As I noted, the caucuses are not run by the state electoral machinery. In Iowa, if there's a primary or general election run by the state electoral machinery, under the aegis of the Secretary of State's Office, there's a 14-hour window when people can go vote at their polling place at a time of their choosing that's convenient for them. So typically, it's from 7:00 in the morning until 9:00 in the evening. That's the rather large window in which Iowa voters, whether primary or general election, can go cast a vote. That's not what caucuses do. As I noted, the caucuses represent, or they function as the business meetings of two private organizations. These meetings typically occur at 7 o'clock on a Monday night, anywhere from the middle of January to the middle of February. In 2020, it will be February 3rd. It's been as late as March, it's been as early as the beginning of January. And the two parties are part of a closed party system. That is unlike some states, where a voter can go in, say, in a primary and ask for a ballot of one party or the other, thinking, I want to vote in this party's primary or I want to vote in that party's primary. In Iowa, you have to be a registered member of the political party in whose caucus you want to participate in. Now, there are two caveats for that. One is that while we can look at the wonderful database that the secretary of state provides the first of every month, we get a new, updated web page showing the party registration for Republicans, Democrats. Independents are called no party in Iowa, and then others. We get an updated number for each of those, the first day or second day of each month. The caveats are, number one, that while you have to be a registered member of the political party to participate in its caucus, you can go to the caucus site and change your party registration on the spot. You're particularly interested in a particular candidate or an issue, you can stay a member of the party after the caucus. Or the very next week, you can go change back to your initial party. So there can be new people coming in who were not recorded previously as being registered members of the party. The second caveat is that a prospective participant can be 17, as long as he or she will be 18 as of the election in November. So there can be a number of people who technically aren't even registered to vote yet because they're not old enough. That being said, each caucus, each party will check people in. And then oftentimes the parties will open the session to representatives from particular candidates who might want to say a few words briefly in favor of their particular candidates. But then the Republicans and the Democrats do things very differently. Now, in 2016, 2008, we had big two-party caucuses. In 2012, there was no meaningful Democratic caucus, though there was a caucus, but President Obama wasn't contested for the nomination. And in 2020, President Trump, at least in any serious way, is not being contested here in Iowa for the nomination. So there will be a handful of Republicans who do show up for those two particular functions I mentioned, platform issues and county-level delegate selection. But in a contested year, what the Republicans will do was essentially conduct a straw poll. In other words, it's not an official vote, it's not an official election. People simply write down the name of the candidate they prefer to be their party's nominee, and it's collected by the officials at the precinct caucus in a box or whatever they happen to use. They're counted and recounted, and then called into the party's central headquarters. This led to a problem in 2012. There's no deliberate fraudulent activity. Nobody's trying to game the system. But in 2012, initially it was announced that Mitt Romney won that statewide preference vote among Republicans. Again, simply an expression of preference, not a formal vote or election of any sort, by something like eight votes. And then a week or two later, a box was discovered in one of the precincts that hadn't been counted and forwarded to the party headquarters. Then it turned out that Rick Santorum seemed to win the statewide preference vote among Republicans by 34 votes. So again, you have, I won't say amateurs, but it's not the official state election machinery doing this with voting machines. So there is the possibility for human error. But the Republican results are fairly clean and straightforward. Okay, here are all the preference votes, preferences expressed statewide, and so and so had this percentage, so and so had that percentage. It's pretty easy. What most people think of when they think of the Iowa caucuses is the Democratic caucuses, because here's where you do what I call a preference dance. At the Democratic caucuses, at the precinct, the first order of business is that a precinct chair is elected. There's a temporary chair, just to get things started, and a precinct chair is elected who can end up being that initial temporary chair. And then the first job of that precinct chair, with assistants, is to get a hard count of everybody registered and eligible to participate in that caucus. That's important because that's a denominator. So they get a count of how many people are registered to participate in that particular precinct's caucus. Then the chair will say, using this year's candidates, okay, everybody in favor of Sanders go to one corner. Everybody in favor of Warren go to another corner, Klobuchar another corner, Biden another corner, and so forth. So it's very public, as opposed to a kind of secret ballot in which you write down somebody's name on a slip of paper, as the Republicans do. Here you visibly have to stand up in public, aligning yourself with a particular candidate. You are entitled to, and you are allowed to go with the group that says simply no preference. One of the things we have to remember is that what put the Iowa caucuses on the political map of the country was the fact that Jimmy who, in other words, Jimmy Carter had worked the state for 14 months under the radar in 1975 into 1976. And he had got the Democratic nomination, and was elected president. Well, the reason that's important is that Carter technically came in second in the Iowa caucus counting. And the counting is very complex, but we'll leave that for the moment. Carter finished second behind no preference. So for the Democrats, you are allowed to mark and stand for no preference rather than a particular candidate. Once this first alignment is done, here's where that denominator comes in. It varies by the size of the precinct. Smaller precincts have a larger viability percentage that's required. But in the large precincts, everybody hears the number 15%. So given the number of people at that precinct caucus who are registered and allowed to participate, if your preference group does not meet that 15% threshold, you're declared nonviable. And what happens is that you can try to entice others to go ahead and realign and join your group. You can dissolve your group and go to no preference. Or the members of that group, who themselves don't make up enough to meet that threshold, they are enticed to go over to somebody else's preference group. So that's the particular dance that's done. Once this is completed, and again, the counting is very complex. We should come back to that. Then the precinct is able to try to figure out some way of reporting results. In terms of how we tabulate these results, as I've said, what the precincts choose is not delegates to the national convention. What they choose is what's called state delegate equivalents, and these are determined by reference to how many delegates each precinct sends to the county convention. How many delegates a precinct might get is determined by the votes for a Democratic candidate in that precinct in the last gubernatorial or last presidential election. But just to give you an indication, in 2016, Iowa, for the Democratic National Convention, had 51 national convention delegates, 51. The number of state delegate equivalents in 2016, the total were about 700 state delegate equivalents assigned to Clinton, and 696 state delegate equivalents assigned to Sanders. So you see, nothing to do directly with those 51 national convention delegates. And indeed, in 2016, those 1,681 Democratic precincts sent 11,000 delegates overall to the county conventions. So you can see how this becomes very, very tricky and complex. This is why we called, or subtitled our book The Making of a Media Event, because reporters want to know a number to figure out who won. And I'll explain something about how we determine who won in a moment. But that being said, this year, there's a bit of a change. In the past, the numbers of people in those preference groups were never reported. All the Democrats reported was the state delegate equivalents. This time around, the Democrats will report the hard numbers for that initial alignment, even if it's less than that threshold, 15%, 25%, whatever it might be for that particular precinct. They'll report that number. Then after that realignment, when everybody moves around after determining some groups are nonviable, they'll report that number, what percentage of the people, how many people supported which candidate. And then they will support, using their formula, that state delegate equivalent total. They're supposed to report all three at the same time. What this does, of course, is take a process that's relatively complex and confusing and, I would suggest, makes it even more complex and confusing. But reporters will have to figure out what set of numbers do they trust, do they find meaningful.
