The American Nations and the States: Change over time

We previously parsed 2020 census data to reveal present day state-level power dynamics. Now we’ve done the same for 1900 and 1950, revealing remarkable stability in most states, and dramatic shifts in others.

By Colin Woodard

Back in February we posted some tools allowing people to see how current state populations are apportioned between the regional cultures in my American Nations model. This is useful because it allows us to better evaluate how state-level polls, data, election results, and policies might (or might not) reflect the regional cultures at work. You can see that work here, which used the 2020 census data and offered a few surprises even for me, like that Left Coast is the least populous of California’s three regions and that Arizona’s El Norte section is almost nine times more populous than its Far Western half.

But that got me wondering about change over time and if political and policy changes in some states might reflect internal shifts in relative population and electoral power between regional sections. So, with the help of our colleagues John Liberty of the cultural geospatial consulting firm Motivf and Bilal Omar Omari, our capable undergraduate research intern, we parsed the 1900 and 1950 U.S. censuses to see how regional control of the various states has changed over time.

At the top of this post is the map for 1900.

In this and the maps to follow, the color coding represents the degree to which any one regional culture influences the state, based on relative populations of the sections they control. The darkest shading is for states that entirely fall within a single regional culture. The next shade is for states dominated by a single culture that accounts for more than 70% of the population; the dominant section can usually govern as it pleases. The third shade is for states controlled by one “nation” with 50.1 to 70 percent of the population – they’re the biggest section, but they usually have to compromise with the others. The lightest shade are the truly contested states, where no single regional; culture is in the majority. And greyed out states hadn’t yet achieved statehood.

Here, for comparison, is 1950:

And here is the situation in 2020:

Perhaps the most consequential changes in regional control over this 120-year period is that California has gone from being controlled by the Left Coast – which in 1900 was by far the most populous section — to being controlled by El Norte (with Left Coast now the least populous section.) This change was accomplished even in 1950 and is of course due to the explosive growth of Los Angeles, fueled by white midwestern and Greater Appalachian migrants in the Dust Bowl and early postwar periods.

Illinois has shifted dramatically since 1950, with the Yankee north (including 50% of Cook County) going from 10.9% of the state population to 45.1% today. Greater Appalachia, the most conservative tier, was the largest section in 1900 (at 38% of the population) and is now very much the smallest (at less than 21%.)

Virginia has gone from being just barely under Tidewater control in 1900 and 1950 to being overwhelmingly dominated by that section today. And both Dakotas have seen a shift toward the Far West and away from the Yankee and Midlands portion.

But the most remarkable thing is just how stable the sectional balances have been most everyplace else despite everything that’s happened since the McKinley administration. Look at Oregon, Washington, New York, Virginia, and even Texas and you’ll see remarkable consistency.

— — Colin Woodard is the director of Nationhood Lab at Salve Regina University’s Pell Center for International Relations and Public Policy.