Latest Harvard Study Example of Garbage In, Garbage Out

Last night, I loaded up Robert Stacy McCain’s site and found a rather poorly conducted study (warning, link is to a PDF document) he had written a post about.  I am planning to fisk both the study and the Atlantic article which first brought it his – and subsequently my – attention.  Because both need it.

Here is the final map from the study by county/parish

Stipulating that the data in the study is complete and accurate, and that everything in the analysis is legit — well, why is there a bright spot on the resulting map in the vicinity of Tuscaloosa, Alabama, but no corresponding bright spot near Athens, Georgia? Why does rural Arkansas look like a beacon of upward mobility, while the bustling economies of Atlanta and Charlotte produce no such effect?

Most of all, why does the map referenced by O’Brien show that impoverished Appalachia offers more opportunity for advancement than any of the more prosperous surrounding flatlands?

To use a social science term: Your data is obviously fucked up.

Stacy has the right of it here.  To quote Page 21, Paragraphs 1 and 2 of the study:

We permanently assign each child to a single CZ based on the ZIP code from which his or her parent led their tax return in the first year the child was claimed as a dependent. We interpret this CZ as the area where a child grew up. Because our data begin in 1996, location is measured 1996 for 95.9% of children in our core sample.

Does anyone else see a problem with this?  By permanently assigning a child to a single “Commuting Zone” – a roughly four county region surrounding a major city as defined by the 1990 Census (yes, you read that right, see Page 20, Paragraph 2 if you don’t believe me) – you are unable to account for geographic mobility.  What’s more, as the above portion notes, a child is assigned to this CZ based on the first tax return they are claimed on.  Even if the child only lived there for a single month.

Now, to be fair, they do point out that 83.5% of children in the United States are raised the same CZ they are counted in, and only 38% live in a different CZ after reaching adulthood.  However, based on demographic trends the Census bureau has been taking (and much to the annoyance of my friends and fellow Kansans out in KS-1, KS-2, and KS-4) the majority of the remaining 17.5% are most likely not moving between various urban areas, but rather moving out of rural areas.  The rate of this movement has increased since 2000, to the point many rural State House and State Senate districts in rural areas both in Kansas and elsewhere have increased in size by almost half as more seats shifted to urban areas.  Further, that 38% (just under 2/5 of the population) will have a profound effect most on those moving out of increasingly poor and abandoned rural towns for urban areas.

The study points this out, in fact, on page 25, “44.6% of children who gr0w up in rural areas live in urban areas at age 30.”  As I mentioned previously, this has a profound effect on the home mobility stats rating, and helps explain why rural areas in the Great Plains have the highest absolute income mobility according to this survey.

This has a major effect.  Let’s take a hypothetical based on the beige colors of West Virginia up there.  A man is a coal miner on a seam which is played out.  Having recently been laid off, he reports a very low-income the year of 1996 (the first year of the survey).  That become locked in as their basis.  Not long after, he gets a job at a local manufacturing firm (perhaps a Toyota factory has opened up in Tennessee) and moves his family there.  His kids are now in better schools, and have better prospects themselves.  In addition, recognizing his attention to detail, the man soon gets promoted off the line and into a supervisory role.  His kids, seeing the hard work does pay off, his kids put more effort into their school work.  2012 rolls around, and the kids have graduated college.  One is a lawyer, the other an engineer, and are both making money.  They are also still counted as living in a county they spent little time in.

And yet, even with the faulty methodology, Matthew O’Brien over at the Atlantic still feels this gives him the authority to proclaim the following.

So what makes northern California different from North Carolina? Well, we don’t know for sure, but we do know what doesn’t. The researchers found that local tax and spending decisions explain some, but not too much, of this regional mobility gap. Neither does local school quality, at least judged by class size. Local area colleges and tuition were also non-factors. And so were local labor markets, including their share of manufacturing jobs and those facing cheap, foreign competition. But here’s what we know does matter. Just how much isn’t clear.

1. Race. The researchers found that the larger the black population, the lower the upward mobility. But this isn’t actually a black-white issue. It’s a rich-poor one. Low-income whites who live in areas with more black people also have a harder time moving up the income ladder. In other words, it’s something about the places that black people live that hurts mobility.

I’m going to stop you right there.  This occurs for the same reason the Boston Federal Reserve found blacks had a more difficult time getting housing loans in the study whose botched report led to the Community Reinvestment Act and the Great Recession – blacks are more likely to have poor credit and low-income.  I do not know why – though I suspect the destruction of the Black Family by LBJ was a major component – but it is a fact.  Communities with lower-income are more likely to be on the doll, and that increases the likelihood one will remain in poverty and upon the doll until the day they die.

You don’t have to like it, but you do have to accept it.

