Profile of Raj Chetty

By Gareth Cook, interesting and excellent throughout, here is one good bit of many:

For example, the strongest correlation is the number of intact families. The explanation seems obvious: A second parent usually means higher family income as well as more stability, a broader social network, additional emotional support, and many other intangibles. Yet children’s upward mobility was strongly correlated with two-parent families only in the neighborhood, not necessarily in their home. There are so many things the data might be trying to say. Maybe fathers in a neighborhood serve as mentors and role models? Or maybe there is no causal connection at all. Perhaps, for example, places with strong church communities help kids while also fostering strong marriages. The same kinds of questions flow from every correlation; each one may mean many things. What is cause, what is effect, and what are we missing? Chetty’s microscope has revealed a new world, but not what animates it—or how to change it.

Here is the full piece.


Chetty says that these are "just" correlations but then often veers into implicitly suggesting causal relationships. This is something economists don't generally tolerate but somehow Chetty gets away with it..

Yes, Chetty gets away with a ton of things that regular economists without all the gold nameplate schools and backers would never in a million years get away with. It's amazing that he's never had to answer the criticisms that Steve Sailer has brought up.

Here's my 2015 critique of Chetty's fascinating database on trans-generational income mobility by county:

I found his results highly plausible: e.g., the worst county in the US for 25th income percentile parents to have raised their kids in 1996-2000 by the kids' income in 2011-2012 was the county coterminous with the notoriously tragic Pine Ridge Sioux Indian reservation. The best county, ironically, was Sioux County, Iowa, a prosperous and conservative farm county that's highly Dutch and Protestant.

But I ended with 3 problems he still faced:

– Regression toward the mean (especially among races)
– Temporary booms and busts
– Cost of living differences.

For example, this Atlantic article talks a lot about Charlotte, NC's low upward mobility for income. But some of that is due to the years Chetty is investigating: North Carolina has a nice long boom before 2008 and attracted a lot of people for jobs in North Carolina's prime industries such as home construction, retail banking, mortgages, golf course construction, furniture-making, and hardwood lumber.

In 2008, all those very North Carolinaish industries got hammered.

So, when he looks at how well parents in North Carolina were doing in 1996-2000 versus their children in 2011-2012, not surprisingly Charlotte, NC looks like a pit of despair.

In contrast, the northern Great Plains were hemorrhaging jobs and people in the 1990s, but in the 2010s, due to energy and China-driven resource booms, they're doing very well. So, Chetty is assuming they must have a lot of innate "opportunity," but some of it is just regional cyclical good fortune based on this pre and post periods.

One interesting aspect of this is that my critique about Chetty ignoring regional booms and busts would appear to be not at all politically incorrect, yet

For example, just compare how his methodology treats a couple of high white counties: Sioux County, Iowa and Horry County, South Carolina. The former came in the best in the country for working class transgenerational upward income mobility, and the the latter was the whitest county in the Worst 25 in the country.

Horry, Just across the North Carolina stateline, is home to the vast Myrtle Beach golf resort complex of about 90 course, last I checked. In Chetty's base period of 1996-2000, Myrtle Beach was booming due to golf course construction for the 1990s Golf Craze. In 2011-12, Horry was as dead as the golf industry.

I don't think that's politically incorrect to point out that Chetty's methodology treats local ups-and-downs due to economic cycles, such as golf going up and going down in expected popularity, as evidence for innate differences in "social capital."

That example of a problem with Chetty's methodology should be usable by anybody, no matter how fearful of the PC mob, but perhaps because I pointed it out first, Chetty's unjustified conclusions remain apparently off-limits to even technical criticism.

I'm pleased Cowen linked this, as my link in an earlier comment is likely ignored. Chetty makes a big deal of social capital: simply throwing money at the poor doesn't work long-term (i.e., there's little if any resulting mobility), it takes building social capital. I'll repeat the sentence from the profile that should have significance to fans of this blog: "Chetty thinks about revolution like an economist does: as a compounding accumulation of marginal changes." Cowen could not have said it better himself. Read the profile.

Is social mobility really a myth. Russ Roberts disagrees and takes issue with Chetty's methods.

I have no love lost for Chetty, but I'd take him over a worthless dilettante like Roberts any day.

