Insider’s experts choose the best products and services to help make smart decisions with your money (here’s how). In some cases, we receive a commission from our our partners, however, our opinions are our own. Terms apply to offers listed on this page.
Imagine deciding you’re ready to buy a home. You’re 30-something with a good job and steady income. You have some debt — a car note, a student loan — but your budget can support it, and you know how much you can afford to spend on housing. Aside from some issues with credit card debt and missed payments in your 20s, you feel good about your financial position. But you just can’t get a decent loan, or even get a loan at all.
There are many factors that go into underwriting a mortgage loan, among them your income and the value of the house you intend to buy.
But a key one is your “character,” and it’s your credit history — including your credit report (which contains information about past payments, accounts you own, aliases you’ve been known by, and more) and your credit score (the number assigned to you based on your credit report) — that tells your lender who you are as a person.
To borrow money for any purpose — a home, a vehicle, a business — a lender will evaluate you as a borrower using a framework called The 5 C’s of Credit.
Capacity is your ability to repay the loan, also known as your debt-to-income ratio.
Capital is your savings, investments, retirement accounts, etc. Basically, any money you could use to repay the loan if you lost your primary source of income.
Collateral is the asset you provide as security for the loan. On a mortgage, the house is your collateral. Same goes for your car if you’re taking out an auto loan.
Conditions are related to the broader economy and interest rates, as well as how you plan to use the money.
Character is about how likely you are to repay the loan based on your past behavior. Today, your credit report is used as a proxy.
Credit scoring, on its face, appears neutral and unbiased — a seemingly appropriate way to measure someone’s ability to repay a loan.
But, according to Camille Busette, director of the Brookings Race, Prosperity, and Inclusion Initiative and a senior fellow in Governance Studies, our credit-scoring model in the US fails to account for the ways systemic oppression and histories of financial exclusion can leave scars on a person’s credit file.
“When you look at credit scoring, what it is essentially doing is asking, based on past behavior, is this person able to pay their bills on time,” she said. “That sounds pretty unproblematic. But when you start thinking about who are the kinds of people who are more likely to pay their bills on time, and what are the factors that enable you to pay your bills on time, one of the major factors is income.”
Throughout history, nonwhite groups have been systemically held back from high-paying jobs that make it possible to build wealth and gather assets — things that help ensure you can pay your bills on time.
Public policy, such as redlining and lack of access to the GI Bill, blocked Black Americans and other people of color from accessing wealth-building tools that were available to white Americans. Over time, that has created a racial wealth gap between white households and households of color.
“Your ability to make a payment on a timely basis over generations is impacted — either positively, if you benefitted from those policies, or negatively if you did not benefit,” said Busette. “When we think about the credit-scoring system being neutral, actually, being neutral is biased because it doesn’t reflect some of the barriers that households of color in particular have faced.”
“Character” has been tied to lending for as long as lending has existed.
Centuries ago, commercial lenders extended credit to business owners after asking around about them, gathering information from their neighbors and communities; they were essentially looking for character references.
Later, in the 1800s, companies began to formalize that reference-gathering process. The Mercantile Agency, developed in 1841, gathered and collated information about prospective business borrowers and stored it in physical ledgers in New York City, then shared that information with subscribers.
These early “credit reports” often contained rumors about a person’s character and assets — not to mention judgment about the individual based on those rumors — as well as personal information like race and ethnicity. Time magazine published an article in 2015 that quoted one illustrative Mercantile Agency report: “Prudence in large transactions with all Jews should be used.”
By the late-19th century, commercial lenders were looking for a more streamlined way to evaluate borrowers. And by the 1860s, the first alphanumeric credit-scoring model was in use, though it still relied on subjective assessments of “character.”
Fast forward several decades and consumer lending was becoming increasingly popular as workers sought access to goods. Many individual department stores and other retailers hired their own “credit managers” to evaluate consumers on a case-by-case basis, but the Retail Credit Company of Atlanta, formed in 1899, was key in bringing the commercial sector’s credit-scoring model to consumer lending.
