The 2024 Election: What Happened in Iowa? Part 2
The Second in a Series on 2024 Presidential Vote Returns in the Hawkeye State
This post is the second in a series of four posts on the presidential race in Iowa. Each post in the series will include visualizations based upon 2024 preliminary returns and presidential vote choice data from 2016 and 2020 obtained from the Iowa Secretary of State’s Office.1 The first post in the series focused on the magnitude of Trump’s victory across the state in 2024 by looking at county-level returns. Today’s post will examine the growth in Donald Trump’s support by examining county-level returns for Trump in 2020 and 2024 as well as county-level returns for Joe Biden in 2020 and Kamala Harris in 2024.
In 2024, Kamala Harris’s popular vote total was significantly lower than Joe Biden’s in 2020. Not only was this true at the state level where she underperformed Biden by over 52,000 votes, it was also true across 97 of Iowa’s 99 counties as well. The figure above shows the change in the Democratic candidate’s popular vote total by county between 2020 and 2024. The change in the total was calculated by subtracting Biden’s popular vote total in 2020 from Harris’s popular vote total in 2024 for each county. Counties shaded in blue represent counties where Harris outperformed Biden while counties shaded in red represent counties where Harris underperformed Biden’s 2020 vote total. The darker the color, the larger the over or underperformance.
Overall, Harris underperformed Biden by an average of 749.2 votes per county in 2024. She only exceeded Biden’s vote total in two counties. One was Dallas County in central Iowa where her popular vote total was more than 2,400 votes higher than Biden’s in 2020. Dallas County, which includes the City of Waukee, a rapidly growing suburb of Des Moines, is a county that has moved more toward the political center as the county has grown more urban. Dallas County has some of the demographic features of a suburban swing county around the country, including the highest median household income in Iowa and the share of county residents who are college graduates is north of 50%. The other county where Harris overperformed Biden was Warren County, located just to the south of the Des Moines metro. Her overperformance was quite small, however, at only 88 votes more than Biden in 2020.
Harris’s underperformance was widespread as indicated by the counties shaded in red in the figure above. Harris underperformed Biden most significantly in the urban counties of the state where she needed to perform well in order to be competitive. She underperformed Biden in Polk County, which includes much of the Des Moines metro, by 6,287 votes. In Linn County, which includes Cedar Rapids, by 4,578 votes. In Scott County, which includes Davenport and Bettendorf, by 4,478 votes. And in Black Hawk County, which includes Cedar Falls and Waterloo, by 4,375 votes. She even underperformed Biden by 405 votes in Johnson County, which includes Iowa City.
Donald Trump, on the other hand, overperformed his 2020 vote totals across the state. His popular vote total was more than 28,000 votes larger statewide in 2024. The figure above, which shows the difference in Trump’s vote total from 2020 to 2024, also shows the growth in his county-level vote totals around the state. In the figure, counties with darker shades of red represent counties where the difference between Trump’s 2020 and 2024 vote totals was the largest.
A couple of trends stand out in the map above. First, while Trump underperformed his 2020 vote total in a third of Iowa’s counties (shaded in a very light shade of blue), his underperformance was small on average. In fact, the county where his underperformance was the largest was Kossuth County in north central Iowa at only 302 total votes.
Second, Trump managed to improve his differentials across a substantial number of rural counties where his differentials were quite large in 2020. In the map above, there are many counties shaded in very light red or even pink. Those counties are counties where Trump’s vote totals grew modestly, but still grew nonetheless. In some cases, Trump’s differential in 2020 was over 60 points and the campaign still increased their vote total in 2024. Take, for example, Sioux County in northwest Iowa. In 2020, Trump’s differential in Sioux County was 66.1 points. In 2024, Trump’s popular vote total in Sioux County increased by 342 votes. What this suggests is that even in the deepest of red counties, Trump was able to squeeze out hundreds of additional votes which improved his vote differentials across the state.
Finally, Trump made significant gains in some of the urban counties of the state where Democratic candidates need to maximize their vote differentials in order to win statewide. In Polk County, he overperformed his 2020 vote total by 5,381 votes. In Johnson County, by 3,144 votes. And in Scott County, by 2,264 votes. In Warren County, one of the two counties where Harris overperformed Biden, Trump’s vote total in 2024 was 1,677 more votes than he received in the county in 2020. Also, please note the color shade for Dallas County. While Dallas County was a bright spot for Harris as she overperformed Biden by 2,474 votes, Trump overperformed his 2020 vote total in Dallas County by 4,342 votes which helps explain why his differential in the swing county increased from just under two points in 2020 to nearly five points in 2024.
So what is the importance of understanding Trump’s widespread overperformance and Harris’s widespread underperformance? In short, it led to Trump significantly overperforming his 2020 differential in county after county. The figure above shows the change in Donald Trump’s vote differential between 2020 and 2024. The change in differential is calculated by subtracting Trump’s county-level differential (i.e., the margin he won or lost by) in 2020 from his county-level differential in 2024. In the figure, darker shades of red represent the largest changes in his county-level differentials.
As displayed in the map above, Trump overperformed his 2020 differentials in 98 of Iowa’s 99 counties, including in the six most urban counties where Democratic candidates typically have advantages. On average, in 2024 Trump overperformed his 2020 differential against Joe Biden by 5.3 percentage points. What this means substantively is that on average his winning county-level margin was 5.3 points higher in 2024 than it was in 2020. The only county in Iowa where Trump did not improve his differential was in Page County in southwest Iowa. But it is worth noting that Trump’s winning margin in 2024 in Page County was only .4 percentage points lower than in 2020 and he still won the county by over 42 percentage points in both years.
The largest growth in Trump’s county-level differentials was over 8 percentage points in three Iowa counties. His differential grew by 8.5 points in both Lee County (from 19.2 points in 2020 to 27.8 points in 2024), in southeast Iowa, and Clinton County (from 10.3 points in 2020 to 18.8 points in 2024), which is located along the Mississippi River just north of Scott County. Trump’s differential also grew by 8.2 points in Emmet County (from 35.7 points in 2020 to 43.9 points in 2024), which is located in northwest Iowa along the Minnesota border.
In sum, Trump saw substantial growth in his support not only statewide but also in nearly every Iowa county in 2024, which helps to contextualize why his overall state margin grew by nearly five points. In the next post, we will look at the change in county-level support for Trump from his entry into electoral politics as a candidate in 2016 to 2024.
If you have comments or questions about the data presented above, feel free to drop them in the comments using the button below.
Vote return data was downloaded on Friday, November 8th. Final certification of the vote will occur in early December.
I created the maps using R and the ggplot2 library of functions. The latitude and longitude data used to draw the counties lines is contained in the maps_data( ) package within ggplot2.
I'm not a numbers person so it's hard to follow all those details. My question still is WHY?