Let’s start with geography. Urban counties showed a bigger swing toward Trump than suburban and exurban counties, smaller metros, and rural areas. Of course, Harris did best — as did Biden four years earlier — in urban counties, but the 10-point swing toward Trump in urban counties was larger than swings in other places.
A more refined county classification from the American Communities Project, which groups counties based on their demographic, economic, and other factors, confirmed that Trump did better in 2024 than in 2020 in all types of communities, with larger swings in some places than others. Big cities, Hispanic centers, and Native American lands swung most toward Trump in 2024. The reddest communities — aging farmlands, evangelical hubs, and working class country — swung less, as did still-blue college towns and LDS (i.e. Mormon) enclaves, where Trump has repeatedly gotten smaller margins than previous Republican presidential candidates.
Going one step more granular to individual metros, many swung more than 10 points toward Trump in 2024 versus 2020, including New York, Los Angeles, Chicago, San Francisco, and Miami, as well as heavily Hispanic/Latino metros in Texas, California, and the Southwest. Just a handful of metros swung a bit bluer in 2024, mostly in the Mountain West and Pacific Northwest, including Salt Lake City, Tucson, and Colorado Springs.
Looking across all counties that have reported election data, the geographic pattern of the 2024 vote was less polarized than in 2020 in some ways. Most notably, counties with a higher share of Hispanic residents were more likely to vote for Harris than for Trump, but by smaller margins than for Biden in 2020. Same with higher density counties: there was a very strong correlation between county density and Harris vote share, though not as strong as in 2020. In contrast, the correlation between county education level and Harris vote share strengthened further in 2024. Density and education are themselves highly correlated, with residents of more urban counties more likely to have a college degree than those of more rural counties, but higher-density counties swung toward Trump, while highly educated counties did not.
This is a different kind of analysis than looking at percentages of urban, suburban, and rural voters and who they voted for. This considers which places changed the most between 2020 and 2024.
One question about this is whether the electoral college outcomes changed if one candidate picked up more votes in cities. If the election came down to key states, were these swings in urban areas enough to win the state? Or maybe they did prove consequential in purple states. Looking at these swings in particular places could help address this. In Pennsylvania, did changes in metropolitan Philadelphia and Pittsburgh decide this or in Wisconsin, changes in metropolitan Milwaukee and Madison?
Additionally, it is less clear what this all means for considering suburban voters. The American Communities Project typology includes multiple suburban settings, Urban Burbs, Middle Suburbs, and Exurbs, in addition to suburban areas that might fit into other categories because of unique traits (such as a college town in a suburban county). Just looking at the three with suburbs in their title in one form or another, the 2020 patterns held: exurbs leaned Republican, suburbs near cities leaned Democratic, and middle suburbs leaned Republican. But voters in each three categories moved toward Trump. Was this shift substantive? Did suburban voters decide the 2024 election?
I am sure there is more analysis to come on this subject and I will keep looking for it.
Legally Sociable 