As goes Clay County … so goes the nation?
The northwestern Minnesota county that borders North Dakota is something of a rarity, one of only nine counties across the U.S. that have voted for the winning presidential candidate every time since 2000.
It’s even more unusual because the number of bellwether counties is rapidly diminishing.
Over the years, Clay County has gone for Democrats Barack Obama and Joe Biden as well as Republicans George W. Bush and Donald Trump.
More recently, Clay County’s seemingly even mix of conservative rural residents and the increasingly liberal population in Moorhead has helped make it a swing county in presidential elections.
“I think basically it encapsulates the whole rural-urban divide in American politics for some time now,” said Paul Harris, who chairs the Clay County DFL.
Bellwether counties on the decline
According to the Massachusetts Institute of Technology Election and Science Lab, the counties other than Clay that voted for every presidential winner since 2000 are Door County in Wisconsin; Delaware’s Kent County; Montana’s Blaine County; New Hampshire’s Hillsborough County and Virginia’s Chesapeake County. The other bellwethers are Essex and Saratoga counties in New York and Washington state’s Clallam County.
The number of bellwether counties dropped dramatically between 2016 and 2020. While 58 counties could be considered bellwethers for the elections between 2000 and 2016, the number dropped to nine in 2020, said MIT political science professor Charles Stewart III, who founded the Election Data and Science Lab.
[image_credit]Creative Commons/Ken Lund[/image_credit][image_caption]Since 2016, Moorhead residents became even more aligned with Democrats.[/image_caption]
Kondik said Trump has cited this as proof the 2020 election was stolen, but changes in demographics and political preferences in some of those former bellwether counties are the real cause.
A recent study of political bellwethers in Political Research Quarterly said there’s an inclination to “readily imagine that some counties demographically/politically mirror the nation, and thus monitoring this small set of counties closely would tell us with high predictive reliability what will happen nationally.”
But the study concluded that most bellwethers were not demographically representative of the nation as a whole, tending to be “whiter, older, less educated, have lower median incomes, have a lower percentage of workers in the labor force, and have higher rates of vacant housing.”
So why are they considered predictive of how the nation will vote in the race to the White House? Pure chance.
“The simple answer is that chance effects can be much greater than most people realize,” the study found.
What about Clay County
But what about Clay County?
Local officials think there are other reasons beyond chance for why the area might have voted for the presidential winner, at least in recent years. And actually, Clay County has sided with the winning presidential candidate for even longer, choosing Bill Clinton in 1996 and 1992. In 1988, Clay County voters picked Michael Dukakis over the actual winner of George H. W. Bush.
The county has about 65,000 residents and around 36,000 registered voters. It has a fairly abrupt split between its largest city, Moorhead, and rural, agricultural, areas. There are not a lot of suburbs – which are increasingly political battlegrounds – in between.
In the past, Moorhead actually had a conservative streak. Morrie Lanning — mayor of Moorhead from 1980 until 2001 — was elected to the Legislature several times as a Republican, serving from 2003 until 2012. Meanwhile, rural parts of the area had favored moderate Democrats.
That has largely been upended, especially since 2016, when rural voters broke in huge numbers for Trump and urban residents became even more aligned with Democrats.
Harris said Moorhead is the largest city in the western half of the state, and part of a “fairly cosmopolitan urban area” centered on Fargo. It’s a college town that has become more progressive in its politics in recent years. It’s also growing, making it a stronger force in regional politics.
Clay County’s presidential vote in 2016 and 2020 really did follow national trends.
When Trump won the presidency in 2016, he won only a slightly larger share of the vote in Clay County compared to Mitt Romney four years earlier. Instead, Clinton underperformed other Democrats, even in and around Moorhead, winning just 44% of the vote compared to Obama’s 52.6% in 2012.
[image_credit]Source: Minnesota Secretary of State[/image_credit][image_caption]Between the 2012 and 2016 presidential elections, Clay County's rural areas became more conservative as Moorhead remained strongly DFL.[/image_caption]
Rodney Johnson, who chairs the Clay County Republicans, said Trump inspired voters in 2016, while Clinton rubbed people the wrong way. “I think it’s a matter of the candidate that gets people the most stirred up or in some cases the candidate that has the least likeability,” he said.
Four years later, Trump actually won a tiny bit larger portion of the Clay County vote. But Biden performed far better than Clinton, winning more than 50% of the vote in the county, and swamping any advantage Trump had in the rural parts of the county with a bigger victory in Moorhead.
That tracks with urban, college educated voters being turned off by Trump’s presidency. Kondik said Clay County has better than average college education rates.