I'd take Chetty over Steve Sailer, who seems oddly jealous for some reason.

Of course, Sailer is jealous. Chetty gets money and awards. Sailer gets grief.

The important question is who is right.

I'm not really interested in rooting for particular academics over others, other than based on the merit of their work. I do think Roberts raises a number of serious concerns about the recent work on social mobility that have been largely overlooked.

Don't let other comments dissuade one from reading the profile. Chetty most definitely is not a class warrior. What he tries to identify in his research is why some rise up and others don't: it's social capital that is the key. Again, throwing money at the poor alone won't work.

It's almost as if traditional values that have survived the test of centuries of adversity across multiple cultures have a role to play in success of people and their communities. Good thing we have linear regression to tell us this.

It helps the survival when the values themselves create a lot of the adversity.

As opposed to what?

thou shall not steal

Kipling’s The God’s of the Copybook Headings.

We used to at least try to pass this distilled experience along to the next generation.

I tell people if you're not going to read a religious text, then read Kipling.

I found "upward mobility was strongly correlated with two-parent families only in the neighborhood, not necessarily in their home" to be most unclear. We wouldn't expect a lot of variation in two- vs. single-parenthood *in the home*, so we wouldn't expect a strong correlation.

It turns out what Chetty et al. found makes a lot more sense (p. 29,

"Consistent with prior work on family structure and children’s outcomes (e.g., Sampson 1987), we find a strong negative correlation of -0.589 (s.e. 0.003)
between the fraction of single-parent households in a tract and upward mobility. Importantly,
this correlation remains similar even conditional on the marital status of a child’s own parents.
In particular, children of single parents have higher rates of upward mobility if they grow up in
a neighborhood with fewer single parent households (correlation = -0.52). This implies that the
correlation is driven not by differences in outcomes between children raised by married vs. single
parents, but rather by ecological (neighborhood-level) factors."

Chetty et al. are correct that house-level single parenthood therefore does not account for the positive effect of being in a low-single neighborhood. But that doesn't mean that the effect of being in a two-parent household is nil! "Not necessarily in their home" makes it sound like there is evidence against the effect, but the effect just isn't reported!

The neighborhood impact makes sense to me, due to likely impact in the education and play setting. Lots of studies show the negative impact of single parent households (predominately single mothers/grandmothers) on discipline and outcomes. If the son of a single (grand)mother is in a classroom where most of his peers and role models have a father and therefore are more likely to face consequences, then it’s a reasonable hypothesis that single mother boy gets less peer support for poor behavior. Mothers (on avg) struggle to provide discipline to sons.

Fewer fatherless boys in the neighborhood = less opportunity for gangs to develop.

More motherless goats = more goat stew for everybody

Chetty and Hendren (2015) write:

"This suggests that there are pockets of places across the U.S., like Baltimore MD, Pima AZ [Tucson], Wayne County (Detroit) MI, Fresno CA, Hillsborough FL [Tampa], and New Haven CT, which seem to produce especially poor outcomes for boys."

New Haven County is a fine place to live if you have daughters and you are a Tiger Mother professor at Yale Law School, but it’s a terrible place to move to if you have poor black sons. Chetty has no data on what percentage of boys who were moved to Baltimore, Detroit, or New Haven weren't earning much in 2011-12 because they were in jail, but it’s obviously a considerable risk.

In contrast, Tucson, Fresno, and Tampa were all home construction boomtowns that got wiped out by the bursting of the Housing Bubble in 2008, a memorable cataclysm whose effects on his data Chetty doesn't seem to have pondered.

Conversely, girls whose parents moved them when they were teens in the 1990s to now booming and low crime Manhattan are likely to pay a penalty in terms of lower family income in 2011-2012 because they are less likely to be married than if they had been moved to Salt Lake City.

You can lead a horse to water but you can't make him drink.

One more reason to abolish the Ivy League.

I'm trying devise a metaphor for this massive display of intellectual back-slapping.

I mean: WOW!

Chetty and his savants balance (Moses above the Red Sea comes to mind) on the precipice of (command and control) bringing 340 million people to upwardly mobile themselves into the New Earthly Paradise - all on his own, 40 year-old, innate genius.

