One of the key points I’ve tried to stress on this blog is that micro-scale population structure – that is, fine genetic variation across populations can have a substantial impact on societal characteristics. We aren’t just talking about continental racial variation. We aren’t even talking just about ethnic variation. Sorting within an ethnic groups can produce distinct regional differences. Founder effects are powerful, as is the converse effect, boiling off.
This means that regional differences across countries like the United States reflect genetic differences between people. Even of those who accept that genes impact behavior, many like to blame these local variations on local “culture” (as if culture was some otherworldly force itself without cause). But our discoveries render that view untenable. These include many newer behavioral genetic studies using nationally representative U.S. samples. These studies find no shared environment influence, which would turn up if local cultural effects existed (see the following: on victimization, Boutwell et al 2013; on domestic violence, Barnes et al 2012; on criminality in adoptees, Beaver et al 2015a; on football participation and violence, Beaver et al 2013; on enlisting in the military, Beaver et al 2015b; on peers and academic performance, Barnes et al 2014; on peers and delinquency, Boisvert et al 2013; on handgun ownership, Barnes, Boutwell, and Beaver, 2014). In addition, there are large extended twin studies looking at politics, religiosity, and marriage that find nothing attributable to local environment (Hatemi et al 2010; Coventry & Keller, 2005; and Zietsch et al 2011, respectively). (These are in addition to the many population-wide Scandinavian studies that generally find an absence of neighborhood effects – future post). These behavioral genetic studies along with quantitative genetics methods (which inform us about rates of selection) demonstrate to us that this…
…results from the genetic variation that exists within it (see my page American Nation Series for a review).
Now a new piece of evidence has emerged, one that speaks to the other myth, of “assimilation”. This evidence will explain other fine-scale variation we see, like on this map of income per capita by county (from here):
Different regions of the United States perform differently on various social indicators. A great many of these indicators were featured in my post More Maps of the American Nations (as well as in the earlier post Maps of the American Nations). The pattern we see above (and many other patterns) – while clearly partially the result of (continental) racial variation, isn’t solely due to such, since we see variation within the White population as well, as can be seen with White poverty rates (from here)…
…and with White teen birth rates (from the CDC)…
…in addition to many other aspects as discussed and mapped in my previous posts. One can’t help but notice that these follow a pattern, one that generally follows the American Nations lines. But what drives the American Nations divide, and is that the only factor at work here? Here now I tackle another persistent myth: the myth of assimilation
While the founding colonial stock was the base for the subsequent differences that would take shape across North America, as argued by David Hackett Fisher and Colin Woodard (see Maps of the American Nations), it isn’t solely responsible for what we see today, because British Americans are not a majority even among White Americans. Subsequent immigrants have greatly altered the American landscape.
The United States received many immigrants – who have largely come in two major pulses. The first was after the Civil War and was mostly European. The second came after the 1965 immigration act and was mostly Asian, Latin American, and Caribbean in origin:
The immigrants settled unevenly. The Scandinavians, Germans, and Eastern Europeans mostly settled into northern areas in the western reaches of Yankeedom and the Midlands. The (Catholic) Irish, the French Canadians, and (especially) the Italians stayed closer to the industrial centers in New England and New York. All immigrants generally avoided the nations that make up the South:
A key problem with tying socioeconomic outcomes to ethnic ancestry has been the difficulty of getting reliable measures of present-day ancestry (see Being the Dutch | West Hunter). Genomic data has helped somewhat, but isn’t yet fine-grained enough to examine matter very well (see More Maps of the American Nations). However, Fulford, Petkov, and Schiantarelli (2015) have come up with a novel way to solve this problem.
Using longitudinal data from the U.S. Census, they were able to track the movement of people across the country. This allowed them to estimate the actual ethnic ancestry in each U.S. county, rather than rely purely on self-report (which they show to be highly inaccurate). Here are the actual ancestry fractions across the country:
(Unfortunately, due to the limitations of their data, the “English” category does not break down the “Four British Folkways” of David Hackett Fischer’s Albion’s Seed but includes them all. This makes comparison of the original American Nations difficult here.)
With a reliable measure of the ancestral make-up of each county, analyzing each county’s characteristics and comparing these characteristics with those of the ancestral nations was a fairly straightforward endeavor:
We see that (with a few exceptions), the performance of the different ancestry groups conforms closely to what we’d expect (given the average IQ of the source country, with selective migration considered). They were able to correlate ethnic ancestry in each county with both characteristics in the source countries and with GDP per capita in each county:
These clearly show the positive relationship between immigrant characteristics and county level characteristics. One drawback of their analysis is that they compare everything against only their determined ancestry coefficient. We already that IQ is associated with all sorts of positive social outcomes. What may have been more useful is to compare these national characteristics with the same characteristics in each county. While many US counties probably lack the appropriate measures, it would have been a far more of a solid statement if they were able to do so.
Nonetheless, this paper confirms the importance of genetic stock to social outcomes, and weakens any case for “assimilation.” Areas across the northern Plains with exceptionally good outcomes have such because they are heavily Scandinavian in origin. The poorer outcomes of the South can be traced to its Scots-Irish heritage. Even more recently, the independent-minded nature of the Far West and the introverted but cosmopolitan nature of the Left Coast can be traced to the traits of the people who founded (and continue to relocate to) these nations. These patterns follow each of these groups even after more than century. As Fulford et al put it:
Every form of the estimation strongly rejects that ancestry does not matter. All estimates include county group fixed effects, so the fixed characteristics of the place of settlement is controlled for. We can also ask whether regional trends—which might reflect evolving factors, such as industrial structure, that may be related both to county GDP and ancestry composition—may affect our answer. However, the inclusion of state specific period effects or county group specific trends, leaves the significance of the ancestry composition intact. Our conclusion that ancestry matters are also robust to the adding county income in the previous period … as a regressor. One might be concerned that ancestry matters only because it reacts to current shocks, yet ancestry matters even when we include it only at a decade lag. The last several columns include other possible explanatory variables including county level education, population density, and fractionalization constructed using our ancestry vector … All of these may be a result of ancestry, for example, if some groups put more emphasis on education than others. Similarly, an increase in density may reflect a higher level of urbanization of the county, resulting in a differential attraction for different immigrant groups. Ancestry continues to matters even after including these controls, and so ancestry matters beyond its relationship to education or urbanization.
This finding serves to complete the case for heritable human differences as a source societal variation. This goes past the well-known (at least in the HBD-sphere) relationship between average IQ (derived from Lynn & Vanhanen, 2006) and national wealth and economic development:
The evidence should show that even the rich regional flavor across North America has genetic roots. This also means that the persistent regional conflicts that trouble us here in America are in good measure intractable, as Colin Woodard discusses in a recent Politico piece.
Woodard touches on another key point in his article: the regional differences are amplified by assortative migration. See the following:
All these facts should makes clear how key demography is to societal outcomes. The people make the society. If you want a better society, you must (in one way or another) get better people. This has obvious consequences for immigration and for many other matters as well. Only time will tell if these facts come to be appreciated for what they are.
Happy Independence Day to my American readers! (And a slightly belated Happy Canada Day to my Canadian readers!) I hope you enjoy your holiday. (Note: the following is NOT an endorsement of Hillsdale College or of their views.)