Pivot Counties: How Obama-Obama-Trump counties voted in presidential elections since 1992
July 21, 2020
Ballotpedia identified 206 counties that voted for President Donald Trump (R) in 2016 after voting for President Barack Obama (D) in both 2008 and 2012. Collectively, Trump won these Pivot Counties by more than 580,000 votes and had an average margin of victory of 11.45 percentage points. These counties are sometimes referred to as swing counties by media and political observers. The political shift in these counties could have a broad impact on elections at every level of government. What those shifts might be, however, depends on several factors, including the nature of the shift, the causes of the shift, and the durability of those causes.
In order to establish a baseline, Ballotpedia examined election results in these Pivot Counties dating back several election cycles before the 2008 presidential election between Obama and Sen. John McCain (R). This analysis looks back to the 1992 presidential election between President George H.W. Bush (R) and Arkansas Gov. Bill Clinton (D).
We selected this date purposefully. First, using four elections prior to Obama’s first victory, this date allowed us to observe what, if any, patterns existed prior to 2008. Additionally, 1992 marked the start of several consecutive presidential election cycles in which the contests were decided by relatively narrow margins, which marked a break in the presidential voting pattern.
From 1972 to 1988, four of the five presidential elections saw the winner earn more than 75% of the Electoral College votes, and two of these—1972 and 1984—were decided by the widest Electoral College margins in the history of the nation. In 1992, by contrast, Clinton won the presidency without winning a majority of the popular vote. This is a feat he repeated in 1996 and that President George W. Bush (R) and Trump also replicated in 2000 and 2016, respectively. These were, in short, close contests by the standards of recent history, and if partisan voting patterns exist, then they would likely be more easily identified in these races than they would be in the previous two decades.
The analysis was necessarily limited to a small portion of all counties across the United States. First, Trump won 2,626 counties in 2016, while former Secretary of State Hillary Clinton (D) won 487. That means that the 95 counties that switched to Trump after voting for, at least, six straight Democratic candidates represent 3.1% of counties nationwide and 3.6% of counties that voted for Trump. Additionally, we have not identified any other trends that might be lurking in the data—for example, the number of counties that were traditionally Republican but switched to Obama and did not switch back to Trump, making them possible long-term Democratic pickups. Therefore, we approach this data carefully and judiciously.
Now, all of that said, there is some interesting information contained herein, indicative of some potentially important, if nascent, trends. Broken down by state, Iowa is the state with the most counties that were long-time Democratic supporters but flipped to Trump, with 20 such counties. This county-wide trend mirrors statewide data. Iowa voted for the Democratic candidate in every election between 1988 and 2012, save one—2004—when it voted for Bush over Sen. John Kerry (D) by less than 1 percentage point. Iowa voted for Trump in 2016, by almost 10 percentage points, which is quite a bit larger than most of the margins it gave Democratic candidates. Given this, and given that these 20 counties represent more than 20% of the state’s total counties, it may be the case that Iowa is or may become a solid Republican state in the longer term.
Of the remaining 75 long-term Democratic counties to flip to Trump, an additional 41 of them were located in what could be called the Greater Midwest – i.e. the Iron Belt, the Upper-Midwest, and the Plains. Illinois had seven such counties (and one of the three that went the other way, flipping to vote for home-state candidate Obama). Indiana had one. Michigan had five. Minnesota had eight. Ohio had five. Pennsylvania had three. And Wisconsin had 12. Taken together, that’s 64% of the total number of counties that flipped from long-term Democratic positions to Trump. Iowa, Minnesota, and Wisconsin alone had 40 of 95 such counties, or 42% of the total.
The other 108 counties had the following voting patterns:
- 18 counties voted for the Democratic candidate once and the Republican candidate three times.
- 40 counties voted for both the Democratic and Republican candidates twice.
- 50 counties voted for the Democratic candidate three times and the Republican candidate once.
Among all Pivot Counties, 35—or 17%—voted for the winning candidate in the four presidential elections between 1992 and 2004. Wisconsin had the largest number of accurate Pivot Counties, with seven backing the winning presidential candidate over that time period.
This can be seen as supportive of the analysis that the upper Midwest is undergoing a political realignment, similar to that which the South underwent a generation ago. If this pattern holds (and that is, indeed, a big if), then it will be another point in the growing data series that suggests that the United States is bifurcating into two distinct regions: an urban and coastal region that votes exclusively Democrat; and an interior, suburban, exurban, and rural region that votes (almost) exclusively Republican. Indeed, if you look at the other states with multiple long-term-Democratic Trump counties—Maine (6), New York (5), and Washington (3)—they are either rural or, in one or two cases, exurban.
In sum then, much of what the Pivot Counties tell us about partisan trends is useful, at present, mainly at the state and local levels. Republicans may find a surge in new voters who could change local politics. These voters would not, however, be able to change presidential voting patterns in and of themselves, particularly in states with large cities that vote overwhelmingly Democratic, cities like New York, Seattle, and Denver.
At the same time, the county data in the states of the Upper Midwest and the Rust Belt suggest that a genuine re-alignment may be underway there. The large number of Pivot Counties in those states, coupled with other trends, and with the lack of cities that dominate state politics, suggest a real opportunity for the GOP to make significant gains.
Data
The following embedded spreadsheet includes data from how Pivot Counties voted in presidential elections from 1992 through 2016.
See also
Footnotes
| |||||||||||||||||||
