Risk & Compliance Do credit scores have a disparate impact on racial minorities?

Sep112012

I am troubled when I read allegations in the press that FICO® Scores discriminate against people of color. That’s because a credit score is nothing more than the output of a mathematical formula built to rank-order the likelihood that a person will repay the debts they have incurred. To say that FICO Scores are unfair suggests that the formula was built with racial bias. To say that the use of a FICO Score is unfair suggests that it would be fair for a creditor to use the score to assess the credit risk presented by most borrowers, but not if the borrower is a member of a racial minority. 

The first suggestion, that the formula was built with racial bias, is untrue. Building a score that’s both “fair” and predictive is certainly a priority for us at FICO. As part of this, FICO® Scores have never used race, gender, marital status or other legally prohibited bases as inputs.

As to whether the use of credit scores has an unfair impact on racial minorities, researchers at the Federal Reserve Board (FRB) have studied the relationship between credit scores and race to see if empirically derived credit scores produce a disparate impact on specific populations. Their finding: credit scoring does not produce a disparate impact.

As part of that study, the FRB team compared credit scores from two types of models. One type produced a more traditional credit score and was developed on a national consumer population. The other type was a series of “race-neutral” models. Specifically, these models were developed for individual demographic populations (e.g., Hispanic, Black, Asian). Disparate impact would be apparent if a minority group scored higher on the race-neutral scorecard than on the baseline credit score. In other words, for each scoring model developed for a specific minority group, evidence of disparate impact would be found if this model yielded higher scores for the group than did the benchmark scoring model.

On page 26 of their 2010 study, the FRB’s Avery, Brevoort and Canner wrote:

“Our results provide little or no evidence that the credit characteristics used in credit history scoring models operate as proxies for race and ethnicity. The distributions of credit scores for different racial or ethnic groups or across genders are essentially unaffected by the reestimation or redevelopment of the baseline credit scoring model in any of the race- or gender-neutral environments. This suggests that credit scores do not have a disparate impact across race, ethnicity, or gender.” (Bold is mine.)

FICO will continue to build credit scores that are objective measures of risk, while at the same time working to demystify credit scores for consumers. We want to empower everyone, regardless of race, to manage their credit scores well.

3 Comments

  • bigguy

    Credit scores are used to discriminate against minority groups because there are many issues that derive from having or obtaining good credit. For example the cost of living, the ratio of jobs per citizen, and the need for governmental intervention. In many ways we as a people are really fighting an uphill battle; prepare yourself , get a good education, get more education to battle for 10 jobs per 100 applicants. More than likely the 10 available jobs will be filled by non minorities, and the rest that are educated and fully qualified would have to settle for less or continue to get educated and make more bills to start the whole process over again. Out of the 90% that are educated citizens and followed the system for success 20% will struggle for the rest of there lives by settling for a job lower than there qualification(s) and the other 70% or so will need some type of governmental assistance program just to survive. There is a systematic approach to obtaining credit worthiness that derives from many approaches. So to suggest that the credit system can be used as a way to discriminate against minorities according to the standards of FICO is unfounded and untrue, but when challenged by actual events of occurrences and looked at without focus on procedure vs. reality one would consider differently if given the same facts.

  • Amnesty

    Good try, guys. But you didn’t answer your OWN query: whether credit scores have a disparate impact across race, ethnicity, gender. (Just because you so conclude doesn’t mean that your evidence supports your own conclusion.)
    It is (may be) true that FICO doesn’t look at race, etc. And it may be that USING the FICO score in and of itself doesn’t have a disparate impact (I don’t agree with that either). But you skipped a step. If you look at whether the compilation of the FICO score itself has a disparate impact – you’re wrong. It does. The race-neutral factors that FICO looks at are disparately more negative among people of color. It can be argued that this is attributable to poverty, not race/ethnicity, but when it is well documented that people of color are disparately poor, it’s a defeatist argument. Because of poverty (and I use that term very widely to include income, education, unemployment, crime, etc.), people of color are less like to even HAVE credit cards, or credit, or bank accounts. Considering the well established tenet that no credit is just as bad (or worse) than bad credit for FICO purposes, then it stands that these people will also have lower FICO scores. Racial profiling and stamping in African American and Latino communities with respect to mortgages and lending practices has been documented, in that these communities receive more unfavorable loans, which leads to higher defaults, Defaults, which are at least partly attributable to racist/ethnicist practices, impact one’s credit score. I could go on. However, suffice it to say, that it’s a bit ridiculous to argue that the process and factors that go into determining FICO scores are not disparately impacted by race and ethnicity.
    The article is blithely vague with respect to “race neutral” models and models “developed for individual demographic populations”. what does this mean? That a Latino with a history of default, low income and no education could possibly score differently on the race neutral and traditional models? That sounds like racism, not disparate impact. No one is arguing that an African American and a middle class white person would both have similar FICO scores if they have similar credit histories. However, the argument is that the average African American or person of color does NOT have a similar credit history. And the very reason for that is rooted in racism, ethnicism, genderism and other negative biases. And FICO scores are created out of that already-existing bias.
    And the most critical problem with that is that FICO scores are now becoming so pervasive, as to affect your ability to not only BUY a house and a car, but also RENT a house and a car, get a job, obtain health insurance, homeowners insurance, student loans, etc. Which means that your FICO score indirectly considers and uses information that in and of itself would not be allowed by employers and providers.
    FICO people can argue that employers and providers have a right to know if a person presents a bad credit risk. But they cannot (legitimately) argue, that their scoring system does not have a disparate impact based on race and ethnicity. That’s disingenuous.