Why It's Important to Acknowledge the IQ Gap

Due to the Watson kerfuffle, there's been a lot of talk recently about the black-white IQ gap. One argument that I hear with some regularity is that even if it is real, we shouldn't talk about it, because no good can come of it, but much bad can.

I don't know for sure whether anything bad would come of widespread knowledge of the black-white IQ gap, but I doubt it. I do suspect that most people lack the rudimentary understanding of statistics necessary to see what does and what does not follow logically from the fact that the median black IQ is 10-15 points below the median white IQ. Certainly this is true of most of the people who have been critical of Watson.

On the other hand, I also believe that most who would use this as a justification for racism are already racist. Even those who don't know about the IQ gap are not blind to its effects: Blacks commit violent crimes at rates much higher than whites, and are also much more likely to have children out of wedlock and to go on welfare. These are not valid reasons to conclude that blacks are inherently inferior to whites, or to discriminate against them, but to someone predisposed to reach that conclusion, they are probably reason enough. Learning of the IQ gap is unlikely to make much of a difference.

What I do know is that considerable harm is done by the left's continued refusal to acknowledge the racial IQ gap. When they insist that the racial achievement gap--the lower average income of black wage-earners and their underrepresentation in higher education and lucrative and high-status occupations--is prima facie evidence of racial discrimination and a need for government intervention, then we have no choice but to counter these assertions with evidence of a more plausible and better-documented explanation.

Opposition to anti-discrimination laws is not simply a matter of freedom to be a bigot. When government can second-guess employers' decisions and force them to spend resources defending them in court, that imposes costs on all of us. Likewise when employers make suboptimal hiring and promotion decisions in order to defend against discrimination charges proactively. And all of this is because of the taboo on the racial IQ gap and the consequent assumption that racism must be behind any significant achivement gap.

My advice to those who would like to keep discussion of the racial IQ gap to a minimum is to cease their loud and self-righteous scapegoating of imaginary white racists for the racial achievement gap, and to stop advocating destructive government interventions designed to solve a problem that may not even exist.

Share this

You say "the market"

You say "the market" should be defended of accusations of racial discrimination by providing evidence that the statistics merely reflect aptitude. While it may be convincing to some, it is in my opinion a dangerous argument... by saying "we're not guilty" you implicitely acknowledge that their is guilt to be had. So there is discrimination. Is is a bad thing? Maybe. Should we use force and guns to stop it? No!

Statistical Discrimination

Imaginary white racists? Even assuming, for the sake of argument, that there is a black-white IQ gap (does it matter if this gap is primarily caused by genetics or social environment?), and that this measure of IQ is important for whatever reasons you happen to think it is, the gap itself does nothing to explain away the solid evidence we already have of pervasive discrimination against blacks.

When we observe employers consistently choosing to hire white employees over black employees, and both potential employees have identical qualifications, we can't necessarily conclude that the employers are all bigots; perhaps they are just engaging in statistical discrimination, which may be less objectionable in terms of how we morally judge the employer, but is little consolation to the black employee.

Track record

According to the reference you linked, "The theory is based on lack of information. For instance, if you don't know an applicant's working ability, how do you know if you should hire him?"

This can only apply to people just starting out. As someone establishes a track record, the lack of information upon which this is based ceases.

The reference you link gives a specific example: "The variation in distribution between different groups might also lead to discrimination. Consider as an example a black and white's ability to "jump" a basketball."

But people are not hired into professional basketball teams by scouts who know nothing about them except their race. Rather, the scouts do not make an offer to a particular player without first becoming familiar with the player's individual ability.

"This can only apply to

"This can only apply to people just starting out. As someone establishes
a track record, the lack of information upon which this is based ceases."

False. Information is always imperfect. No matter how extensive the record, there is always some uncertainty regarding both past and future performance. The greater the level of uncertainty, the greater the room for statistical discrimination. There is no way to entirely eliminate the possibility of statistical discrimination so long as information remains imperfect; we can only reduce it by degree.

Bah, quibbling

False. Information is always imperfect.

You are quibbling.

It is as though I had argued (based, e.g., on experimentation) that water starting out as ice under certain conditions reached room temperature after X minutes and you had responded, "false, the temperature of the ice approaches room temperature asymptotically and therefore never exactly reaches room temperature."*

Such a response would be appropriate only in certain narrow situations. In almost any real situation, such a response would be, at best, a joke. By a certain time, any difference between the water and room temperature becomes small enough not to be worth worrying about for almost any purpose you might have. And similarly with statistical discrimination as information accumulates. By a certain point, the complaint becomes small enough that anyone still complaining should be told, "welcome to the club of people with some slight, lingering injustice to complain about" - which turns out to be absolutely everybody.

*(assume whatever additions will block smart aleck answers, such as the smart aleck answer that evaporation will keep the water measurably lower than room temperature - assume a sealed container preventing evaporation, for example)

If their's a statistical gap

If there's a statistical gap between black and white applicants, skin color does allow you some inference of suitability for the job, however I think it would dwarfed by the other sources of information you have, such as the level of education achieved or the experience. 

One possible source of rational age or race discrimination for an employer that seems more likely to me is that homogeneous groups of workers might be more efficient than heterogeneous. Since at many geographical scales, in many places in the US white outnumber blacks, and eventually qualified whites outnumber qualified blacks even more, the balance tilts in the direction of all-white employees.

But then again it may be cowardy to look for excuses to fight those who intend to stop racial discrimination by force.  

Statistical Discrimination

Even if some employers are just plain racist and flat out refuse to hire blacks (and I'm sure that some are and do), that shouldn't be sufficient to account for the income gap, as a small difference in expected salaries would create an incentive for any nonracist employers to hire blacks preferentially. It doesn't take many non-racist employers to arbitrage away the gap. For racism to cause an income gap so large would require that a large majority of employers discriminate against blacks.

Statistical discrimination I find more plausible, particularly at the low end of the job market, but I believe that the taboo on the racial IQ gap is to some extent responsible for statistical discrimination. I actually have a follow-up post on this topic planned--I'll put it up early next week.

I don't know whether the IQ gap is caused by genetics or by environment, but I'm not sure that it makes that much practical difference anyway. Whether or not the gap is due to genetics, it is pretty clear that it's due to broadly heritable factors, and unfortunately, I don't think there's any intervention short of adoption that's been shown to produce any persistent increase in the IQs of children of low-IQ parents. If it turns out that the gap is purely due to environmental factors, what can we do about it?