Disparate Impacts

There’s a new Uncommon Knowledge interview with Thomas Sowell, with the release of a new book—his fifth since turning 80. The book is called Discrimination and Disparities. I’m glad he keeps writing, even though he’s no longer doing a weekly column; it’s never enough.

Thomas Sowell, Uncommon Knowledge, May 3, 2018

Much of the discussion is about the term disparate impact, which Wikipedia defines this way: 

Disparate impact in United States labor law refers to practices in employment, housing, and other areas that adversely affect one group of people of a protected characteristic more than another, even though rules applied by employers or landlords are formally neutral.

It’s a controversial idea, because it’s ostensibly about fairness while being literally unfair. And it introduces into law the idea that a person with no intention of discrimination is held accountable as if there were mens rea (intent to commit a crime) based on outcomes that simply appear discriminatory.

Interviewer Peter Robinson starts the discussion with a quote from the book:

The disparate impact standard represents a major departure from American legal principles where the burden of proof is usually on those making the accusation.

Disparate impact is a tool used in cases of employment, housing, and credit, among others. And, for the sake of setting up the discussion Robinson asks,

the latest book by
Thomas Sowell
cover image from Amazon

But, Tom, what about the notion that we need a disparate impact test because discrimination, particularly racial discrimination, particularly against African Americans, is so deeply embedded in the fabric of this country that people discriminate all the time without even being aware of it?

And Thomas Sowell answers:

If you're going with that assumption, then you don't need the disparate impact theory. You just simply say what you've just said. To dress it up as the disparate impact theory, the disparate impact theory depends upon the truth of the assumptions.

What assumptions?

This implicit assumption that all of the groups are very similar in their capabilities, what they want to do and so forth. When you look at facts, you find disparate impacts everywhere.

For example, he looks at three groups of immigrants: Irish, Jewish, and Italian. Disparities depend on what you’re looking at:

If you look at things like politics, the Irish were so far more advanced politically than either the Italians or the Jews that for generations you had Irish politicians representing neighborhoods that were overwhelmingly Italian or Jewish.

Many disparities are natural. Like tornadoes.

You find 90% of all the tornadoes in the entire world occurring in one country, namely the United States. And only in a part of the United States. You don't hear about tornadoes in Maine or in the Pacific Northwest. So think how much land area there is in the world, and 90% of them right in this one little place.

If disparities are not an inherent evil, but simply part of nature, do we accomplish something good by adjusting to get rid of the disparities? Or do we accidentally cause some other problem?

Maybe we should start by questioning the assumptions. Are differences necessarily the result of bigotry? Or are some differences the result of personal choice and ability?

Sowell’s book refers to a 50-year study of 1500 people with IQs in the top 1%.

What I point out in the book is that the disparities within that narrow range, the top third, for example, had more than 10 times as many post-graduate degrees as the bottom third, among people who were all in the top 1%. So there were obviously many other things that had to come together. The other thing was that two people who failed to make the 140 IQ cutoff ended up getting Nobel Prizes in physics. There's nobody among these 1,500 that did, so obviously there have to be a lot of things coming together.

So intelligence is not the differentiating factor. What is? Family background.

It doesn't matter how much brain power you may have, if you're not raised in a home where people are thinking, where they're doing intellectual things, you're not in the same position as someone with the same IQ who's in a family that has that kind of background.

I talked about that in the last post. It’s not really about victimization, or a victimized underclass. It isn’t even much about innate ability. It’s about family.

Government interference—on the assumption that society is doing something evil, such as bigoted oppression—doesn’t get the positive results it claims to want, because it isn’t addressing the actual problem.

Dr. Sowell even gives examples of societies where there is actual racism, and government interferes by enforcing that racism, and yet people flout the law when it makes economic sense to do so. It happened in Apartheid South Africa. More black workers were hired than white, even in jobs where it was illegal to hire blacks. More blacks than whites lived in white-only neighborhoods.

Quoting the book, “Black incomes in 1900 were almost half again higher than they had been in 1867 to ’68.” So, in the American South, after the Civil War, where employers had agreed to suppress earnings for black workers and sharecroppers, black earnings had a higher rate of growth than the American economy overall.

The market is more powerful than racism, or any other bigotry.

Dr. Sowell explains this related to minimum wage laws:

If you have a minimum wage, and that's set above where it would be in a free market, then that means you're going to have more people applying because there's a higher wage, and there are going to be fewer people hired because of the higher wage. So you're going to have a chronic surplus of applicants.

Now, in a market where there's, say, a chronic surplus of qualified people of, say, 200, and there a hundred blacks, for example, who are qualified, then if the employer refused to hire all hundred black qualified people, he still has 200 others he can call on, and that's it. And it's cost him nothing. But if there's no minimum wage now, and there's no chronic surplus, every time he turns away a qualified black person, he has to have someone who's not black who's also qualified that he can hire….

