Biden’s Iowa Problem: Our Poll Suggests His Voters Aren’t the Caucusing Type

With one week to go until the Iowa caucuses, Joe Biden finds himself in a seemingly unimaginable position: needing to mobilize older voters who are usually the most consistent voters.

This helps explain why recent polls diverge so much on Mr. Biden’s standing in Iowa.

He trailed in the New York Times/Siena College poll of Iowa caucusgoers released Saturday. Yet he very well might have led a Times/Siena poll of a hypothetical Iowa primary, because he had a substantial advantage among Democrats in the survey who regularly turn out in primary elections.

But the Iowa race is a caucus, not a primary. And while analysts know very little about the nature of caucus electorates, there is reason to think that caucuses attract very different kinds of voters.

On the one hand, many regular voters don’t participate in caucuses. One-third of people who participated in a recent Iowa Democratic primary — usually considered all but certain voters — said they weren’t likely to caucus, and these voters backed Mr. Biden by 11 points in the Times/Siena poll. The large number of Democratic primary voters who seem uninterested in attending the Iowa caucuses is all the more surprising given that recent Iowa primary elections for Senate, House, governor or local offices were not particularly competitive and certainly did not attract national attention.

On the other hand, a presidential caucus — and perhaps especially Bernie Sanders — draws a large number of voters to the polls who do not vote regularly. As a result, it was Mr. Sanders who led the Times/Siena poll — by seven points — despite weakness among regular and consistent voters. His supporters may not vote so often, but he led among those who said they had caucused before, including in 2016. Most important, he led among those who said they would show up next week, and therefore led the poll.

This mismatch — between the voters who say they will participate in a caucus, and the voters who typically show up in primaries — may be at the heart of the wide split in recent Iowa polls.

Many pollsters rely, in some way, on past vote history to conduct their surveys. Some pollsters use it to define which voters could be selected to participate in a survey, like a recent Monmouth University poll that selected registered Democrats or independents who turned out in 2018 or in a recent primary, or who registered since 2018. A Neighborhood Research and Media poll was even more limited in its model for who was likely to vote: voters who turned out in either the 2016 or 2018 primary. These polls in Iowa showed Mr. Biden with the lead, and the Times/Siena poll also found Mr. Biden tied or ahead among these groups.

The Times/Siena poll does not use past vote history to define the universe of voters eligible to participate in the caucus poll. Neither do the Ann Selzer/CNN/Des Moines Register or CBS/YouGov polls, which also show Mr. Sanders in the lead.

The question is whether it is reasonable to assume that the caucus electorate is largely confined to these groups of highly regular voters. It’s an understandable assumption: Caucuses are very low-turnout elections, and so one might assume they are mainly composed of the most reliable voters. One might even note that the 2018 midterm primaries in Iowa had about the same number of voters as the 2016 Democratic presidential caucus, and suppose they’re probably the same people as well.

If that assumption is right, the Times/Siena poll and others showing Mr. Sanders in the lead include many irregular young voters with little chance of actually showing up. Over all, voters over age 65 made up 42 percent of the Iowa primary electorate in 2018, compared with 24 percent in the Times/Siena poll. And in fairness, it is hard to know with certainty either way, given the lack of authoritative public data on the composition of caucus electorates in Iowa or elsewhere.

But there is reason to think caucus electorates are fundamentally different from primaries or even typical elections.

Just consider the actual results of recent caucuses. In both Democratic and Republican contests, they are plainly different from the results of primaries in demographically similar areas, or even in the same states. They tend to favor outsider, activist-backed candidates, including Mr. Sanders. The only plausible explanation for the difference is that low-turnout caucuses tend to attract the sort of highly informed political activists who are passionate enough to sit through a laborious, multi-hour, public caucus and favor ideologically consistent candidates.

One reason for the difference seems to be age. Caucus electorates appear to be fairly young, and primary electorates fairly or even very old. This shouldn’t be a huge surprise. The kinds of Democrats who have excelled in caucuses in recent years, like Mr. Sanders and Barack Obama, were backed overwhelmingly by younger voters. The young represented a large share of the caucus electorates in the 2004, 2008 and 2016 entrance polls — the only publicly available data on the composition of the caucus electorate, even if it’s of uncertain quality.

The results of the Times/Siena poll of Iowa also indicate that past caucusgoers are quite different from past primary voters — at least if you believe our respondents, since we can’t prove they actually participated.

Past caucusgoers back Mr. Sanders by four points, even as past primary voters support Mr. Biden. Voters who say they participated in the 2016 Democratic caucus, specifically, back Mr. Sanders by nine points; he’s at 25 percent and Pete Buttigieg at 16 percent. These 2016 caucusgoers say they voted for Hillary Clinton over Mr. Sanders, 49 percent to 46 percent, with 3 percent for Martin O’Malley — very close to the actual result, perhaps lending credibility to the overall Times/Siena finding.

The poll indicates that caucusgoers are poised to behave differently from the broader electorate yet again in 2020. The Democrats who say they will caucus back Mr. Sanders; those who say they won’t attend a caucus back Mr. Biden by a six-point margin.

Mr. Biden’s turnout disadvantage persists even among the voters who said they would caucus. In a second likely-voter question — which asks Democratic caucusgoers to evaluate their likelihood of voting — Mr. Biden does worst among the voters who say they’re “almost certain” to vote, and falls to just 15 percent of the vote. On the other hand, he is at 30 percent among those who said they would caucus but specified in the follow-up question that they were “somewhat” or “less likely” to caucus.

In the Times/Siena poll, younger voters are likelier than older voters to say they will participate in the Democratic caucus. Among registered Democrats, 53 percent of those 18 to 29 said they were “almost certain” or “very likely” to caucus, compared with 35 percent over age 65. It’s not because this is just a group of exceptionally high-turnout young people: 58 percent of these same 18-to-29-year-old Democrats turned out in 2018, according to data from the L2 voter file, while 85 percent of the respondents over age 65 were voters in 2018.

Self-reported vote intention is far from perfect; people can mislead pollsters, intentionally or unintentionally. But the finding should be taken seriously. Of the more than 100 pre-election Times/Siena surveys in the last four years, the two Iowa caucus polls stand out as the two cases where self-reported turnout among young voters significantly outpaced that of older voters, and it is not a close call. In the only two other cases where self-reported youth turnout was higher, the samples of young voters were far smaller, and so was the difference.

Of course, there’s no guarantee that youth turnout will be so high. But for Mr. Sanders, it might not need to be. He would have led the Times/Siena poll even without any 18-to-29-year-old voters at all. His strength among older voters is enhanced in a caucus setting as well, because even older caucusgoers tend to be more liberal than older Democrats who say they intend to stay home.

If Mr. Biden can mobilize older, regular voters over the next week, it might be a different story.

Source link