I don't want to give a metaphor that mirrors pornography, like visualizing this as an intellectual daisy chain.

It's more mental masturbation.

"The attempt to make heaven on earth invariably produces hell." - Karl Popper

Sometimes it rains because there is steel, but the truth about snowstorms is that there are snow leopards.

At night the ice weasels come.

Cracks me up how the untermenschen now have a problem with smart people being smart, in and of themselves. Just doing research on something is now a bad thing to them.

And the deplorables wonder why we have all the money and control the media. LOL.

Your elite are pathetic last men who have secularized salvation into "science" to produce banal findings.

Ich bin ein untermensch.

Are you elite who? Jeff Bezos? Alexandria Ortega-Castro? Ilhan Omar? Elizabeth Warren? Jeffrey Epstein? Hillary Clinton? Bill Clinton? Barack Hussein Obama? Rev. Al Sharpton? Joe Biden? Maxine Waters? Corey (Spartacus) Booker?

FYI - The few people that listen to/read the fake news are misinformed. And, I have plenty of money in cash and equivalents, stocks and bonds, real estate (multiple commercial parcels and residences), precious metals, no leverage.

Anyway, America survived eight years of community organizers, PhD's, professors, elites, politicians, racial racketeers, America-hating black/brown nationalists that never did anything ruining America. In 2016, we took a sledge hammer to it. His name is Donald J. Trump.

Yeah he's elite too.

"(Swoon.) Ahhh, Harvard will save us! Harvard will discover an algorithm for 'stratified egalitarianism'--one 'equality' will suit the aspirations and abilities of poor people Harvard otherwise has no use for, the other 'egalite' will continue to enforce the privilege and the pay accorded to Ivy League elites."

Does Harvard U. qualify as an exponent of egalitarianism or has its career as a poor victim of an egalitarianism it cannot believe in or consent to already begun? COULD the South Side of Boston swarm Cambridge and Brookline one fine day soon?

Harvard was home to John Rawls, an ardent egalitarian who managed to justify Harvard's privileged status with his maxi-min principle (Chetty is concerned about making life better for the least advantaged among us!). Or as the late Allan Bloom put it:

"I thought of the old joke about the farmer who hears a thief in the chicken coop. Substituting the Harvard Coop, I imagined myself yelling, 'Who's in there?' and getting the answer, 'There's nobody in here but us anti-elitists!'"

Spontaneously forced upon my imagination is this vision of the deputization of the faculty and students of the Sociology Department at BU by the chosen delegates of the Greater Boston People's Party with the aim of commandeering and securing all bridges spanning the Charles, perhaps more intent to keep nefarious influence out than eager to facilitate pillage, plunder, loot, or burn along the river's north shore.

Faculty are far too self-important to do such mundane tasks, which they leave to their students and indentured servants...err...teaching assistants.

Yet children’s upward mobility was strongly correlated with two-parent families only in the neighborhood, not necessarily in their home. There are so many things the data might be trying to say. Maybe fathers in a neighborhood serve as mentors and role models? Or maybe there is no causal connection at all. Perhaps, for example, places with strong church communities help kids while also fostering strong marriages. The same kinds of questions flow from every correlation; each one may mean many things. What is cause, what is effect, and what are we missing?

Middle and upper class people tend to cluster in middle and upper class neighborhoods, and their children are acculturated into the norms and values of the people around them. This is hard to figure out?

No, it’s just politically inconvenient to admit there is a dosage effect.

But it is easy to adulterate when it comes to race. Black people are shitty because they have low IQ, it can't possibly be due to the norms and values of the people around them.

It is mostly because of the people they have around them, namely, other black people who have allowed their customs and habits to degenerate to a state worse than under slavery. But somehow "white supremacy" managed to cause that degeneration only AFTER the end of slavery and segregation. If it took whites that long to figure how to do that to blacks, whites must have even lower IQs.

Segregation ended? Citation needed.

But sure, how society treated black people after 1865 had nothing to do with how their customs and habits formed. #whitegenocide


Surrender accepted.

Gheyer still.

You don't have to keep surrendering.

Yes, segregation ended. Jim Crow is gone. Don't bother with so-called de facto segregation.

