NYT Getting Chetty Wrong Again
Print Friendly and PDF

From the New York Times oped page:

An End to the Class vs. Race Debate


Ralph Richard Banks (@rrbanks), a law professor at Stanford, is the author of “Is Marriage for White People? How the African American Marriage Decline Affects Everyone” and is working on a book tentatively titled “Why Black Boys Fail.”

A new study rebuts a widely shared view that racial disparities in social mobility are economic inequalities in disguise — the belief that if we address class issues, we can fix racism.

The report, by the Stanford economist Raj Chetty, the Harvard economist Nathaniel Hendren and colleagues at The Equality of Opportunity Project, provides an empirical basis for an economic susceptibility that black parents like me have sensed: Across generations, we are less likely than whites to rise and when we do, are more likely later to fall. We seem unable to grasp or preserve economic gains as other groups do, including Latinos and Asian-Americans. …

Black families trace our economic insecurity in part to a gender divide that we see but often don’t discuss. We know that African-American daughters tend to do well. They climb the socioeconomic ladder as high as their white peers, if not higher.

No, as I point out in my new Taki’s column column, that’s a widespread misperception of the new study by Chetty and Hendren. What Chetty found was that parents’ family income predicts daughters’ average individual income about as accurately for both black and white daughters.

So, a class-based model works okay for predicting the individual incomes of black women, but that doesn’t mean that black women are doing as well as white women overall. They’re not.

Why not?

Because on average, of course, black women have parents who earn lower family incomes than white women, so, overall, black women tend to earn less as individuals, much less as families.

It’s the boys who fail.

What Chetty found was that black sons earn less as individuals than you’d expect from their parents’ household income relative to what white sons earn relative to their parents’ household income.

So a class model needs to be supplemented by a race model to predict the incomes of black sons.

This difference of a difference kind of thinking is tricky, so it’s not surprising that it’s been widely misunderstood this week.

… Black women may surpass their white counterparts in individual income, but they lag in household income.

These kind of sentences mislead readers. Black women do not on average surpass their white counterparts in individual income … unless you adjust, as Chetty does, for parents’ family incomes, which of course vary sizably between black daughters and white daughters. The average black household or family income in 2016 was $39k compared to $65k for whites, so the average black woman is expected to make less as an individual than the average white woman.

On the other hand, according to Chetty’s data, a black woman who grows up in a family earning, say, $100k in the 1990s, is expected to have an individual income at age 27-32 in 2014-15 about 1% higher than a white woman who grew up in a family earning $100k.

Of course, a far higher proportion of white women than black women grew up in families earning $100k.

So, if you know the household income of a woman’s parents, you don’t really need to know whether that woman is black or white to forecast her individual income. Yet, to forecast the woman’s current household or family income, it helps to also know her race because black women are so much less likely than white women to have high-earning husbands.

In contrast, a black son who grew up in a family earning $100k is likely to have a lower individual income than a white son growing up in a family earning $100k. To more accurately forecast just the individual income of sons, it helps to know the race as well as the parents’ income.

Why? See my Taki’s Magazine column for some obvious reasons.

I can see already that this is going to become a common misperception among elites, especially as Chetty and the New York Times reporters who wrote it up use the misperception to claim it disproves The Bell Curve.

[Comment at Unz.com]


Print Friendly and PDF