Capitalism has a disparate impact
Matt Yglesias has an excellent post discussing Ibram X. Kendi’s attempt to redefine the term ‘racism’ from personal animosity against another racial group to advocacy of policies that widen the gap between two racial groups:
When the book was first published in 2019, that’s not how it was received. Kelefa Sanneh’s excellent review in the New Yorker heads straight for what I think is the core weirdness of Kendi’s ideas. If we accept the definition that a racist is a person who supports racist policies, and what makes a policy racist is that it “produces or sustains racial inequity,” then determining which policies are racist requires exhaustive analysis of controversial empirical questions. Sanneh uses the example of “ban the box” laws which prohibit employers from asking about past criminal convictions. Many activists and the National Employment Law Center regard this as an important anti-racist measure since African Americans are more likely to have prior convictions and thus be disadvantaged by this question.
But Jennifer Doleac and Benjamin Hansen find that “ban the box” laws lead to worse employment outcomes for Black men because absent specific information about past criminal records, employers engage in statistical discrimination.1
“Are these laws and their supporters racist?” Sanneh asks. “In Kendi’s framework, the only possible answer is: wait and see.”
Just to be clear, there are public policies that have a disparate impact and are also motivated by racism. That is probably true of our crack cocaine laws, and also some zoning regulations. Nonetheless, Yglesias points out that it is a very odd definition of racism where a person might end up being called racist for advocating laws that they thought were anti-racist—such as “ban the box” or “defund the police”.
In fact, I think it’s even worse than Yglesias suggests. Why should we worry about how laws affect inequality? Isn’t the actual goal to maximize aggregate utility? And if that’s not the goal, if you have a Rawlsian value system, then surely a public policy that improves the utility of disadvantaged minorities is not racist, right?
In 1959, Mao implemented an egalitarian program called the Great Leap Forward, which equalized the pay of Chinese peasants regardless of their productivity. Unfortunately, that program also reduced the incentive to work hard and output fell sharply, resulting in the death of roughly 30 million people from starvation. Not good. Indeed, maybe the worse disaster ever.
This example might seem pretty far removed from our society. Don’t most actual policies that increase economic disparities also make the poor worse off? Actually, not as often as you’d think.
If you polled most American economists they would probably agree that:
1. Capitalism makes the poor better off than socialism.
2. Capitalism results in more income inequality than socialism.
And I don’t think it’s just economists that have this view. Over the years, I recall reading a number of studies suggesting that African-American voters tended to be skeptical of candidates that advocated socialism, and often voted for more centrist Democratic candidates.
I agree with Yglesias that it’s a bad idea to define “racist” as someone who advocates programs that increase racial disparities. (And who gets to make that judgment? Kendi? Trump?) But I would go even further. Reducing racial disparities should not be the primary goal of public policy. I don’t want to live in a country where everyone is paid the same, because in that country the average level of real wages will fall to a very low level. Instead, the goal of antiracists should be raising the utility level of disadvantaged groups.
Lots of obvious policy initiatives like decriminalizing crack cocaine or eliminating residential zoning restrictions can do this without making society worse off. But I would go even further. There are non-obvious policy initiatives that can improve the well being of disadvantaged groups, such as lower tax rates on capital income and free trade agreements. Those initiatives won’t necessarily reduce racial disparities; they might even widen them. But they will improve the well being of disadvantaged groups.
To summarize, the anti-racist movement is not merely off track in redefining racism from personal animosity to economic disparity, it is also infected with zero sum mentality, which assumes that any policy that widens disparities reduces the welfare of the lowest paid groups. And that’s just false.
PS. This is a good example of someone motivated by anti-racism, taking an action that makes racism worse.
Oct 13 2021 at 6:20pm
Tougher drug laws were supported by the Congressional Black Caucus among many other notable black leaders. Are you asserting they were motivated by racism? Things get confusing quickly on this topic.
Oct 13 2021 at 6:58pm
I would suspect that the members of the Congressional Black Caucus might be focusing on their primary job. Getting reelected. Better or worse for people that happen to share their skin color well down the priority list. Rhetoric aside of course.
Oct 13 2021 at 7:32pm
Because of majority minority districting the vast majority of the CBC come from heavy black districts.
If they are voting for these crack cocaine rules to get re-elected, it suggests it was popular amongst largely black communities.
Oct 16 2021 at 12:34pm
They initially supported them, then saw the impact when actually carried out. They no longer support them.