And what sort of fool thinks that whites were somehow unable to break up the black families during slavery and segregation to the extent that we see today under far better legal, economic, and social conditions.

Nearly all 'experts' never consider that it's not the values that change the people, perhaps it's that certain people are attracted to those values intrinsically. Those people like all peoples tend to gather together in similar communities when possible. Perhaps nothing can be done to improve the lot of some.

It would be useful to discuss this in an environment where race isn't a factor (although the data sets Chetty uses may not be available).

I'd suggest a starting point is Theodore Dalrymple's Life at the Bottom. All the social and cultural pathologies often attributed to race, or racism, are here in the largely white English underclass.

There is a bit of a summary here:

See also Sowell's "Black Rednecks and White Liberals." The latter need to shut up and do nothing for a couple of generations.

Maybe divorce is an incidental event when it comes to children’s mobility. Parents in a good neighborhood have the traits which allow them to impart social mobility onto their kids. Those same traits are correlated with a lower divorce rate. If divorce happens, it’s more likely to be due to a genuine unforeseen personality mismatch between parents rather than fatal character flaws (low social capital) which adversely impact both parent and child. The latter scenario is more likely to cause divorce, which explains the co-occurrence of a high divorce rate and low social mobility. This effect is boosted by the fact that parents with high social capital are more likely to continue to support their kids after divorce. Moreover, the whole village steps in to make sure a “good kid” doesn’t go to waste. Quotes added to highlight how depressing this is.

Affirmative action is just window dressing. The true solution is a systematic uplift of communities with low social capital via a partnership between local and federal government:

- Increase funding for poor schools, either by diverting property taxes from wealthy zip codes or tapping into federal funds
- Increase school hours and extracurricular activities. Include the sort of moral instruction that “high social capital parents” give to their kids. Try to involve kids’ parent(s) as much as possible.
- Legalize every drug that causes as much or less harm than alcohol, to decrease gang activity
- Fix the proverbial “broken windows” in neighborhood cores and public institutions
- Increase recruitment of police from the neighborhoods they patrol, then step up crime enforcement. Never agree to civilian oversight of police, but definitely listen to community activists and try to find a happy medium. Make sure all police interactions are recorded, and that the recordings are released as freely as possible. Encourage greater taser use where appropriate.
- Scale sin taxes to account for the increased damage they cause in bad vs. good neighborhoods. The liquor tax should be some multiple higher, maybe 2x ore more, in bad zip codes. This would both discourage the operation of liquor stores and casinos in places where they don’t belong, and discourage locals from patronizing those businesses where they exist.
- (Controversial) Ban music glorifying violence from the radio. Ban commercial distribution of music with explicit gang references. There’s a case to be made that this wouldn’t stifle constitutionally protected free speech, but I’m open to less risky solutions that don’t involve banning. Humans are social creatures. If you’re a kid in a “bad” neighborhood, those songs aren’t ironic fun like they are to an edgy suburbanite. They depict your reality, and then perpetuate it as they glorify something that you’d otherwise be inclined to hate.

The list could go on, but I think that’s a good start.

> The true solution is a systematic uplift of communities with low social capital via a partnership between local and federal government:
> - Increase funding for poor schools, either by diverting property taxes from wealthy zip codes or tapping into federal funds
> - Increase school hours and extracurricular activities. Include the sort of moral instruction that “high social capital parents” give to their kids. Try to involve kids’ parent(s) as much as possible.

More flight of fancy and wasting other people's tax money. The New York city "Renewal Program" involved pouring USD$773 Million into 50 "borderlined" schools for 5 years, i.e. on average USD$ 3.09 Million per school per year on top of the usual school budgets. The reason for the "Renewal Program" is that very few Black and Hispanic students were selected into the "Elite8" special high schools by the objective "SHSAT" standard test and the cause is alleged to be the socio-economic status of the students and school. The money goes into providing special remedia teachers, education psychologists, school counsellors, career advisors, school equipments and facilities, special test prep classes, free nutritious lunch, longer school hours, mentors, etc, etc.

