Previewing the Iowa Caucuses
Thursday, January 3, 2008; 11:00 AM
Former majority staff director of the Iowa Senate Christopher C. Hull, who recently wrote " Grassroots Rules: How the Iowa Caucus Helps Elect American Presidents," was online Thursday, Jan. 3 at 11 a.m. ET from Iowa to answer readers' questions about the caucuses, how they work, and how he sees them playing out Thursday night.
The transcript follows.
Washington: Let's get to it: What are your predictions for both caucuses tonight?
Christopher C. Hull: Sorry for the delay, all. We're having technical problems.
I just happen to have completed my rough statistical estimate of the results, which follows:
Estimated Democratic Results
Obama 35 percent
Clinton 25 percent
Edwards 23 percent
Richardson 6 percent
Biden 4 percent
Kucinich 0 percent
Estimated Republican Results
Romney 39 percent
Huckabee 32 percent
Paul 11 percent
McCain 9 percent
Thompson 5 percent
Giuliani 0 percent
Know that I don't claim this as particularly accurate or prescient. It IS the result of a complex statistical model built over many years, though, so might be of some interest. Happy to answer further questions about it.
And, for the record this is the tagline:
Christopher C. Hull, Ph.D., Andrea Mayer, and Briana R. Morgan
Georgetown University Department of Government
Statistical Estimates of 2008 Iowa Caucus Results
Completed January 3rd, 2008, 6:09 AM
Methodology from Grassroots Rules: How the Iowa Caucus Helps Elect American Presidents
Stanford University Press, 2007
Dayton, Ohio: It seems to me that the Iowa Caucuses are a boon for the state with questionable return on investment for America. Candidates spend tens of millions to attract 4 percent to 6 percent of dedicated Iowans. How can we better balance the process? Thanks for your consideration.
Christopher C. Hull: Certainly there's a concern anytime so much money is spent in one state at a single time. There are two ways of balancing the process, in my view.
One is to move in the direction of a national primary -- whether that is by compressing states into regional contests or by actually allowing all states to compete at once. My serious concern about that is that it eliminates the winnowing process that currently takes place, during which different states learn about the candidates sequentially.
The other way would be to improve the Caucus' process itself. In the book I recommend tying delegates to the percentages, bringing the state in to regulate the process, and eliminating the Democratic caucuses' silly 15 percent viability threshold.
The return on investment for America, at that point, is to cull out candidates with little support from the grassroots activists of the party.
That may not address your fundamental concern -- but we do need to be careful, it seems to me, not to sacrifice the upsides of the current process in our drive to change it.
Lyme, Conn.: Because caucus meetings allow for shifting of people who leave candidates who fail to achieve a threshold, or perhaps shift to allow a candidate to achieve a threshold required to obtain a delegate, are these initial results tabulated and kept, or are only the final results reported? If these initial results are not kept, is the press observing, and has the press recorded initial votes and how much shifting occurs? If so, how much shifting does one see at these meetings? It would be interesting to see a political science study of these shifts and to learn the factors behind these shifts.
Christopher C. Hull: The initial results are not kept, nor are the raw total vote counts on the Democratic side, and the press is not recording what those totals are, as there are 1780-plus precinct caucuses going on. Jonathan Alter from Newsweek just asked me this very question, and I had to admit we as scholars have no way of seeing this crucial information.
My recommendation to the Iowa Democratic Party, actually, would be to eliminate their 15 percent viability threshold altogether. That would take care of our curiosity!
Washington: Around what time tonight will we start getting results? Will they trickle out like in normal elections, or do caucus results arrive all at once?
Christopher C. Hull: Oh, this is hard-core trickle. It will take a long, long time to get the Democratic results in, as the caucuses can take as long as three hours, and then the precinct leadership will have to get the results reported back to headquarters -- and then the party still has to total up their State Delegate Equivalents (SDEs), the measurement of support which they actually report. Expect the GOP results to be in first, as the contest is more simple both at the precinct and state-wide level.
I'd guess 9 p.m.-10 p.m. for Republicans and 11-midnight for Democrats.
Iowa: Good morning, looking forward to the caucuses today. I am from Scotland, so it is interesting reading how they work! Hope they were not counting on the weather helping the attendance -- it's hovering around six degrees right now. How do you feel the people with the candidates that get less than 15 percent will realign to? It seems that if you are for Hillary then you are already in her corner; not many people who were originally with Richards, Biden or Dodd will migrate to her. Even if she goes in level, if she does not attract those others, then it's a straight race between Obama and Edwards.
Christopher C. Hull: My sense is that the beneficiaries of the second-place phenomenon will be Barack Obama especially, and John Edwards to a lesser extent. I totally agree with you that few non-Hillary backers will move to her if their candidate falls below the 15 percent threshold of support.