Oct 14 2021 at 11:03am
My recollection of the 80s was that there was a perception that crack cocaine was unusually harmful to the black community. Some of that is just my memory, but there are plenty of reports of the theory that crack was a CIA plot specifically designed to destroy black communities, and my understanding is black leaders wanted crack off the streets.
If I remember correctly, the disparate sentences on crack dealers were intended to help the victims of the crack epidemic. As Scott points out, under the Kendi analysis, you can still argue that it was racist in effect. Thirty years later, it’s believed that by putting a lot of crack dealers and possessors in jail, the laws did more harm than good.
Oct 14 2021 at 2:59pm
There is a very big difference between “racist in effect” and “racist in intent” (which Scott claims includes tough crack cocaine laws). We as a society seem very confused as to which we care more about. I would argue the entire BLM + “Defund the Police” movement was racist in effect as it resulted in significant damage to black neighborhoods and a spike in homicides (more pronounced in cities where BLM was active). And yet the movement is generally praised. If we care more about appearances (which can be very deceiving, such as in the case of crack cocaine laws) than effects, we are never going to solve much of anything.
Oct 15 2021 at 11:20pm
I’m inclined to reject the very idea of ‘racist in effect.’ I don’t think one can be ‘accidentally’ racist, only consciously or subconsciously (which I’m differentiating from accidentally) racist. Of course, a well-intentioned, non-racist policymaker with really stupid ideas can hurt the people he’s trying to help worse than a more competent racist, but that just shows that racism is not the only vice (or even necessarily the most dangerous one, in some contexts) in policymaking. But I guess it’s a question of definitions.
Juan Manuel Perez Porrua Perez
Oct 13 2021 at 9:02pm
A problem is that the existence of “racism” (in Kendi’s sense) encourages racism (defined as prejudice) as a justification or apologetics.
That there are public policies that have a disparate impact and are also motivated by racism is true (c.f. Lee Atwater), but it’s also true that racism (again, defined as a kind of prejudice) is behind a lot of the opposition to improv[ing] the well being of disadvantaged groups., not some concern for the general welfare.
You can never rule out ill will. It’s not the best thing to assume ill will, but you can never rule it out.
Oct 14 2021 at 12:23am
Talk about “reducing inequality,” “making the poor better off,” “making Blacks better off,” etc., is mostly *loose talk*, because it does not specify the time when the desired result is to occur. Take *making the poor better off*, for example. A policy of radical redistribution, instituted as quickly as possible, might make the poor better off in the near term (as soon as the redistribution had actually been carried out), but after the passage of some more time the poor (and almost everyone else) might be worse off than at the start. If the objective was “making the poor better off a month from now,” the redistributive policy would be a success; if it were “making the poor better off five years from now,” a failure. If the statement of objective omitted temporal specification, no verdict would be determinable.
Of course, in politics *loose talk*, and fuzzy thinking, is the rule; this is one of many examples.
Oct 14 2021 at 7:19am
“Reducing racial disparities should not be the primary goal of public policy.”
I guess this is right, but I have a couple of comments.
(1) If you think it shouldn’t be, then you should be pleased, because reducing racial disparities is definitely not the primary focus of either existing policies, or the administration’s current agenda. The tax credit and the infrastructure bill aren’t focused on race at all. So, yes, but who are you arguing with? (Don’t argue with Twitter, it has inexhaustible wells of idiots.)
(2) Policy discussion is almost always marginal and reactive, not idealist. That is, policy discussion answers the question, “What is the problem now?” Not the question, “What should policy aim for?” Saying that racial inequality is/isn’t a fundamental goal is not necessarily germane to the issue of what policy questions are going to get a lot of air time.
(3) There is an efficiency argument for racial equality. Plus, it’s a tough nut to crack, given the deep historical roots. So perhaps this problem just needs a lot of working through in the public sphere. That’s what the public sphere is there for! The conversations have to be had, or the issue will never go away. (Maybe!)
Finally, an argument against what you say. One of America’s great founding myths is that it was born out of the rejection of oppression – the pilgrim fathers, whose Protestantism meant they were not fairly treated in the corrupt old world, came and created a new world where no religion could be established as an instrument of oppression (I may be eliding a coupla hundred years of details). Rejection of inequity is a pretty fundamental American value. So it’s not obvious to me that the state shouldn’t combat inequality. In fact, that may be the state’s most important role.