“(Clay County’s) four-year college attainment is a little over the national average —36% to 34% — which probably helps explain why the bottom hasn’t dropped out for Democrats there,” Kondik said.
Johnson said he felt people were more fired up to vote against Trump in 2020 because of unfavorable media coverage and campaigning from Democrats. Kenneth Foster, a political science professor at Concordia College in Moorhead, said the sizable city of Moorhead and the large rural area is one factor that creates “considerable variation in voter interests and priorities.”
But with such an evenly split county, “shifts by a small number of voters can change the outcome.”
What does it all mean for 2024, if anything? Johnson said voters will be motivated against Democrats because of this year’s legislative session, where DFL lawmakers passed a broad progressive agenda on issues like taxes, spending and abortion despite a narrow majority. “I really think the Democrats have overstepped by far any bounds of reasonableness to the point where they are going to turn some people off,” Johnson said.
Harris said he doesn’t like to make predictions. But a backlash isn’t out of the question. “I think in some ways it’s the political winds of the time that people tend to vote against the party in power,” he said.
How this applies to the 2016 el ection is interesting. It seems the county went for Trump, in an election where Clinton won the popular vote. So it seems it’s bellwether status is not based on any sort of ability to predict real numbers, but it’s ability to predict the entirely arbitrary vote in the electoral college.
Which is to say these paragraphs are the meat of the story:
“So why are they considered predictive of how the nation will vote in the race to the White House? Pure chance.
“The simple answer is that chance effects can be much greater than most people realize,” the study found.”
The county vote serves as a functional model of the results of the electoral College Nothard to understand, the county mirrors the results in swing states that provide the margin for the EC winner.
The interesting dynamic with Morehead of course, is that Fargo is right across the border. It seems to me that if you’re a republican from Morehead, you’ve probably moved to or are considering moving to Fargo to take advantage of their lower taxes, cheaper real estate, vibrant business climate, booming economy and other conservative governing principles, conceding Morehead as a liberal-infested college town and even more democrat-leaning relative to Fargo.
How did we get here in America, where social conservatives align with financial conservatives?
It just doesn’t make sense.
I’m just never going to really understand why working-class people increasingly vote in lock-step with wealthy elites.
They can’t seriously think their Guns/Gays/God/Race agenda is so important that they’re willing to just keep funneling their hard-earned money to millionaires and billionaires.
I mean, come on, how can they happily accept small short-term tax breaks for themselves while allowing permanent big tax breaks for the rich?
Because most working people know where their jobs come from and how they’re created. Government employees know their jobs are created by confiscating money from someone else.
Johnson says that voters will be motivated to vote against Democrats because of overstepping their bounds, so that seems to be the message Republicans will be using in the upcoming elections. I just hope the Democrats/DFLers can counter by showing how great this past legislative session was for Greater Minnesota… better rural broadband, increased aid to towns and cities, increased funding for wastewater and infrastructure and education, and plenty for farm programs. Rural voters were not benefitting from the stalemated outcomes of the previous legislatures, inaction mostly caused by Republicans.
Assume that there are no bellwether counties. In that case, we would not be having a conversation about bellwether counties.
Now assume an alternative where a reporter notices that dozen or so out of 60-some counties correlate with the presidential race two cycles in a row. The reporter files that away in the “maybe next cycle I’ll report that if it is still true” file. The next cycle a whopping 10 of those counties still predict the outcome! The phenomenon gets reported and it is now a thing. Or is it?
Now over the course of four or five cycles, the number of bellwether counties drops, for this reason: There were never bellweather counties. Just counties that happen to be right a couple of times in a row. But the story is not “oh, this is an expected outcome of how probability works” but rather “magical counties predict unknowable outcomes multiple times,” and now, “the well documented magical phenomenon is declining!”
Most years there are only two viable candidates, and most years it doesn’t take too much work to predict which one is slightly to moderately more likely to win. Under those conditions, any given set of “correct” counties will duplicate that result the next cycle, and over time, a small number of counties will, for utterly meaningless reasons, appear to have been correct several times in a row.
This is how zoos do it: They get all the animals, through some simple behavior, to predict an outcome. Most of them get it wrong but a handful get it right a few years in a row. Next thing you know, you’ve got some octopus on the Today Show predicting the outcome of the Academy Awards.
Why did I mention the case of no bellwether counties? Because that is what most things are. There are no bellwether telephone poles, bellwether painted turtles, or bellwether mosquito, not because these things don’t have magical powers (though they actually don’t) but because nobody has interrogated their stochastic outputs to see what they spuriously correlate to.