3 Ways to Improve How We Calculate Credit Scores
Shouldn’t our system include the 45 million Americans who are credit invisible?
We all have a number on our head assigned to us by a nameless entity impacting our lives without regard for our individual situation that we cannot influence: our credit score. Credit scores are a shortcut to trust for creditors — Can I trust you’ll pay me back? — and as a shortcut, they do some good. But, the scoring algorithm hasn’t changed much since its debut in 1989 and we’re due for an update.
We can do better than an opaque, overly granular score that divides us. Here are three ways we can meaningfully improve the credit scoring system, so more people have access to credit:
1. Decrease the granularity
Before the introduction of FICO’s universal credit score in 1989, credit was evaluated by human beings who read and assessed long-form credit reports, a time-consuming and labor-intensive process that didn’t quickly separate the Fairs from the Goods and the Excellents — and one I’m certain no one would be willing to return to. Technological advances heralded changes to the system. By the late ‘80s, increases in computing power made it possible to collate all of the necessary data for millions of people to create a numerical score. Three decades later, the algorithm remains remarkably similar.
Putting a number to something can be useful, and not just for making faster lending decisions. It’s also helpful for identifying discrimination. Say there’s a group with an average score of 750 and another group with an average of 720, but the 720s have greater access to credit because of some other factor, such as where they live. Then discrimination is evident. It’s a lot like an SAT score. A poor kid from southern Mississippi like myself has a much better chance with the admissions team if their application includes a standout score. Without the score, the application might not appear special.
However, try to pinpoint the difference between a credit score of 650 and a 670. It’s a tough task because there’s not a substantive distinction. Credit scores are overly granular and the breakpoint is arbitrary. Sure, scores are somewhat correlative with default rates, but since the scoring model is not public, we don’t really know the difference.
2. Share the algorithm
Consider this: Do you know how to move your score up or down one hundred points? I certainly don’t. None of us know exactly. Each of our scores is calculated based on an opaque and individual combination of payment history, length of credit history, credit mix, newness of credit, and credit utilization.
Not only do we have no way to influence our score, an action that benefits us financially can hurt our score. Right now, my credit score is 840 because I’m carrying more debt than I’d like to and I make payments regularly. When I pay off the debt, it’ll be good for me, bad for my score.
As long as median credit scores reveal persistent racial disparities and there is no clear lever we can pull to increase our scores, the score is discriminatory. Our score impacts our lives — what credit we qualify for, what we’ll pay to borrow that credit, even if we will land a job — so we should know exactly how to impact our score.
3. Shift to community-based underwriting
We share the sports team we root for, the type of car we drive, our hometowns, our political affiliations, even our blood types, but our credit score, like our social security number, is ours alone. Our score is not linked to others, not even within our own households. Even though our partners, parents, and adult children all contribute to our true financial picture, currently our scores are completely independent.
The credit scoring algorithm measures our ability and willingness to pay by tracking payments on debt. Without debt, there are no payments; without payments, there’s no score. The result? 45 million Americans who are credit invisibles — they have no score because their credit history is too thin, it’s gone stale, or they’ve never taken on debt. Shouldn’t our credit system include these millions of people? Adding in a layer of community to the credit scoring system would remedy this.
45 million Americans have no credit score.
Shouldn’t our credit system include these millions of people?
If we’re not ready to look at the community to assess credit, we can at least take steps in the right direction. As a beginning, couples should be able to link their scores. Federal student loan repayments already link married couples filing taxes jointly when they calculate income-based repayment plans. They acknowledge that a couple’s capacity is more accurate than the individual’s capacity. So should our credit scores.
Can you imagine a future without individual credit scores?
Credit scores do more harm than good, yet what would we do without them? A common thought is that without scores, we’d have no way to track payment history, and people would skip out on debt without consequence. That’s the same line of reasoning as, If you don’t give grades, no one is going to do anything in school. Well, I went through an education system that didn’t give grades: Stanford Business School. I can attest that people still did the work — we were beholden to one another as students and classmates. We’re interconnected. Our financial measurements should reflect that.
At Sawa, we believe it’s a lot more accurate to look at the capacity of the community rather than the credit history of the individual. While any one person may be at risk of missing a payment in any given month, there’s always enough capacity to surge when you look at the community as a whole. This is the consistent, but mitigable problem of payment ability poorly assessed. I believe that by adding a layer of community underwriting to the current system, we could grant access to more credit to more people.