Glenn links to Stacy Mccain, who is fired up about this new Harvard/Berkekey study on income inequality and intergenerational economic mobility. I will revert to Mr. McCain momentarily but first let me deplore the lead sentence to the NBER working paper:
The United States is often hailed as the "land of opportunity," a society in which a child's chances of success depend little on her family background.
Says who? This is certainly a convenient definition for a team of economists who are embarking on a measurement of outcomes and hoping to pass it off as a measure of opportunity, but who really believes that parents have no influence on the success of their kids? Just by way of example, one might suspect that financial success is influenced by some combination of intelligence, good looks, height, good health, and high energy. All of these factors are subject to genetic inheritance, so one would not be surprised to see the child of successful parents having a bit of a head start. Other traits, such as self-discipline (aka impulse control) or a love of learning can be taught, but are more likely to be exemplified by parents who already possess those traits.
So, to pick an example almost at random, one might expect Sasha and Malia Obama to become successful women based on the genetic gifts and cultural values passed on by their parents. Who among us will attribute any of their future success merely to their parent's high income?
Gary Becker has lots on this; a snippet:
The relation between intergeneration mobility and meritocracy becomes still more complex after we recognize that earnings in a meritocracy would depend not only on cognitive abilities, such as IQ. For it depends also on investments in education and other human capital, on getting to work on time, on being able to take criticism, and on many other psychological characteristics. Families that are more educated and have high earnings tend to invest a lot in their children’s human capital, and in various non-cognitive traits. In a merit-based economy where earnings depend on the totality of abilities and skills, children of high earning parents would also tend to be high earners because their parents would pass on both cognitive skills and investments in various forms of human capital.
Even after including parental investments in education, non-cognitive traits, and other human capital of children, an economy where success and failure are determined by merit would still have low intergeneration mobility. To be sure, investments in education and many other types of human capital are not only determined by parents, but also by government policies and by philanthropists. To the extent that governments and philanthropists invest more in the human capital of children with less successful parents (as appears to be the case for governments in Scandinavian countries), a merit-based economy could have relatively high intergenerational mobility since children from poorer and less educated families might have high levels of human capital investments.
Nevertheless, a big jump is still required to make inferences from the intergeneration mobility in a country to the role of merit in determining success and failure in that country. In particular, although the United States has considerably lower intergeneration mobility than many Western European countries, this does not imply that merit is a less important determinant of success in the American economy than in these other economies.
And all of that said, I further dispute that "the American dream" was ever a claim that one parent's were irrelevant. This is James Truslow Adams, no relation to the Oresidential family:
Adams coined the term "American Dream" in his 1931 book The Epic of America. His American Dream is "that dream of a land in which life should be better and richer and fuller for everyone, with opportunity for each according to ability or achievement. It is a difficult dream for the European upper classes to interpret adequately, and too many of us ourselves have grown weary and mistrustful of it. It is not a dream of motor cars and high wages merely, but a dream of social order in which each man and each woman shall be able to attain to the fullest stature of which they are innately capable, and be recognized by others for what they are, regardless of the fortuitous circumstances of birth or position."
By way of contrast with a Europe of landed nobles and hereditary guildsmen, America was notably lacking in barriers to self-improvement and advancement.
Now to pick up on Stacy McCain's point - he is deeply dubious of a study that tells us that West Virginia is more of a land of opportunity than, say, Silicon Valley or New York City. Good point!
By way of illustration, here are San Jose, NYC and a few beacons of opportunity from West Virginia:
|New York||New York||10.5%|
These results reflect the probability that a child who was "raised" in the locale above in the lowest economic quintile (really, living there at about age 15 regardless of prior or subsequent movement) will eventually rise to the top quintile *of their age cohort* by about age 30. Parts of West Virginia have it all over the Big Apple or the biggest city in Silicvon Valley, which activates Mr. McCain's BS detector:
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.
Well, that's as maybe. Eric Mertz commented over there with a follow-up on his blog, noting that the decision to fix children to one locale based on where they were at approximately age 15 is fraught with implications.
I will say this: the authors attempt to check the validity of that assumption and conclude that all is well. I am a non-buyer and suspect that a deeper dive into the data would unearth trouble in paradise. But first, their comments in anticipation of this objection:
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 in 1996 for 95.9% of children in our core sample. For children in our core sample of 1980-82 birth cohorts, we therefore typically measure location when children were approximately 15 years old.
For the children in the more recent birth cohorts in our extended sample, location is measured at earlier ages. Using these more recent cohorts, we nd that 83.5% of children live in the same CZ at age 16 as they did at age 5. Furthermore, we verify that the spatial patterns for the outcomes we can measure at earlier ages (college attendance and teenage birth) are quite similar if we define CZs based on location at age 5 instead of age 16.
So that is reassuring. This, however, is far less so:
Importantly, the CZ where a child grew up does not necessarily correspond to the CZ she lives in as an adult when we measure her income (at age 30) in 2011-12. In our core sample, 38% of children live in a different CZ in 2012 relative to where they grew up.
That is a lot of movement, so they offer a bit of a breakdown:
44.6% of children who grow up in rural areas live in urban areas at age 30. Among those who rose from the bottom quintile of the national income distribution to the top quintile, the corresponding statistic is 55.2%.
So the poor move even more than the average. This final test reassures them but not me:
In row 8 [of Table V, p. 68], we assess the extent to which the variation in intergenerational mobility comes from children who succeed and move out of the CZ as adults vs. children who stay within the CZ. To do so, we restrict the sample to the 62% of children who live in the same CZ in 2012 as where they grew up. Despite the fact that this sample is endogenously selected on an ex-post outcome, the mobility estimates remain very highly correlated with those in the full sample. Apparently, areas such as Salt Lake City that generate high levels of upward income mobility do so not just by sending successful children to other CZs as adults but also by helping children move up in the income distribution within the area.
"Apparently"?!? As best I can follow, they looked at the aggregated data for Commuting Zones ("CZ") both large and small and and concluded that dropping the kids who eventually moved didn't change the results much. But that means that CZs with a large population, such as New York, will drive the population-weighted result and swamp whatever story the data might be trying to tell about Spencer, West Virginia.
So when they write that "apparently" Salt Lake City is seeing a combination of kids succeeding in the area and kids succeeding after moving away, my not-so-unreasonable question is, what did the data actually show for Salt Lake City specifically, as opposed to the aggregate? In the data tables all I can find is that without weighting for population the result for "8. Children who stay within CZ" has a correlation of 0.87 with the baseline estimate. However, when they re-weight by estimated population, the correlation rises to 0.95. That makes me think that rural areas are showing a much weaker correlation than urban areas. That overlaps with their comment that rural areas show a lot more movement, but the overall impoact is not to reassure.
And I am not at all sure what to make of this. Are we to conclude that Welch, WV is a great place to be born because it is easier to leave and prosper elsewhere? Or that being poor in Atlanta is dreadful but not so dreadful that people actually leave? Baffling.
GO AHEAD - MAKE MY DAY. Ask about Denmark.