[And he may not be able to find that person] at that price. Therefore, the price will have to go up, so it's costing him. And if he doesn't raise the price, he's going to have to keep his customers waiting because he doesn't have enough people to do the job.

In short, minimum wage laws are going to have an intensifying effect on existing racism, while free markets tend to lessen discriminatory hiring.

Another example of unintended consequences is housing in northern California, where home building is restricted in the name of saving the environment. The result, of course, is higher housing costs. And then officials ask, “What can we do to get more affordable housing?” The obvious answer is, build more houses. Instead, they appoint a committee to study the issue. Dr. Sowell quips,

It's like appointing a blue-ribbon committee to go out there and find out why the ground is wet after the rain. I think it's almost miraculous the way they can avoid the obvious.

So, what is the disparate impact of the higher housing costs resulting from government imposed home-building limitations? Blacks, among others who started out lower income, instead of moving up in the economy, get priced out—what Peter Robinson calls a sort of “soft version of Jim Crow.”

The progressive/liberal/tyrannists tend not to apply their disparate impact rules on their own “good intentions,” however. Because they want to keep feeling good about their intentions regardless of the harm they do.

Here’s another quote from the book:

The plain fact is that the black poverty rate declined from 87% in 1940 to 47% in 1960, prior to the expansion of the welfare state that began in the 1960s under the Johnson administration. There was a far more modest decline in the poverty rate among blacks after the war on poverty began.

He lists several more disproportionately negative effects for blacks. Quoting the book again:

In the United States, murder rates, rates of infection with venereal diseases, and rates of teenage pregnancies were among the social pathologies whose steep declines were suddenly reversed in the 1960s. Nowhere was rampant violence and other social pathology as common among low-income people in the first half of the 20th century, when they were more deprived, as in the second half, when the welfare state had made them better off in material terms.

Since we’ve been talking about sex education in schools recently, let’s look at that. Sex education was implemented in schools in response to disease and teenage pregnancy—both of which were going down on their own. By 1960 the rate of venereal disease was half what it was in 1950. Then they started teaching it in the schools, and rates rose dramatically, as did teenage pregnancy. And that hit blacks particularly hard.

In 1960 two-thirds of black children were being raised in families with both parents. By 1995, two-thirds were born out of wedlock. Today 85% of black children live in single-parent households. And that’s while, in some places, like New York City, more black babies are aborted than are born.

We know—the data is astoundingly clear on this—that children fare best in homes where they are raised by their two parents. The results of failure to these children is calamitous.

As Thomas Sowell puts it,

It’s not the legacy of slavery that destroys the African American family. It’s the legacy of the welfare state that destroys the African American family.

And, if you want to destroy or oppress a people, the simplest way is by destroying the family—the basic unit of civilization.

Did they do it on purpose? I’m not willing to go that far. But, if they had intended it, they couldn’t have done a more thorough job.

Dr. Sowell puts it this way:

For one thing, it [the welfare state] makes it unnecessary for fathers to support their offspring. And in fact, it makes it counterproductive in many cases. A very poor man who might be able to support his family realizes his family will be better off without him. But on the other hand, someone who's strictly irresponsible, either the man or the woman or both, now pays no price for being irresponsible. The taxpayers pay the price. And actually, the harm done to the taxpayers, which is serious, still is not comparable to the harm done to the families, especially the children.

Thomas Sowell is an economist; he’s about data. Let the data take you where it takes you, and then see what conclusions you need to draw. You don’t start with a premise and then dismiss data that doesn’t fit. Sometimes there’s an outlier, but sometimes it’s a clue that your original assumptions are wrong.

Disparate impacts—or simply differences—are not necessarily the result of a nefarious plot even so secretive that people don’t know they’re thinking it up. The simpler answer might be that there are a lot of factors—including personal choice—that go into various differences.

But allowing for freedom, and free markets, along with encouraging strong families—that gets us to better, more positive outcomes overall, than interference gets us, no matter how well intentioned.

Maybe we ought to switch it around: hold government accountable for negative impacts of their interference, and put a stop to it. And stop holding individuals and businesses accountable for discrimination that was never intended, thought, or practiced.

Disparate impacts, then, are really just a narrow, limited view, of one portion of a larger picture that, if you could see the whole picture, would become nothing worth acting on.

How do you know whether a particular policy is one of those, or something really worth acting on? Try applying this Spherical Model axiom about unintended consequences:

Whenever government attempts something beyond the proper role of government (protection of life, liberty, and property), it causes unintended consequences—usually exactly opposite to the stated goals of the interference.