What would Brexit look like if it happened in the United States?
An American Brexit wouldn’t split the nation as deeply as you might think. Sure, most Americans speak English as their first (or only) language and are taught from an early age about English colonies such as Plymouth and Jamestown. But the share who actively claim English, Scottish, Welsh or Northern Irish ancestry on government surveys is small and geographically contained. It shows that any focus on early white settlers doesn’t represent the reality according to the self-reported identity of most Americans.
The ultimate results depend on how you slice the numbers, but one thing remains constant: If the split were based on ancestry alone, Utah and eastern Idaho would be the first to pull a British exit. As a rule, areas of Mormon influence report the strongest ancestral ties to the U.K., and both Utah and eastern Idaho have large Mormon populations. By some metrics, northern New England and much of the Deep South and southern Appalachia would also be part of a Brexit bloc.
For this experiment, we focused on comparing the ratio of people claiming U.K. ancestry in each county with the ratio of people claiming ancestry from the other 27 countries in the European Union. That has obvious limitations. The United States, United Kingdom and European Union are multicultural societies, but our metric can only capture the overwhelmingly white subset of Americans claiming British or E.U. ancestry.
In addition, support for Brexit within the U.K. varies. In the 2016 referendum, voters in England opted to leave the E.U. Voters in Scotland and Wales wanted to stay. But for these purposes we will consider Americans claiming roots in England, Scotland, Wales and Northern Ireland to be a single bloc that would exit together, based on the U.K.'s narrow vote in favor of leaving.
The initial result is a bit too messy for a theoretical national boundary, and probably would contain only Mormon country. But once we broaden the definition to anywhere that’s atypically tilted toward U.K. roots and consider each county’s nearest neighbors to smooth things geographically, we get a loose coalition of places that are the most British (Utah and Maine) and places where the European population is smaller, but tends to have stronger U.K. ties (the Deep South and southern Appalachia). They’re linked by a stretch of Arizona, New Mexico, Texas and Arkansas where the ratio of U.K. to E.U. is slightly higher than would be typical. Emphasis on “stretch.”
Few Mormons will be surprised that a U.S. Brexit quickly becomes a Latter-Day Secession. The Mormon Church has deep roots in England, where it expanded in its early days. By 1850, English Latter Day Saints outnumbered those in the United States. About 100,000 of them would emigrate to Utah Territory in the 1800s, according to estimates by Mormon historians cited by the BBC. In 1870, about 27 percent of the territory’s residents had been born in the U.K. That’s more than ten times the rate seen in the rest of the country at that time.
That total is padded by American Mormons, many of whom came from Upstate New York, northern New England and other areas with U.K. ties. The church’s most notable (and prolific) early leaders, such as Brigham Young (55 wives), came from English stock.
Folks with U.K. roots also outnumber those reporting E.U. backgrounds in the South. This is in part because the South received a smaller share of the European immigrants who poured into the United States in the 19th and early 20th centuries. Many of those immigrants came from countries that are part of the European Union, first from Germany and Ireland, and later from Italy and Poland.
The South is much more diverse than Utah, as you may have noticed. By our measure, the most common self-reported ancestry or origin in the region is “Black.” In Appalachia, the most common self-reported ancestry is “American,” another broad category that shows how messy ancestry can get. In Appalachia, a concentration of Scots-Irish help tips the ancestry balance toward the U.K. The group includes Scots who migrated to what is now Northern Ireland under a planned colonization program in the 17th century, and later left for the New World.
Ancestry isn’t a perfect measure of someone’s origin. To be as fair as possible given the well-documented issues with how the Census Bureau has treated ancestry, race and ethnicity, we’re including several aspects of self-reported identity. Because many people report more than one ancestry or ethnicity, for example, we’re counting people with English mothers and German fathers twice, once on each side. Some may be counted even more often. In a typical category, it should balance out.
We are not counting folks from former British colonies such as Canada or Australia as originating in the U.K., nor do we include people from former Spanish colonies such as Mexico as originating in the E.U., unless they also list separate Spanish or English ancestry.
The most common country of origin in our data is German. That doesn’t include those who report German ancestry indirectly, such as the 25,000 Volga Germans and others who are recorded as “German Russian,” or the 306,000 Pennsylvania Dutch who are marked as “Pennsylvania German.” The next-largest single groups are Mexican, Irish, English, American, Italian, Polish, French and Scottish.
The Census Bureau’s forms tend to provide extensive detail on European ancestries, such as Carpatho Rusyn (primarily hailing from mountainous border regions of what is now Slovakia, Ukraine and Poland) or Alsatian (from what is now eastern France, adjacent to Germany), but far less on the many U.S. residents who don’t report links to Europe. To get additional detail on these Americans, we have combined parts of tables that ask related questions about race and ethnicity.
If we could properly account for all Americans in our thought experiment, instead of focusing on the nations involved in the Brexit debate, the entire conflict would seem comically minor and regional — akin to a spat between Utah and the Dakotas, perhaps, rather a continentwide rift.