As these are "borderlined" schools some might have been closed down during that period and had been replaced by some other schools. I can only obtain data on 24 schools from the NYTimes feeder school data. Although there could be many fringe objectives, the unsaid main target is how many of those students can get into the "Elite8". The tester number weighted average percentage of Black and Hispanic students in the 24 schools is 95.96%. Not all students had judged themselves to be academically strong enough to sit for the SHSAT. For the 24 "Renewal" schools with data, the number of SHSAT testers per school for 2018 ranges from 0 to 45, totalled 270 testers, thus the estimated number of testers for the 50 "Renewal" schools is 563. The number of students from the 24 "Renewal" schools selected by the objective SHSAT for the Elite8 for 2018? Big fat zero.

MinTested | AvgTested | MaxTested | Nschool | SumTested | EstTotalTested | AvgPctHisBla
7 | 16.88 | 45 | 24 | 270 | 563.0 | 95.96

The independent RAND report:

The report, prepared by the RAND Corporation, is the most rigorous study to date on the effectiveness of the mayor’s turnaround program, which cost $773 million and has been widely criticized for producing disappointing results. The city recently announced the program will not continue.

First some bright spots: The turnaround program resulted in students attending school more often. Renewal boosted daily attendance by 1.5 percentage points and led to a 5 percentage point reduction in chronic absenteeism, defined as missing 10 percent or more of the school year. The initiative also led students to accumulate more credits in high school, a sign that students were more poised to graduate on time.

But many of the report’s findings are not as positive, often pointing to only slight improvements that were not statistically significant. The program did not produce clear effects on graduation rates or test scores. Worse, the program slightly raised high school dropout rates. And the researchers, looking at variables such as curriculum, “high expectations,” and teacher teams and leadership, did not find any clear positive effects on instruction or school culture.

The consequences of the waste, i.e. estimated at USD $274,600 per effective student per year (i.e. est. 563 sat for the SHSAT but none seemed to have qualified for the Elite8 SHS),

"‘Everything is on the table’ — Carranza talks budget cuts to NYC schools at City Council hearing"

The Renewal Program was flawed from the start. De Blasio felt sympathy for failing schools rather than just their students. NYC should have just closed the worst schools. Flooding them with stupid money was obviously not going to do anything. I’m for an iterative, well-designed, empirical overhaul of urban schools. If one strategy doesn’t work, we shouldn’t just give up and conclude something insidious about the students.

"Flooding them with stupid money was obviously not going to do anything."

It might not have helped the students, but it kept the teachers at the schools and the teachers union happy.

Did you read the report??? They are reporting on a program that is just two years old, as they stressed.

But, more importantly, here are some of the conclusions:

"In conclusion, we attest that through our rigorous quasi-experimental methodology, the New York City Renewal Schools appear have some positive impact on important leading indicators of attendance and chronic absenteeism among students at all grade levels. These effects grew over the years of implementation, with some additional outcomes becoming significant in the second year. For high schools, the effects are stronger in schools with the highest levels of economic needs....

In spite of these limitations, we believe that the evidence regarding the positive impact of the RS program on attendance outcomes is very strong. Other possible impacts, such as the positive impact on credits earned among high school students and the differential impact based on schools’ level of economic needs, suggest that the program is having a positive impact on lagging indicators of academic success, particularly among the most vulnerable students."

These are quotes from the conclusion section.

Did you ever read the report?

By the way, TrashDiver, I can see from your quote that you are quoting someone else's interpretation of the report, since the quote refers to Rand as a third party.

Thus, you misrepresented the authorship of your quote. Please provide the link to the article you are quoting.

If we don't have civilian oversight of police, who's watching the watchers? The police are civil servants, not a military division nor a monastic order. Of course they should be under civilian control!

Civilian as in non-bureaucratic. People have been calling for jury-style panels which dole out punishments on police for excessive use of force. That’s been tried, and it makes police soft on crime because they become so scared of a witch-hunt.

>There are so many things the data might be trying to say.

No, the data is saying that traditional families are the way to go.

There are however many other things that YOU want the data to say.

