A new study suggests that laws banning drug tests and credit checks may hurt black applicants:

Why were African-Americans put at a disadvantage when states banned employer credit checks? It could be that black job-seekers found it harder to meet the increased education and experience requirements that employers started to impose. Or it could be that employers simply started to become tougher on black applicants because they couldn’t verify their credit histories and assumed the worst.

A powerful study published last year in the Review of Economics and Statistics shows something of the opposite happening: When employers began to require drug tests for job applicants, they started hiring more African-Americans.

“The likely explanation for these findings is that prior to drug testing, employers overestimated African-Americans’ drug use relative to whites,” the study’s author explained in an op-ed. Drug tests allowed black job applicants to disprove the incorrect perception that they were addicts.

This corresponds with a thought that I’ve been having for a while now on a related issue: IQ tests. While IQ tests are not universally banned in hiring, they do leave companies with hoops to be willing to jump through if they’re challenged, and so a lot of companies that might utilize them don’t.

Which, as the article points out, can lead to an increase in requiring credentials that aren’t challenged. I’ve been wondering if we could poach the higher ed bubble (if there is one) by simply applying disparate impact to that as a job requirement, leaving it to employers to demonstrate that the job really requires a degree. But in addition to potentially contributing a bit to credential inflation, the thought had occurred to me that it could actually hurt high IQ black applicants. Potentially by requiring a college degree that they don’t have, or by leaving it to (possibly unconscious) racist hiring manager judgment.

Which is to say, if allowed to take a test, David Alexander can demonstrate his intelligence. So a hiring manager that subconsciously looks at a black man and thinks “probably dumb” can have his concerns in that area satisfied. If the manager is systematically underestimating black IQ’s, this can act as a corrective! At least in individual cases. Now, you don’t even have to believe in the validity of the IQ test, so long as he does. If you don’t, you can try to disabuse him of that notion, but it might be better for David (or any other individually intelligent black person) to simply be able to produce a good score.

Instead of using an IQ test, you could use “successful at Super Mario Bros 2.” If some employer believes that’s a worthwhile metric, then that gives minority applicants, poor applicants, and whatever else something to strive towards. Only if they can get ahold of the game, though, which is a concern. Also a concern is that if this became widespread, you’d start to see training classes and it might become a part of the curriculum in well-heeled suburban schools. Asian-Americans might become unusually good at it. Then you might run into a Disparate Impact problem as black and Hispanic kids are disproportionately be unable to buy the game, unable to afford SMB2 tutors, and won’t have playing that game ingrained in their culture. But even then, at least it would provide an opportunity to answer the important-to-the-employer “Can play Super Mario Bros 2” metric. And it would be vastly less expensive than the alternative, which might be “Has a Bachelor’s Degree.”

Of course, that’s not what we want employers to do. Because as we know, SMB2 performance bears no resemblence to the ability to do all but a few jobs. And we want to be fair. In a perfect free market economy, we might say “Employers that make their hiring decisions based on a lackluster video game will be at a competitive advantage and so they’ll weed themselves out.” But that’s probably not what would happen. What we would be left with is a hiring qualification with the only three advantages being (a) less susceptible to stereotypical impressions than subjectiving interviewing, (b) not as reliant on networking as recommendation hiring, and (c) less expensive than college.

Which, come to think of it, isn’t the worst list of advantages I’ve ever heard. But it’s transparently dumb. Less transparently dumb is the subject of the Washington Post article, credit checks. You can at least see the rationale for using that as a criterion. But by its very nature it’s discriminatory towards those we as a society don’t want discrimination against: Poor people, those down on their luck, people who have gotten sick, and so on. Like companies refusing to hire people that are unemployed, it may make sense for any given company (whether the metric itself has empirical foundation or not) but is not good for society as a whole.

In this sense, the laws against credit discrimination continute to make sense. They are still too unfairly discriminatory. That they are not as discriminatory towards black folks in particular as the next most likely alternative may be unfortunate, but stands in suggestion that maybe bad, discriminatory policies don’t have to disproportionately affect non-Asian minorities in order to be bad policy.