Workers at RCC, known today as Equifax, effectively went door-to-door asking local merchants about the character of consumer borrowers they’d loaned to. They then aggregated that data into a guide and sold it for $25, making it easier for lenders to make decisions about borrowers based on their history as described in their credit file.
But just as in commercial lending, the system was deeply unfair, and “character” had prejudice built in.
“In the past, people who had access to credit were usually people who were seen as well-established members of the community,” said Busette. “In general, at the consumer level, Blacks, Latinos, and nonwhites were cut out of the credit system. It was pretty systematic; the only people who were really going to get mortgages or auto loans were people who were white and pretty well-established.”
The Civil Rights movement brought big changes to credit reporting in the United States with the goal of eliminating subjective character judgments and making lending equally accessible to all. With increased attention to economic disparities and pressure from Black leaders, three key laws were passed to increase fairness in lending: the Equal Credit Opportunity Act; the Fair Housing Act; and the Fair Credit Reporting Act.
“As a result of the Civil Rights movement, there was a focus on how this failure to extend credit impacted Blacks and other people of color negatively,” said Busette. These new laws “were intended to end the discriminatory practices of consumer lending that had been operating up to that point.”
The final step in the credit-score-as-proxy-for-character journey took place in 1989 when the FICO score was adopted by the Big Three credit bureaus — Equifax, Experian, and TransUnion. The goal of systematizing and automating credit scores was to eliminate the appearance of bias that had plagued older character-based lending and scoring models.
But while FICO touts itself as “unbiased” and “impartial,” and says it “replaced hunches with calculations,” experts say that’s not always how things shake out in practice.
As it has throughout history, the credit-scoring model is evolving. One common complaint about the current model is the limited number of variables taken into consideration to generate a credit score. For example, mortgage payments are counted, but rental payments are not. This means individuals who do not own homes but are paying their housing costs in full and on time every month are not being evaluated in the same way as homeowners.
The credit bureaus are beginning to respond. Experian recently added on-time rental payments to its Boost feature, which previously allowed consumers to include on-time utility bill payments in their scores. Consumers must actively use Experian Boost; rental and utility payments are not counted automatically. TransUnion also offers a feature to landlords to report on-time rental payments, and Equifax offers one to some renters.
These steps are helping to “bulk up” credit files for those without much credit history, and also help those considered “credit invisible” to develop a profile. But some experts say it’s not enough, especially given how widely credit reports are used to evaluate “character.”
“One clear step is to prohibit credit reports from being used in landlord-tenant decisions,” said Terri Friedline, an associate professor of social work at the University of Michigan whose research focuses on financial system reform. “Stopping medical debt from being included in credit reports. Not allowing employers to not hire you based on your credit report. Those, I think, are clear and important steps on the road to re-envisioning what this system can look like.”
Beyond that, Friedline said, stopping the credit-scoring system from further overreach is critical. There is currently research being conducted into the use of social media data in credit scoring that, she said, is beyond what many people feel comfortable with.
She explained, “There’s a theory about the number of friendships you have, for example, on Facebook, and about how close or far they’re perceived to be, that that has some bearing on your connectedness to a community and what that means for your ability to repay.”
Busette said she’d like to see other lending models considered that don’t depend so wholly on the credit-scoring system. She cited a new program from Bank of America that tosses out the old lending model to evaluate mortgage applications as an example to watch.
And, she said, the Consumer Financial Protection Bureau could take action. “The credit bureaus are highly regulated by the Consumer Financial Protection Bureau, which could do a study to figure out how to eliminate the biases in the credit-scoring system. That might be a politically more palatable way of addressing it,” she said. “I think this is an opportunity for the Consumer Financial Protection Bureau.”
Editorial Note: Any opinions, analyses, reviews or recommendations expressed in this article are those of the author’s alone, and have not been reviewed, approved or otherwise endorsed by any card issuer. Read our editorial standards.
Please note: While the offers mentioned above are accurate at the time of publication, they’re subject to change at any time and may have changed, or may no longer be available.