Read on Twitter
So... I've been seeing these kinds of remarks from my esteemed colleagues in the twittersphere for a while now. And my... perspective is that (A) most of the studies that purport to prove systemic racism are naive or mendacious, but (B) it also nonetheless definitely exists. https://twitter.com/a_centrism/status/1329134883722113024
I'll explain what I mean in a moment- and what qualifies as 'systemic' from my PoV- but first let me give some examples of what doesn't count.
Example 1: People of colour earn less money than whites even after adjusting for occupation, experience, hours worked, and education.
Example 1: People of colour earn less money than whites even after adjusting for occupation, experience, hours worked, and education.
This statistic gets thrown around a good deal (including in the Biden admin's policy website addressing racial equity), but honey, I'm afraid that IQ remains a powerful independent predictor of job performance even after controlling for all of those factors.
There are mountains of evidence for persistent and pervasive IQ gaps between racial groups, on the average, and even in settings that filter by IQ, lower-scoring groups get in by slimmer margins, thereby creating within-setting racial differences (albeit to a lesser degree.)
(Amy Wax got into hot water for commenting on this phenomenon in law school- she had never seen a black student in the top quarter of the class, and rarely in the top 50%. Such is the tyranny of the bell curve.)
I would add, of course, that 'people of colour' are a highly heterogeneous group, and higher-performing east-asians and indians, along with relatively high-performing black immigrant groups, are usually omitted from this narrative.)
The educational system in the US has itself also been distorted by the pressure to guarantee equality of outcomes. (Put bluntly, if you make it easier for black and hispanic kids to get into college or high school, their qualifications become correspondingly less impressive.)
So... unfortunately, this particular statistic doesn't really prove that employers are racist in the sense of judging individuals unfairly on the basis of their particular track record of real performance. You'd need to measure real job performance to establish that.
Example 2: Black americans are more likely to be restrained using force during an arrest.
This one is trivial- what makes you think black americans aren't more likely to resist arrest? Can you think of *any* factor, cultural or otherwise, that might make them angrier at cops?
This one is trivial- what makes you think black americans aren't more likely to resist arrest? Can you think of *any* factor, cultural or otherwise, that might make them angrier at cops?
Example 3: Black and hispanic children are more likely to face disciplinary action at school.
Here's a bomb: I'm afraid that there's pretty good evidence for large genetic effects on childhood misbehaviour, including in racially heterogeneous samples. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1913189/
Here's a bomb: I'm afraid that there's pretty good evidence for large genetic effects on childhood misbehaviour, including in racially heterogeneous samples. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1913189/
Not that one should necessarily have to establish a non-environmental *cause* for a child's misbehaviour in order for discipline to be proportionate and reasonable. This could be purely a function of poverty or culture and yet not be the school's fault. https://www.city-journal.org/html/who-misbehaves-15811.html
(One interesting thing I'd note is that The Wire, surely the darling of left-wing television dramas on the subject of dysfunction in mixed-raced american cities, explicitly calls out this problem, although I doubt David Simon thinks the cause is genetic.)
Okay, so- having delved, at some length, into examples of social disparities and studies thereon that do not, to my mind, furnish dispositive evidence for the existence of systemic racism... what am I talking about, then?
Let me give you an example that was actually furnished by James Flynn in his paper defending the hereditarian position on B/W IQ gaps (Flynn himself is perhaps the best-known environmentalist researcher on this topic.)
https://scottbarrykaufman.com/wp-content/uploads/2018/12/Flynn-2018.pdf
https://scottbarrykaufman.com/wp-content/uploads/2018/12/Flynn-2018.pdf
Flynn gives a hypothetical scenario where a landlady is deciding whether to lease their apartment to a 25-year-old black ADoS male or to a 25-year-old korean-american woman. (Let's say the landlady knows nothing else about the applicants.) Who does she rent to?
I'll give you a clue: One of these applicants has about a 1 in 3 chance of being a convicted felon and the other has virtually none. So, if she wants to minimise the chance of her apartment being wrecked and maximise the chance of rent being paid on time, who does she pick?
...She's going to pick the 25-year-old korean-american woman.
What's more, if the landlady, as a business owner operating in a competitive rental market, wishes to remain in business, she *must* adopt this practice in order to remain competitive with other landlords.
What's more, if the landlady, as a business owner operating in a competitive rental market, wishes to remain in business, she *must* adopt this practice in order to remain competitive with other landlords.
Is this going to result in a significant number of 25-year-old black-american men who have never committed a crime in their lives, and are otherwise upright and decent citizens, being denied rental opportunities? Yes. It absolutely will.
This is, virtually by definition, racial discrimination at a systemic level (insofar as market competition and cost-cutting under conditions of imperfect information is a kind of 'system'.) And it penalises individuals for the stereotyped behaviour of their larger racial groups.
Note this does not require that our hypothetical landlady hold any particular opinion as to *why* 1-in-3 black american males are convicted criminals: All that is required is that the average difference in criminality *exists* and is predictive of future behaviour.
Our hypothetical landlady might be in total agreement that some combination of redlining, lead exposure, food deserts and the school-to-prison pipeline explains black crime rates. She might even be 100% convinced that white racism is to blame, and recognise her own complicity!
Nevertheless, she *must* discriminate. She *must* exclude black applicants (or charge them higher prices, or give them lower-quality rooms to rent), if she wants to remain in business in competition against other landlords.
(This is why, for example, black-owned banks are at least as likely to discriminate against black applicants as white owned banks. Black ownership does not render them suddenly impervious to all the usual market forces at work.)
https://www.researchgate.net/publication/5150640_Do_Black-Owned_Banks_Discriminate_against_Black_Borrowers
https://www.researchgate.net/publication/5150640_Do_Black-Owned_Banks_Discriminate_against_Black_Borrowers
There aren't really any simple or painless fixes to this problem. A partial solution is to give landlords, employers and other discriminators (for discriminate they must) more information about the individual- criminal record, employment history, educational attainment, IQ, etc.
But this will not really result in less discrimination against minority applicants *on average*. It will simply mean more discrimination in cases where it is more-or-less merited and less discrimination in cases where it is not. Minority advocates don't want to hear this.
(This is why, in states like California, there is currently legislation in the pipeline making it illegal for landlords to inquire as to someone's criminal record before considering a lease. Which, of course, will force landlords to fall back on group-based stereotypes.)
Another solution is to try and force the outcome- make it illegal for landlords (or other businesses) turn down minority applicants, at least until a certain % of their rooms/jobs/loans etc. go to those applicants.
Unfortunately, this doesn't make the actual behaviour of felons go away, so this would be an excellent way to make landlords, banks and businesses flee your state for somewhere they can make more money (especially if there are price controls on top.)
It's a genuinely thorny problem. But the key point I want to emphasise here is that, yes- members of underperforming racial groups will routinely wind up in situations where they are judged, as individuals, based on the stereotyped characteristics of their group.
(Does this outweigh the effects of affirmative action and diversity quotas, which are much more overtly systematic and clearly punish white applicants regardless of their background? Maybe, maybe not. That's an empirical question for another day.)
But regardless, one can extrapolate from first principles that 'systemic racism' more-or-less does exist, and will exist independently of any particular ill-will or animosity on the part of the people who practice it.
