Sometimes, a narrative is told about American history that goes like this: in the 19th and 20th century the US received a huge number of European immigrants. Initially, they had high levels of crime and poverty, and reliably voted for the democrat party. With time, these groups economically assimilated into country, deemphasized their ethnic origins, and became an indistinguishable part of White America.
This narrative is used by liberals to argue that we should expect the same thing to happen with Hispanic immigrants today. Race realists also use this narrative to argue that European ethnic differences are of relatively little importance, while the same can obviously not be said of the racial differences that have allowed Black, Hispanic, and Asian, Americans to live here for generations and never assimilate. The later is closer to the truth than the former, but in this post, I am going to argue that both interpretations are significantly wrong because the historical narrative they are interpreting is significantly inaccurate.
Today, there exist substantial per capita income differences between different groups of European Americans. Below is data from the census showing the per capita income of different groups, and how many dollars each group makes per dollar made by German Americans.
|Ancestry||Per capita income||Ratio to Germans|
|French (except Basque)||$36,559||0.96|
Several of these gaps are as large as the income gaps between races and sexes that at the center of identity politics.
These differences aren’t random, either. Fulford et al. (2018) analyzed data on the ancestry composition and GDP per capita of 1,154 US counties tracking census data for all the years between 1850 to 2010 to create a much more accurate measure of European ancestry than is normally available. Controlling for the effects of race, and the effect of immigrants choosing to move to economically well-off counties, Fulford et al. found immigrants who came from richer nations had more positive effects on county GDP. In the long run, it was found that a 1% increase in the GDP per capita of an immigrants nation of origin predicted a .3% increase in their effect on county GDP per capita.
The paper further shows that the impact that immigrants from a given nation will have on the economy can be predicted by how long that nation has had an independent state and the degree to which it has a culture characterized by trust and cooperation. Once these variables are controlled for, the formal education of an immigrant group does not predict its impact on the economy.
So, differences between European ethnicities were real back then and they are real now. At the same time, it is important to realize that the popular story of certain groups, the Irish and Italians, for example, slowly rising up the American economic latter over multiple generations is largely false.
In the time of mass European migration, it was common for immigrants to move to America, try it out, and if things didn’t go well they would return home. Between 1850 and 1913, roughly one in three American immigrants returned to their nation of origin. Between 1890 and 1920, 54% of Italian immigrants returned to Italy. The poorer an immigrant was the more likely they were to return home.
Abramitzsky et al. (2014) show that if you look at long staying immigrants, even in the early 20th century most European immigrant groups, including the Irish, Italians, and Russians, had above-average incomes, and this was true within the first generation of immigrants.
In fact, for most European ethnicities there was little difference in income between the first and second generations:
It’s easy to make a group richer if you get all the poor people to leave. That is largely what America did. The story of a multi-generational climb to the middle class is largely untrue.
Crime and Literacy
Turning to crime, in popular culture it is often implied that European immigrants used to have high crime rates similar to those of certain groups of modern immigrants. This is completely untrue. Moehling and Piehl (2009) show that in 1904, 1923, and 1934, native Whites had higher crime rates than foreign-born Whites.
Breaking the data down by ethnicity, and correcting for age, some European groups had high rates of criminality, but this was due to minor crimes like public drunkenness. Group differences in serious crime rates were minuscule.
At least, this was true for Europeans. Mexican immigrants, on the other hand, had a rate of major offense committing that was over 5 times that of similarly aged native Whites.
It is also often implied that the groups coming from outside the British Isles were overwhelmingly illiterates who couldn’t even speak English. Again, this is untrue. The same paper shows that 72% of Italians, 85% of Russians, and 69% of Poles spoke English. Literacy rates were higher than 85% in the vast majority of groups.
Once again, the biggest exception to this trend was Mexico. The literacy race of Mexican immigrants was only 58%, and the rate at which they spoke English was 30%.
Clearly, some assimilation between European ethnicities has taken place. The minority of them that didn’t speak English now do, and a few of them used to have elevated rates of committing minor offenses and now they don’t. But the actual assimilation that did take place is much lesser than what is often implied
With respect to politics, the story is that European immigrants used to be reliable Democrats, but now they vote like the rest of White America does. As I documented in a previous post, this is not accurate. On net, many European ethnic groups don’t identify as Republican.
This chart is based on an analysis I did of GSS data that was collected between 1972 and 2014. I also looked at how different groups reported voting in US elections between 1968 and 2012, finding that Irish, Italian, and Scandinavian Americans continued to be, at best, unreliable Republicans right up to the 2012 election when they all reported having voted for Obama over Romney.
The claim that Polish Americans are, on net, Democrats, is backed up by research done by Radzilokski and Stecula (2010).
The claim that Italian Americas are, on net, Democrats, is supposed by a 2011 Zogby Poll.
BuzzFeed did some polling in 2016 replicating the finding that the Irish identify as Democrats but finding that Italians ID as Republicans. It is highly plausible that this reflects a real change in Italian American opinion, likely brought on by Trump.
(They also asked about who people intended to vote for, but huge numbers of people said they weren’t going to vote for either major party candidate, something that obviously didn’t happen, and so that analysis doesn’t seem to have much value.)
With respect to the Irish, there is some evidence that they voted for Clinton in the 2016 election. A poll reported on by The Irish Times found that 47% of Irish Americans voted for Clinton and 27% voted for Trump. The same survey found that 41% identified as Democrats and 21% as Republicans. However, this poll was biased in that its respondents were disproportionally from the North East, possibly more so than is appropriate for a poll of Irish Americans.
Another analysis looked at the 14 counties in which the Irish comprise at least 25% of the population. Clinton won 8 of them. This means that Clinton won 57% of Americas most Irish counties while she only won 15.5% of US counties generally.
Before, we saw that the economic impact of immigrants could be predicted based on the countries they came from. Analyzing the 24 European ethnic groups I had data on, I found that their political behavior can be likewise predicted.
The relationship between intelligence and partisanship was weak though so it was dropped from the model. Protestantism and economic freedom both continued to predict republicanism after controlling for each other.
It is noteworthy that my numbers on economic freedom and Protestantism are contemporary, meaning that the political viewpoints of people who immigrated here some time ago and be predicted on the basis of what their country of origin is like now.
The differences between European Americans of the past are often exaggerated. The differences between them today are often minimized. Some assimilation has happened, but it has been massively overblown. Differences still exist. They are moderate in size and have been for a long time. These differences are predictable on the basis of the home country because these differences are stable.
If Americans were conscious of these ethnic differences, they could fuel bitter ethnic conflict. They did in the past. We are not conscious of them, but not because they don’t exist. It seems plausible to me that the suppression of White racial consciousness has probably had an inhibiting effect on White ethnic consciousness as well, though that is just speculation. In any case, this is an interesting example of how group differences only matter if people decide to care about them, and how the category “White” contains a good deal of heterogeneity.