GEICO Sued for Racial Discrimination

Before reading this article in the St. Paul Pioneer Press, I would have thought that plaintiffs against GEICO would use a disparate impact theory. That is, the facially neutral underwriting criteria has a disparate impact on African-Americans.  With the disparate impact theory (under the Civil Rights Act) which does not apply to insurance contracts, the business need only show business necessity for a successful defense.  Risk classification, arguably, is a good business necessity reason.  However, the plaintiffs seem to be asserting a disparate treatment approach which for racial discrimination is likely illegal under every state’s insurance law.  The plaintiffs claim, GEICO

“"knowingly and purposefully used occupation and level of education as proxies for the race of its insured," leading to higher premiums for African-American customers.”

This requires a level of proof that seems to be high.  The plaintiffs will need to find a smoking gun where GEICO employees talk about this openly in memos and the like.  What seems to be missing is that if an African American had high income and a high status job, he or she would be eligible for GEICO’s lower rates too.  Similarly if a white person had low education and low income, that person would get the higher rates, all other things held constant. 

One would have to prove that GEICO knew that it was going to “punish “ some majority racial customers just to discriminate against some minorities and that they would set up an underwriting system which benefit some minorities just so that they could discriminate against other minorities.  I may be simple, but I think that if GEICO was really trying to discriminate against minorities it wouldn’t set an underwriting rule that allows high income/high education minorities to benefit form the underwriting rule.

 

Comments

[Bob]

"One would have to prove that GEICO knew that it was going to 'punish' some majority racial customers just to discriminate against some minorities and that they would set up an underwriting system which benefit some minorities just so that they could discriminate against other minorities."

In fact, when you consider that non-minorities (a category that, for the purposes of this lawsuit, comprises everyone who isn't black) constitute at least 85 percent of the population, the number of non-minority customers who are "punished" under this rating scheme is almost certainly much larger than the number of minority group members who are "punished"-- even if the percentage of blacks that is adversely affected is greater than the percentage of whites that is adversely affected. Under these circumstances, it seems ludicrous to suggest that the challenged rating variables were intended by GEICO to serve as a "proxy" for race.

[Megan Mahan]

Good comments here, gentleman. Your commentary adds perspective to my previous opinions of GEICO's underwriting practices. And while I see where you both are coming from (and agree that the chances of GEICO facing legal ramifications are slim), I think this suit will hurt GEICO's credibility in the eyes of consumers...but will it be enough to affect GEICO's bottom line?

Of that, I'm not totally convinced. However in this case, I think it's clear that not all publicity is good publicity.

[martin]

Good point Bob. I am sorry I didn't think of it.

You are right too Megan--there is a cost in terms of negative publicity. In fact, I know one other company (from hearsay) that decided specifically not to use certain legal rating variables because of how they might be perceived by the public. However, if companies like GEICO persist and win the good risks in terms of market share becasue of their better pricing model, tehn other companies will have to do it to compete.

[insurance guru]

I think you guys have done a good job dissecting some of the uphill battles for GEICO, but I think you may actually find a smoking gun on GEICO here. I don't think it'll be too hard to find at least 1 email from one major executive or underwriting manager that says something to the effect of "these are bad areas" to insure drivers, or "we don't want to be insuring these people." and have it refer specifically to predominantly black areas each time. I think this would be enough, along with recent stats that show that GEICO does not have favorable rates to predominantly black territories in each state. In fact, recent statistics in NJ show that the most populous city, that has the largest black population (Newark NJ) has 1/10th the amount of vehicles insured than the most white populous areas of the state. Yes, this smells like a disparate impact theory not recoverable under a Civil Rights Action, but I think if you can connect PURPOSEFUL INCOME rating - you might find GEICO loses here. The reason: If you look at Federal case law that has rendering decisions dealing with INCOME based criteria- in the Federal Housing cases - it is unambiguous that the Supreme Court of the US has held that use of such "income" criteria is racially discriminatory. I don't think it takes a rocket scientist to realize that using education and occupation - is a proxy for income, which is a proxy for race. Is it purposeful? I would have to say each one of them are so obvious, its hard to say how one couldn't see it. It's like saying that "I am sorry, I needed this for risk based business decisions" while basing rates upon home ownership, knowing that poor people don't own homes in the U.S. This is another trend that will lead to similar cases.