Financial Inclusion

Scoring The Unscorable

Creditworthy but unscorable. What gives? Old credit scoring models, for one. VantageScore sat down with PYMNTS to explain how using new credit scoring models can expand the pool of borrowers without increasing the risk to the lender.


Creditworthy, yet unscorable.

At least that’s the situation that arises when using legacy credit scoring methods, which were devised decades ago when life, credit access, the economy and consumer behavior were all quite different.

Unfortunately for both borrowers and lenders, those conventional scoring models ignore about 30 million to 35 million U.S. consumers  – nearly a third of whom are prime or near prime, with credit scores of 620 or better.

How’s this possible?

“Traditional” credit scoring models only assign scores to consumers who have used credit and/or had an account update in the past six months. That means those models can ignore consumers who use credit infrequently (a sizable slice of the population in the wake of the recent recession) and newcomers to the credit marketplace.

While many of these “unscorable” consumers are actually highly creditworthy, their inability to get credit scores shuts them out of “traditional” credit channels or forces them to pay more than they should for the credit they get.

So what gives?

The old credit scoring models, for one. Legacy scoring models can’t accurately assess older tradeline data to suss out whether or not a gap in credit usage is due to a consumer simply not wanting to use credit, despite being able to repay. Including this older tradeline data also helps prevent excluding new entrants who are also creditworthy, but perhaps have not used credit for a long period of time.

That scenario – creditworthy but unscorable – was the genesis for the development of  VantageScore’s new credit model. The goal was to be as inclusive as possible in setting the criteria for whom it could score, while still providing the most accurate measure of consumer credit risk possible – protecting borrowers and lenders alike.

The VantageScore 3.0 model recognizes the predictive value of data more than six months old and in some cases will accept data that is more than 24 months old, and it also can provide a score after the first tradeline update. Incorporating this data allows the VantageScore model to account for new entrants in the credit market, and those who may use credit infrequently.

The objective was not to lower the bar — but to widen the window of opportunity – for consumers and lenders.

Not surprisingly, widening that window opens the door to critics who question the accuracy of this new approach to assessing the creditworthiness of consumers. But this modern approach to scoring today’s modern consumers is working — and the proof is in the numbers.

Based on the annual testing of its scoring models, which assess that the model performance aligns with expectations, VantageScore found that in the score range of 600 to 850, its model outperformed its benchmark by a double-digit improvement in mortgage performance as measured by a Gini coefficient – a common standard of measuring predictiveness. A Gini coefficient of 45 or more indicates effective rank ordering by industry standards, and VantageScore reported that the coefficient on its validation sample of newly scored consumers was an impressive 52.3. In addition, newly scored consumers demonstrated nearly identical first-payment default rates as those that are scored using traditional credit scoring methods.

But ultimately, the biggest testament to scoring these creditworthy but unscorable consumers is the fact that nearly 2,000 heavily regulated lenders have tested it and now use VantageScore to assess risk and make their own lending decisions.

Clearly, there is goodness all around. VantageScore’s new models end the punishing and vicious cycle of not extending credit to the highly creditworthy people who need credit – and want to build up their credit scores – but can’t get it or who do not want to always have lines of credit open. New models that address the new realities of consumers and their relationship with credit mean that these consumers not only get credit, but don’t have to pay more for the loans that they are given.

And for the lender, they are able to serve a consumer who is no less the profile of a “mainstream” borrower than any other prime credit customer.

Talk about a win-win-win.


New PYMNTS Report: The CFO’s Guide To Digitizing B2B Payments – August 2020 

The CFO’s Guide To Digitizing B2B Payments, a PYMNTS and Comdata collaboration, examines how companies are updating their AP approaches to protect their cash flows, support their vendors and enable their financial departments to operate remotely.