I do think that Sen. Clinton's support is strong enough to push her past Edwards even with this effect, however. We'll see if I'm right!
Boston: Aren't only hardcore partisans represented in the Iowa caucus if only 6 percent of eligible voters go to what is a fairly time-consuming and cumbersome process? And who came up with the idea of making people stand up in front of the community and to state who they support? So much for making whatever choice you wish anonymously. If these caucuses are as democratically challenged as Karen Tumulty in Time suggests, why should the rest of the country and the national pundits care what the results are tonight? To see what the fringe of the various parties in some small rural state thinks about the candidates?
Christopher C. Hull: Gosh, this 6 percent figure is everywhere. In the best-attended years the figure can be 10 percent or 11 percent, and I'd expect the Democratic caucus this year to potentially even top that. The figure sounds low, granted, but it's very high relative to other caucus states. Imagine a primary in which you would be required to show up at one exact time -- and potentially give a speech in favor of the candidate you voted for. It's an intimidating process that keeps turnout to, as you say, hardcore partisans.
Regardless, that begs your question: in fact, absolutely, the Iowa caucus attracts only the most committed supporters of either the party or a given candidate. Don't forget the latter, though -- Barack Obama, for instance, is drawing enormous support from political independents in Iowa who may Caucus for him as Democrats, but their loyalty is to him, not to the party. Likewise, Hillary Clinton is drawing many women into the Caucus who are inspired by her, not by partisanship.
You're also right, though, that the Caucuses are not the most democratic of contests. The answer to your question of why people should care is exactly because Iowa is an acid test of partisanship and/or mobilization. Only those who care passionately about their party or their candidate will show up -- which means candidates who don't have a draw to either will suffer badly. Rudy Giuliani might be an example -- someone with very high favorables within the GOP, but without a core of activists that support him.
If you are opposed to partisans having their say about their party's nominees, Iowa is a very serious concern. But it is, after all, their party.
Anyway, just to point out the other side of the argument.
Philadelphia: I notice recent polls have even some major candidates like Thompson at 12 percent, Giuliani at 12 percent, Paul at 9 percent, etc. If this holds consistently statewide and they consistently fail to obtain 15 percent at caucus meetings, doesn't this mean their final percentages may be far below what they are polling? Will the public understand this, or might this potentially be fatal for the candidates who collapse in the Iowa caucuses?
Christopher C. Hull: Well, the first important note here is that the 15 percent viability threshold is only imposed on the Democratic side. So Thompson, Giuliani, and Paul will be unaffected by it.
That said, your point is absolutely correct - the examples would be Gov. Bill Richardson (the former U.S. ambassador to the United Nations), Sen. Chris Dodd (a 26-year veteran of the Senate), and Sen. Joe Biden (a 30-year veteran of the Senate and Chair of the Senate Foreign Relations Committee).
I think you are right that each of these candidates could be mortally wounded in Iowa, in large part because of this one Caucus quirk. And I don't think that's right.
Dallas: Do you think Huckabee is getting his approval with his approach in Iraq? I think he is the only candidate who will not "stay the course"' as this is such a hot topic with voters, could this be the reason that he is in the lead?
Christopher C. Hull: Can't say as I know. My own guess would be that Huckabee's social conservatism is driving his support in the state. But you may be right.
Arlington, Va.: Mr. Hull, do you think the Iowa Caucuses distract from and substantial discussion of candidates' policies? It seems to me that TV is saturated with too-short interviews of candidates eating breakfast at diners, and radio and the print are full of who-said-what-about-who type commentary and horse-race-type analysis. Frankly I think it's going to take real work for the voter to find the information to compare and contrast candidates' policy stances.
Christopher C. Hull: Certainly the media is heavily focused on horse-race politics in Iowa, agreed. That said, I'd say that if we were to actually peel back one of these town hall meetings, we'd find very serious policy discussions going on. Remember, these are committed and heavily ideological activists.
Try C-SPAN's coverage, for instance, and see if I'm right.
Baltimore: I understand (sort of) the complex system the Democratic voters use with the second choices counting and viability, etc. But what's the Republican system "straw poll" like? How is it different from a normal primary vote?
Christopher C. Hull: Good question. The GOP straw poll system is different from a primary because it takes place at a set time, is held in the context of a party meeting where other business is conducted, and is surrounded by speeches of support for various candidates. The feel is very different from a primary.
That said, you're on to something -- the Republican process is much more like a primary than the Democratic one is in the Hawkeye State.
Washington, D.C.: Can you give an indication of what kind of timeline we should expect for getting results? During the caucuses, I guess we'll get speculation based on turnout but when do hard numbers start coming in and when should we see the process finalized? Thanks!
