A lot of voters who skipped 2016 turned out in 2020 in Wisconsin. The data suggest that boosted Trump

This story was republished on Jan. 13, 2022 to make it free for all readers  

One of the hallmarks of last fall’s election was the multitude of new voters it drew to the polls. 

And that has left both parties wondering: How many of them will vote again in 2022?

Of the roughly 3.3 million Wisconsinites who cast a presidential ballot in 2020, almost a quarter — about 800,000 — did not participate in the 2016 presidential election, the state’s voter file shows. 

That is an enormous pool of new voters.  

It includes many people who weren’t eligible to vote here in 2016 because they were too young at the time or lived in another state.   

But it also includes a lot of Wisconsinites who skipped the 2016 election and other elections before that.

An analysis of the state’s voter file sheds some light on this latter group, often referred to as “lower-propensity” voters.

It doesn’t tell us how they voted, which is a private act. It does tell us where they came from, however.  

The data show that lower-frequency voters made up a larger share of the turnout in places where Donald Trump gained ground between 2016 and 2020.  Many of these were small, rural Wisconsin counties that saw bigger than average spikes in turnout last fall.

This pattern wasn’t powerful enough to carry Trump to victory here. Democratic turnout surged, too.

But it helps explain how some rural counties that are flat or declining in population generated much bigger turnouts and even bigger vote margins for Trump than they did when they swung so sharply behind him four years earlier.

This feat would have been a very big story had the election turned out slightly differently.

“The rural areas were almost the hero that Trump needed them to be,” said Brian Kind, a Republican data strategist based in Wisconsin.

“There were new voters on all fronts,” Kind said. But in his analysis, he found that rural areas made up a larger share of the statewide vote in 2020 than they did in 2016 — a difference he called small but significant.

(A Journal Sentinel analysis found that turnout — total votes cast — in all Wisconsin cities combined grew from 2016 to 2020 by 7.9%, while turnout in all Wisconsin towns combined grew by 13.6%).  

Will Trump voters keep turning out?

One obvious question coming out of 2020 is whether Republicans can mobilize lower-propensity Trump voters in a mid-term election when Trump is neither on the ballot nor in office.

“You can see from wards and counties with the largest share of first-time voters and of same-day registration voters, a lot of them were Republicans who weren’t answering calls from pollsters but did show up and vote” in 2020, said state Democratic chair Ben Wikler.

Wikler expressed skepticism that these pro-Trump voters with inconsistent voting histories would turn out in the 2022 mid-terms, when Wisconsin has marquee contests for US Senate and governor.

“We’ve seen even in the Georgia (US Senate) run-offs that when Trump is not himself a candidate, even if he’s holding rallies and endorsing people, it doesn’t have the same kind of electric effect (on his supporters) that it does when you can cast a ballot for him,” Wikler said.

State Republican chair Andrew Hitt agreed his party faces a turnout challenge in that regard, though he also views it as an opportunity, since these voters have added greatly to the party’s potential pool of support.

“Even when President Trump was still in office, there were challenges in bringing those people to the ballot box in 2018,” said Hitt, noting that more than 100,000 voters who turned out for Trump in 2016 failed to turn out for losing GOP Gov. Scott Walker in the 2018 midterm election.  

“And that’s the margin of about every major statewide race in Wisconsin,” Hitt said. “We need to be getting them engaged year-round, not just in presidential years.”  

This look at where Wisconsin’s legion of new voters came from last fall is based on the state voter file — a public database of who voted in 2020, whether those same people voted in past elections, and where they live.  

For this analysis, I used an enhanced version of the voter file compiled by L2, a nonpartisan election data firm.  L2 takes the pretty skeletal information on voters recorded by the state (names, addresses and individual turnout history) and adds its own layer of demographic and personal data.        

There are real limits to what the public voter file can tell us about turnout patterns because it contains no record of how individuals voted — only whether they voted.  

The limits are even more severe in Wisconsin because unlike many states, there is no record of which party an individual voter “belongs to,” since voters don’t register by party here. Without party registration, it's hard to say with certainty which side did a better job mobilizing casual voters. 

Instead, we’re left with home addresses, which allows us to compare shifts in the electorate in different kinds of communities — Republican, Democratic, urban, suburban, rural.