“Traditional families are the way to go” is not a very edifying comment because some people don’t have a choice in the matter. From a metaphysical standpoint, poor kids are born with such terrible resources that they are 80% destined for a broken life and to start another broken family (note: I’m using arbitrary numbers just to illustrate my point). Their choice to pursue a traditionally successful life occupies 20% of their decision space. They still feel agency with critical life choices, but in reality they were mostly destined to make bad decisions. Only exceptional intelligence, a good role model, and/or a stroke of luck can save poor kids from the vicious cycle encircling them. You don’t solve this issue by saying or implying “keep it in your pants, raise your kids, and go to work, you deadbeat.” This is a systematic problem which requires a systematic solution. See my comment above. If anything, Chetty’s study proves that divorce and a traditional family are incidental to a child’s mobility. Of course divorce is still bad, but in other ways.

Won’t wash. Almost everyone was poor a few generations ago, yet 80% don’t seem to have
been condemned to a broken life and poor decisions.

A lot of the people you’re talking about, like German, Irish, and Italian immigrants, had good social capital. If they were transplanted to the modern day, they would have climbed the socioeconomic ranks more quickly. That’s why immigrants from high social capital areas tend to fare well even if they are relatively poor when they come to the U.S. Indians, who came in waves as recently as in the 70’s/80’s, are really having a moment now.

I used “poor kid” as a proxy for “kid with low economic *and* psychosocial resources.”

"No, the data is saying that traditional families are the way to go."

Oh, God. The far-right is really getting desperate.

What does "traditional" even mean here? Two parents? A non-working mother? Male headship and submissive wives? The Amish? Polygamy?

Chetty’s research was fascinating but one major issue was how he measured household income as the sign of adult success in his landmark study. That measure depends very highly on how often and early people get married as married households no surprise tend to have higher incomes due to having twice the number of earners. As a result, the original Chetty study showed that urban areas, even wealthy ones like Manhattan, had the worst social mobility in the nation. I highly doubt kids who grow up in Manhattan are disadvantaged compared to ones who grow up in the Great Plains; their lower adult household incomes is probably due mostly or even entirely to the fact that they are less likely to be married as adults as their culture promotes later marriage.

The northern Great Plains are the motherlode of Magic Dirt, according to Chetty's research. That's partly because they've been doing quite well in this century due to the China boom (just as Australia has) and because they are quite white, and white people regress toward a higher mean income than do blacks.

The northern Plains have a significant Native American population. They are not lily white.

And you can see the big Indian reservations in South Dakota stand out in Chetty's maps as islands of downward mobility in a sea of upward mobility.

But, yes, Chetty's methodology tends to understate the life prospects of, say, 31-year-old Manhattan single women whose dads are paying their rents while they work at NGOs in low paying but respectable jobs while their fiancees make 6 or 7 figures on Wall Street. In contrast, Chetty finds 31 year old Salt Lake City women doing well in terms of family income, because such a high % of them are married.

Reminds of and seems to support Judith Rich Harris' idea that the kids that kids grew up around influence them as much as or more than their own parents (to include marital status or physical propinquity (and apart from genetics, obviously). So growing up around kids with two parents can be as good as having two parents yourself. So being in the right place at the right time can be conducive to desirable outcomes, it would appear.

An amusing aspect of this article is that the author keeps putting in details about what high IQ ancestors and siblings Raj Chetty has, as if he's trolling his readers with an Alternative Explanation.

That's not trolling. That's Straussian.

Will Chetty get the joke?

His parents and sisters might.

From reading the comments and criticisms, it seems that some who don't like his conclusions

Are making


The New


I'm a big fan of Chetty's data-gathering.

Chetty's analyses of his own data tend toward the inane, but I encourage anybody with a quantitative eye for pattern-recognition to take a look at what he has dredged up. I've learned a lot from it and so can others, as long as they don't take Chetty's descriptions of his findings on faith.

That said, Chetty's papers are less politically correct than the stories whipped up by journalists flacking for him, like the Upshot people at the NYT.

Nice, this article gives a partial answer to a question that I'd been wondering: how did Chetty get access to all that IRS data?

He evidently still operates under some restrictive research conditions. But I presume the data are now more accessible than they were before -- and to more researchers?

His Mobility Report Card provided an unprecedented look at colleges and income mobility, his research team presumably could add two or three years of additional data so it'd be nice if they updated that study.

Here's my analysis of Chetty's college data:

Highest median income graduates: MIT.

There's a lot of fun information in Chetty's databases, but he never, ever finds any of it.

It's called the welfare state that destroys families.

Comments for this post are closed