As of July 1, credit score changes are coming that are estimated to impact up to 14 million Americans — and according to many industry experts, the change will be positive. Then, a brand new scoring model is coming this fall that will broaden the types of information considered in consumers’ credit scores — equalizing the playing field even further.
As noted personal finance expert Jean Chatzky explained, scores could jump 25 points or more, which could “translate into hundreds, if not thousands of dollars in savings as you quality for better rates on any money you borrow.”
This is a big deal in part because 56 percent of Americans have a credit score under 680, which is generally considered subprime. A subprime score makes a lot of things more challenging and expensive, like the simple act of getting credit card or renting an apartment. Not to mention getting reasonable rates on a home loan or having access to a rewards credit card.
Often times low credit scores are due to factors like income volatility — the single mother coping with an emergency doctor visit or the rideshare driver who had a few slow nights. An inability to predict income makes it hard to pay bills on time, and as a result, is a major contributor to low credit scores. Another challenge is being new to credit (having a “thin file”) — for example people new to the United States, or young people who haven’t established a credit track record. Simply put, credit scores are important to daily life in the United States.
We think it’s great that the methods for calculating people’s credit scores are being refined, with some unreliable data being removed, new sources of data being considered, and a general recalibration of what it means to be “creditworthy.” If you haven’t heard about the various changes, here are the highlights:
- As of July 1, the credit score formula will begin excluding select public-record information, like civil judgments, medical debts and tax liens. This is because many consumers have complained this data is inaccurate on their files — leaving many with unfairly low credit scores due to administrative errors.
- Starting this fall, a new model will take into account how a person’s debt balance changes on a month-to-month basis, rather than taking a snapshot of one moment in time. Called “trended data,” this new technique allows lenders to see how a consumer’s financial health is trending — vs. a one-dimensional point in time. This gives borrowers a fairer shake — for example, if an unexpected expense dinged their credit but they’ve been steadily working toward getting it back, the new scoring model will reflect that.
- Also starting this fall, the new formula will lean more heavily on new techniques to help score people with minimal credit history, or “thin files.” The formula will use machine-learning techniques, examining many more consumer behaviors to identify who is likely to pay their bills on time. This is good news for those who are just starting to build their credit.
While seen as good news for some consumers, many industry experts have cited concerns; for example, the changes will make it more challenging for lenders to know who to approve and who to decline. And we can see why: much of the banking industry relies heavily on third party credit scores like FICO and Vantage to make their credit decisions. The changes — both removing some information that lenders previously relied on, and adding new information and unfamiliar sources of data — will require financial institutions to tweak their algorithms and realign their models around who to deem creditworthy. For traditional lenders, this is no easy feat.
But for us here at LendUp, we see it differently. Our mission has always been to provide financial solutions to more Americans — those traditional financial institutions overlook because of poor credit scores — by using innovative underwriting models. We already look at the bigger picture when assessing whether to extend credit. So in addition to gathering data from the three big bureaus to build our own score, we also already consider other variables that may help us determine a borrower’s creditworthiness. Additionally, because we build our technology in-house, we’re able to be more agile and respond to changes in the model much more quickly than many others.
We think removing unreliable information that may be incorrect from a score with such an impact on people’s lives — which the July changes aim to do — is simply the fair and right thing to do. And we believe that a broadening of the way the big three look at people’s creditworthiness — the changes coming in the fall — will make it easier for everyone in the industry to move toward our approach: of looking at the bigger picture, so that we can find ways to offer more people access to the safe financial products they need and deserve.
Ofer Mendelevitch is the head of data science at LendUp.
Disclaimer: LendUp is not providing financial, legal or tax advice. If you need or want such advice, please consult a qualified advisor.