and type of customers served, you may want to see how such a platform could work for you. Doing so could help you grow relationships, as these customers build their credit and buy more products.
Subscribe to the BID Daily Newsletter to have it delivered by email daily.

Related Articles:

Pros and Cons of Basing Early-Stage SBA Loans on Gross Income
Borrowing a page from the Paycheck Protection Program, some banks have begun offering SBA loans to early-stage businesses based on their gross income. We consider the pluses and minuses of this underwriting method.
The Importance of Fairness Testing on Artificial Intelligence
As financial institutions increasingly rely on artificial intelligence and machine learning to determine lending worthiness, fairness testing software is becoming an important component in verifying that the algorithms lenders are relying on don’t have their own biases built in.