A closer look at 800,000 'new voters'

Let’s start with those 800,000 “new voters” in Wisconsin last year — more than 23% of the 2020 vote. These weren’t all first-time voters, but they were “new” in the sense they didn’t vote here in the previous presidential race.

Nationally, the share of “new” voters in their states was even higher than it was in Wisconsin — about 29%, according to a recent election report by the progressive data firm Catalist.

The turnover in the electorate was unusually big in 2020, with its record turnout.  But the churn in the voting pool from one election to another is routinely larger than many people realize, even in states like Wisconsin that aren’t growing very fast.

“The magnitude of these changes is poorly understood,” Catalist noted.

Who were all these non-2016 voters who cast ballots in Wisconsin in 2020?

A small chunk lived and voted in a different state in 2016.  But most were either young people who came of voting age after 2016, or were older, inconsistent voters who sat out the first Trump election but not the second one.

We don’t know how big each of these groups was. But the L2 voter file tells us that this third group — of older, lower-frequency voters — numbered more than 264,000 people. That's how many 2020 voters who skipped 2016 had some history of being registered to vote in the state during or prior to the 2016 campaign. In other words, they could have voted in 2016 but didn’t. Many of them voted in the 2012 presidential election or in the 2018 mid-terms.  But they weren't consistent November voters.

This group of more than a quarter-million Wisconsinites understates the number of low-propensity voters who skipped 2016 but voted in 2020, because it’s impossible to identify them all in the voter file. But it represents a significant slice of them, and comprises about 8% of the 2020 vote in Wisconsin.  

These voters were disproportionately Republican, according to L2’s estimates of voters’ partisanship based on other data it has gathered.

And their share of the vote last fall was bigger in counties where Trump gained ground between 2016 and 2020, an analysis of the data shows. 

The story of a small, rural county

Take for example, rural Grant County (pop. 51,000) in the far southwestern corner of the state. Democrat Barack Obama carried it by 14 points in 2012. Then Trump carried it by 9 points in 2016, a 23-point swing.  

Between 2016 and 2020, Grant County lost more than 1,000 people, according to census estimates. Yet it cast 1,240 more votes in 2020 than it did in 2016.  

The county’s more than 25,000 voters in 2020 included about 7,000 who hadn’t voted in 2016. Many of these were low-propensity voters who sat out the 2016 election. That helped Trump improve his raw vote margin in a county that got smaller. He won Grant by 3,144 votes in 2020, up from 2,299 votes in 2016. That’s not a big spike. But it came on top of the massive swing toward the GOP that occurred in 2016.

And Trump made similar gains in other small counties. Those gains added up in a state where four of the past 6 presidential elections have been decided by under one point.   

One other note about Grant County: roughly 12% of its voters registered on election day, the third highest figure in Wisconsin.  As Wikler noted, same-day registrations played a bigger role last fall in pro-Trump counties than in Democratic counties — the reversal of both 2016 and the state’s traditional pattern.  

One reason is that Democratic counties did more voting by mail in the pandemic than Republican counties. But another may be the surge of low-propensity voters in pro-Trump counties — people who were not registered to vote until they went to the polls.

Of the 20 counties with the highest share of voters who registered on Election Day, Trump won 19. And he won 18 of them by bigger margins than he did four years earlier.

The voter file shows that Democratic counties — and GOP counties where Trump lost ground in 2020 — had a lot of new voters, too. But their mix of new voters was different. Some of these counties have younger populations with more people aging into the electorate. Some have more people moving in for the first time.  And some have a history of higher turnout, so low-propensity voters were a smaller share of their 2020 vote.

Dane County is an example of all three. It’s also an example of how one big blue county that is adding lots of people can offset the Democratic Party’s erosion in lots of small counties like Grant.

For Republicans, there are two sides to their turnout success in many small and relatively rural parts of Wisconsin in 2020.  The considerable up side is that Trump drew lots of non-habitual voters to the polls, seemingly raising the party's ceiling of potential support.

The sobering side is that it wasn’t quite enough to win here last time due to other trends, especially a loss of support among suburban and college-educated voters. 

The shift in the Republican coalition may be making the party more reliant on a demographic group — non-college whites — that has historically lower voting rates, and includes a lot of voters who might not turn out without Trump on the ballot.

Neither party in next year's mid-term election will come close to the turnout levels they achieved last fall. But they'll both be scrambling to hang on to every new voter they got in 2020.