2. Segregation. Something like the poor being isolated—isolated from good jobs and good schools. See, the more black people a place has, the more divided it tends to be along racial and economic lines. The more divided it is, the more sprawl there is. And the more sprawl there is, the less higher-income people are willing to invest in things like public transit. That leaves the poor in the ghetto, with no way out for their American Dreams. They’re stuck with bad schools, bad jobs, and bad commutes if they do manage to find better work. So it should be no surprise that the researchers found that racial segregation, income segregation, and sprawl are all strongly negatively correlated with upward mobility. But what might surprise is that it doesn’t matter whether the rich cut themselves off from everybody else. What matters is whether the middle class cut themselves off from the poor.

Seriously, Mr. O’Brien, you are blaming racism here?  Do you know why folks in Kansas City don’t want mass transit?  Because as much of a drain on public coffers in Chicago and D.C., it would be even more of one here.  It would be cheaper for the Johnson County Board of Commissioners to issue every citizen a debit card which could be used only with certified taxi companies than it would to continue running the JO.  This is not hyperbole.  The only mass transit system in the world which turns a profit, is the MTR Corporation in Hong Kong (oh, the irony).

As to the Bad Schools, that is not a funding problem.  Kansas City Missouri Public Schools received $15,000 per pupil in 2012.  Across the state line, the Shawnee Mission school district received $12,041, and it only cost the Diocese of Kansas City-Saint Joseph $5,000 per student.  And yet, in spite of receiving the most spending per student in the area, the KCMO school district has gone unaccredited for the second year in a row.  Washington D.C. and New York are similar, and yet a recent study showed a majority of graduates from New York’s public schools were functionally illiterate.  The problem is not spending, the schools get plenty of money, it is the culture which emerges in the area.

You know, there was a time when the Black Side of town was just the place where blacks lived.  During this time, blacks were well underrepresented in the criminal justice system (even allowing for racist Democrats and Segregationists arresting blacks on sight when a crime had occurred until evidence was found proving them innocent).  During this time, the Black Middle Class was growing, and the statistical nuclear family was black-skinned.  Then LBJ’s War on Poverty began.  Now, I don’t know for a fact his war on poverty is the cause of black poverty in the United States, but LBJ certainly hoped to get them addicted to the doll and thus become faithful voters for the DNC.  Perhaps this had the unintended side effect of destroying social mobility for blacks in this country?

3. Social Capital. Living around the middle class doesn’t just bring better jobs and schools (which help, but probably aren’t enough). It brings better institutions too. Things like religious groups, civic groups, and any other kind of group that keeps people from bowling alone. All of these are strongly correlated with more mobility—which is why Utah, with its vast Mormon safety net and services, is one of the best places to be born poor.

There is a problem here, but I will let Arthur C. Brooks answer that one.  You should buy the book, its good.

4. Inequality. The 1 percent are different from you and me—they have so much more money that they live in a different world. It’s a world of $40,000 a year preschool, “nanny consultants,” and an endless supply of private tutors. It keeps the children of the super-rich from falling too far, but it doesn’t keep the poor from rising (at least into the top quintile). There just wasn’t any correlation between the rise and rise of the 1 percent and upward mobility. In other words, it doesn’t hurt your chances of making it into the top 80 to 99 percent if the super-rich get even richer.

Mr. O’Brien, I have to ask.  [Annoyed Sisko voice]Did you even read the study Chief?[/Annoyed Sisko Voice]

But wait, he does get one thing right.

5. Family Structure. Forget race, forget jobs, forget schools, forget churches, forget neighborhoods, and forget the top 1—or maybe 10—percent. Nothing matters more for moving up than who raises you. Or, in econospeak, nothing correlates with upward mobility more than the number of single parents, divorcees, and married couples. The cliché is true: Kids do best in stable, two-parent homes.

And, as I pointed out previously, the black and poor-white nuclear family which had once been the statistical image of the American family was shattered at the same time as LBJ’s great society was put into place and our nation’s problem with a growing addiction to the doll began.  Which might be one of the reasons that poor city areas with high black populations are as poor off in terms of income mobility as poor whites who remain in rural areas.  Those currently tend to be the areas where the problem is most pronounced.

Now, there is another important factor here which needs to be addressed.  The Chief here has only spoken on the mobility from the bottom 20% to the top 20%.  How does mobility as a whole look?  Take a look at this chart.

Only a maximum of 1:3 children will remain in their parents tax bracket by adulthood

The chances of remaining in the same Quintile of Income Earners in the United States as one’s parents are less than 1:3.  We may not have the social mobility of Denmark, but we are also a far larger nation with a far larger pie, meaning you have to work harder to get a smaller share.  But even a small increase in one’s share in the US is a greater increase than you would get in Denmark.

And that map, the one showing absolute changes across the entire nation?  Here’s what it looks like when you chart relative changes in comparison to one’s neighbors.

DNC cities aren’t looking so good now…

Wow, that is a much different map.  Suddenly, the Democrat controlled areas got a lot darker.  Hmm, I wonder why…

Advertisement