Oct 14 2021 at 11:43pm
The founding myth of American equality was specifically equality of individuals before the law or equality of opportunity, not equality of outcome between demographic groups. I think America’s founders – or for that matter most Americans until the last couple decades – would have found this use of ‘equality’ in the context of politics as mainly meaning equality of outcome (and especially equality of group averages) a somewhat eccentric use of the word, and were quite comfortable with great inequality of outcome. A minor pedantic note: America’s early settlers weren’t mistreated because of their protestantism – they were almost all from protestant countries – but rather because they belonged to marginal sects of protestantism – puritans, Quakers, Mennonites, Methodists, etc. – that were disliked by the official protestant churches of Europe, like the Church of England.
Oct 15 2021 at 3:50am
Thanks, I had forgotten that part of the history.
“America’s founders…equality of outcome…eccentric use of the word…quite comfortable with great inequality of outcome.”
This seems like a lot to put on the shoulders of 17th century farmers. I’m willing to bet that a number of them were also concerned with inequality of what you call “outcome” and they would have called livelihood. They wanted land.
Of course, equality before the law and economic equality are distinct; but they’re also connected. There are two obvious connections in this context: (1) when inequality follows racial lines, it suggests that racism, either social or legal, is driving the “outcomes”; (2) the connection that Kendi argues for, which is that policies which treat different races differently create racial strife. So, yes, the argument requires an extra step, but there are plenty of reasons to think that the argument can be made complete. I’ve just downloaded Kendi’s book, so I’ll have a look and see if he does successfully make it!
Oct 15 2021 at 12:45pm
Everyone wants land/money. Everyone wants a better life for themselves and their family. That doesn’t mean they want “equity”/equality of outcomes. Would they have been satisfied if richer people lost their land? If not, equity was not their concern.
So racism is driving the success of Asian Americans and Nigerian Americans? Sexism is driving the success of women in college and all the men in prison?
Oct 14 2021 at 9:39am
Like most things in life, this reminds me of a Scott Alexander post where he anticipates a lot of this discussion. It was posted in 2017, a couple years before Kendi’s book was released, but already there was a trend to redefine “racism” into what Alexander called “Definition by Consequences”, which is more or less what Kendi argues and was being treated like a new, Sophisticated Definition. He goes on to point out all sorts of weird implications of this definition, such as:
And for what it’s worth, Kendi’s response to the weird implications to his definition has been to ignore them. I’ve watched videos of him giving talks and later in the Q&A, people will ask about these sorts of things. One that particularly comes to mind is a student who first brought up Kendi’s assertion that “racial discrimination is not inherently racist” and Kendi’s definitions of these things (and here I’m quoting directly from Kendi’s book):
The student pointed out that by almost every metric Kendi is concerned about (wealth, income, educational status, criminal conviction and incarceration rates, life expectancy, etc) Asian Americans are by far the most successful racial group in America. So if racism is defined in relation to equity and equal outcomes, and being properly antiracist requires engaging in racial discrimination to bring about those outcomes, doesn’t antiracism require that we must racially discriminate against Asian Americans? Kendi just flat out dodged the question, though.
Oct 15 2021 at 9:57am
Don’t colleges and universities already discriminate against Asians?
Oct 15 2021 at 10:44am
Many do, yes. The curious thing is that people like Kendi refuse to acknowledge this kind of anti-Asian racial discrimination and affirm that, according to this theory, racially discriminating against Asians is a good thing. Kendi defines “racism” entirely in terms of equal outcomes, and says any policy which does not produce identical outcomes for all groups is therefore racist. But again, the awkward implication is that if you want to equalize according to racial groups, you really shouldn’t be particularly worried about white people, because in America white people are pretty middling as a group by all these measures. As usual, Scott Alexander was able to explain this well too:
Kendi is even more explicit in his book Stamped from the Beginning, where he says:
But if this is true, the it can’t also be the case that America is a “white supremacist” society, because white people aren’t at the high end of any particular metric when looked at by race. If the only possible reason for one racial group to do better or worse than another is racial discrimination, then you must believe that America heavily racially discriminates in favor of people of Asian ancestry. I think this is obviously nonsense, but if someone accept Kendi’s framework and definitions, the conclusion seems unavoidable.
Oct 14 2021 at 2:56pm
Who gets you judge what “aggregate utility” is, since it doesn’t exist?
Comments are closed.