In just over a month, voters will finish casting their ballots for candidates for the U.S. House. In most cases, the districts in which those voters live will have changed since 2020, expanding or contracting as states adjust the distribution of House seats following the most recent census. After a decade of mostly static boundaries, who and what members of the House represent has shifted to better reflect the population.
That means not only a change in the political composition of the House — state legislatures in charge of drawing those new lines are very attuned to the political repercussions of doing so — but also to how the country itself is represented. For example, redrawing district lines means moving some neighborhoods out of one district and into another, changing the composition of the district. Shatter a suburban district and you might create a new rural one, for example — probably changing who might get elected.
In 2018, David Montgomery created an index of how urban or rural congressional districts were for CityLab, now part of Bloomberg. It uses a six-category scale from “pure urban” to “pure rural,” with various iterations of suburban density in between. Usefully, Montgomery also explained the methodology he used for his calculations, allowing The Washington Post to replicate it with the newly drawn congressional districts.
As it turns out, there’s not much change at the margins. Here’s how Montgomery’s 2018 analysis compares with our new one in terms of the distribution of seats.
There are now three more rural districts than there used to be, but also three more districts that sit in the urban-suburban overlap. The biggest change is in suburban density, with a drop in densely suburban districts and an increase in sparsely suburban ones.
These terms are admittedly vague. The best way to explore the political differences is, instead, to compare how residents of the districts voted in 2020 and key factors that correlate to political decision-making, such as income and education.
Adding up the votes cast in each type of district in 2020, we see that the four densest population categories all preferred Joe Biden, by margins ranging from 57 to 2 points. The least-dense districts preferred Donald Trump, which is what we’d expect.
Within those categories, though, there’s variation. Three of the “pure urban” districts, for example, were won by Republicans in 2020; one of them also voted for Trump. That’s New York’s 11th district: Staten Island and a sliver of Brooklyn.
At the very bottom, there are similar outliers. In the most rural districts, there are seats won by Democrats in 2020 and won by Biden that same year.
Interestingly, the median household income in these clusters of districts doesn’t vary that much, save in the most rural ones. There are much wealthier districts included in the “pure urban” and “urban-suburban” mix categories, but the median incomes in the four densest groups are all within about $3,500 of each other.
You’ll remember that those latter two groups, where median incomes are lower, were the two that preferred Trump in 2020.
But that’s not solely because of income. Population density overlaps with other factors, too, like education. Considering the percentage of the population that has a college degree (or more), the four most densely populated groups of districts all sit at about the 30 percent mark. In the more-rural groups, that percentage drops.
Those two factors are correlated. In each district, more education and higher incomes go hand-in-hand. You can see that below. As dots (individual districts) move to the right (have more residents with college degrees), they also generally move up (the district has a higher median income).
If you watch, you’ll see that we’ve isolated the most rural and most urban groupings of districts. Even in those, the correlation exists: Rural districts with higher education have more income, even if both generally trail the values for more urban districts.
Those rural districts with higher levels of education also were more likely to vote for Joe Biden in 2020.
The best indicator of how redistricting will affect the politics of the House, of course, will come in November. But just by considering this limited demographic lens, we see two different Americas: one wealthier, more educated and more urban. The other, less of each of those things — and more likely to vote for Trump.
Adrian Blanco Ramos and Harry Stevens contributed to this report.