As the first voting event of the 2016 Democratic presidential nominating process grows nigh, polls from Iowa are pouring in, but they reflect a fundamental difference of opinion in the standing of two late-septuagenarian men. In the last couple of weeks, surveys from Focus on Rural America, USA Today–Suffolk, and Monmouth showed Joe Biden in the lead; all but FRA has Bernie Sanders running second. ISU–Civiqs, Emerson, New York Times–Siena, and CBS–You Gov showed Sanders in the lead, with Biden ranked anywhere from second to fourth.
It’s becoming increasingly clear that these variable assessments of the relative strength of Biden and Sanders reflect different assessments of who’s going to show up on Caucus Night. As I noted earlier this week, the New York Times’ Nate Cohn makes a compelling case that Biden-friendly polls tend to assume past Democratic primary voters will dominate the caucuses, while Bernie-friendly polls (including the Times–Siena survey) tend to focus on self-reported interest in caucusing, which is a very different animal. As Cohn acknowledged, the universe of validated past voters (not caucusgoers, but voters) leans heavily toward Biden:
Overlaying this big gap in the prior experiences and future intentions of poll respondents backing these two men is the rather shocking age divide between them. I’ve never seen anything quite like the numbers in this recent national poll:
But there were splits of a similar nature in a fairly recent set of Iowa caucuses in which the surprising size and shape of the universe of caucus attendees was the big story: 2008, when Obama’s win in Iowa placed him on the path to a two-term presidency. As Politico noted at the time, it was a turnout tsunami:
[A]bout 234,000 Democrats turned out Thursday night, shattering the previous mark of 124,000 …
[A]mong those independents who participated in the Democratic contest, Obama won them by roughly a 2-to-1 ratio over Clinton and John Edwards.
It was first-time caucus-goers in the Democratic contest that set a new record: 56 percent, 11 percent higher than 2004.
According to entrance polls, Obama beat Clinton with under-30 voters by a 57–11 margin. These voters represented 22 percent of caucusgoers, compared to an estimated 17 percent in 2004. If you read accounts of that campaign, what basically happened is that the Clinton campaign — under the direction of legendary field organizer Teresa Vilmain — hit all its turnout “marks,” but was overwhelmed by the wave of young, first-time caucusgoers (including an unusual number of indies) organized and deployed by the Obama campaign.
This has to be the model for Bernie Sanders, whose strength among young, first-time caucusgoers, particularly independents, may be even stronger than it was in 2016, relative to his overall levels of support. In 2016, Sanders won under-30 voters by an 84–14 margin, but the share of total caucusgoers they represented was back down to 18 percent, and total participation was down from 234,000 in 2008 to 171,000. Had total (or under-30) turnout been at 2008 levels, Sanders would likely have won the caucuses comfortably. Meanwhile, Clinton won over-65 voters, who represented 28 percent of caucusgoers, by a 69–28 margin.
So the big question for the February 3 Caucus Night is: Who will show up? Every indication of enthusiasm and engagement points toward a big turnout among both old and new caucusgoers, including (perhaps) older primary voters who may not normally sit through the hardy rituals of the caucuses.
Campaign organizations will, of course, work to shape that turnout, and then the final results could depend on the second choices of candidate supporters who don’t reach the viability threshold in the first round of caucusing at any given place. This is where, in the post-caucus spin, the possibility of multiple “winners” could become a factor, since the Iowa Democratic Party will be announcing total first-round “raw vote” support and total second-round “raw vote” support along with the traditional measurement of delegates won, as I recently explained:
Iowa’s coerced decision to begin reporting raw caucus votes was directly the product of post-caucus spinning in 2016 by supporters of Bernie Sanders who claimed their candidate had “really” won in terms of the total number of caucus participants backing him. Since the state party did not compile, much less publicize, raw votes, there was no way to confirm or deny that claim, which of course fed accusations the process was “rigged” for Hillary Clinton.
Something like this could happen again, notes Cohn:
If you are an Iowan or an observer at the caucuses, check out the lines to sign in (where, among other things, independents and even Republicans can change their party-registration status on the spot in order to participate), and their length and composition may tell you a lot. There are even some signs the famously pale complexion of the caucusgoers could notably change this year as Iowa’s growing Latino population makes its presence known. But most of all, if caucus sites are flooded by young people, as they were in 2008, it’s likely to be a Bernie-riffic night.