Christopher C. Hull: In case you didn't see before -- I bet you were caught in the delay along with me -- my guess is 9 p.m.-10 p.m. CT for the GOP and 11 p.m.-midnight CT for the Democrats.
Oradell, N.J.: Which counties in the state have the most disproportionate influence in the delegate distribution on the Democratic side, and which candidate will benefit most from that distribution of delegates?
Christopher C. Hull: Ooh, fabulous question! It depends on what you mean by disproportionate. In terms of housing large numbers of Democrats, Polk County -- which contains Des Moines is the titan of the Caucus. Democrats are heavily implanted in that area. My own bet is that Sen. Clinton is pretty popular there (here!).
However, the other way to look at disproportionate impact is that the Iowa Democratic party weights rural areas more heavily than urban ones. So expect candidates who spent more time in rural areas like John Edwards to outperform on that score.
Annandale, Va.: How is it fair that a couple of states determine who gets to run for president? They do not reflect my values or politics, yet play one of the most important roles in the process.
Christopher C. Hull: I think you're right that "fair" is not the first word that comes to mind when considering the geographic distribution of influence in the presidential nominating process.
In my book ("Grassroots Rules: How the Iowa Caucus Helps Elect American Presidents," for the record) I argue that the Hawkeye State plays a particular role, forcing candidates to rely on retail politics in a way other states don't.
But, still, it probably doesn't meet your test of equity, regardless.
Princeton, N.J.: Gail Collins in the New York Times has a column on why we should ignore Iowa. Here is a brief quote: "People, ignore whatever happens here. The identity of the next leader of the most powerful nation in the world is not supposed to depend on the opinion of one small state. Let alone the sliver of that state with the leisure and physical capacity to make a personal appearance tonight at a local caucus that begins at precisely 7 o'clock. Let alone the tiny slice of the small sliver willing to take part in a process that involves standing up in public to show a political preference, while being lobbied and nagged by neighbors."
Sitting in a town that I am sure is quite different than any town in Iowa, she makes a lot of sense to me.
washingtonpost.com: The Slice of the Sliver Speaks (New York Times, Jan. 3)
Christopher C. Hull: Princeton, N.J., is actually very like a lot of the academic communities in Iowa -- Iowa City comes to mind. I adore Princeton, and think it's one of the most terrific places in the country, actually. But you'd be surprised, I think, by how coastal the Iowa Democratic Party is: highly educated, anti-war, sympathetic to progressive causes.
Feel free to ignore the results if you like, of course. Ms. Collins certainly has a point.
Mt. Lebanon, Pa.: Let's cut to the chase. Re: Returns -- what time will be too early to see anything important happen tonight on TV? What time will be too late? No, I don't want to hear noodles from the press opine about who could do what. Nor do I want to hear them explain in a thousand ways what happened. I only want to see the victors crowned and the losers led away in suck-it-up dignity. When will that occur? Thanks much.
Christopher C. Hull: (Know I already answered this, but just to keep things clean, again it's 9 p.m.-10 p.m. CT for the GOP and 11 p.m.-midnight for the Democrats. Boy, I hope I'm right on this!)
Not Iowa: Everybody goes on about "ground games" and "organization" and "get out the vote," but if Huckabee does well with little such organization, does that invalidate the millions of dollars others spent on their ground game? Or does that mean that if he'd had the money and people to combine with his charm he would have won handily?
Christopher C. Hull: There is a serious question as to the value of just dumping large dollar figures in Iowa. Television, the spendthrift's choice of tactics there, may have a negative effect on candidates' performance -- at least, according to the analysis in my book.
Also, don't discount the Christian community's internal organization in the Hawkeye State. Though Huckabee didn't create it, it still exists, and is powerful.
Cedar Rapids, Iowa: As an Iowan, I haven't made my mind up yet, and for me it comes down to "who would you rather have a beer with?" Have you had a beer with any of these candidates, and if so could you tell me which one was the best one to have a beer with?
Christopher C. Hull: LOL -- no, I haven't had a beer with any of them. I'd sure like to have a beer with Joe Biden, though, wouldn't you? And certainly Barack Obama would be on my list -- seems like both an inspiring and approachable guy. And Rudy Giuliani. I'm not sure whether Mike Huckabee drinks beer, but if he did, I'm sure that would be fun, too.
Triangle, N.C.: What's your take on the rumors that Biden and Richardson will tell their supporters to vote for Obama if they fall under 15 percent, as Kucinich did?
Christopher C. Hull: My understanding is that Biden's people are being directed to vote "undecided" if they fall under the threshold. I don't know about Richardson, but I would doubt that -- I think he's campaigning to be Vice President, and Hillary Clinton would be the safer bet if he were to do something like that.
Avon Park, Fla.: I must say that I'm a little disturbed by the fact that the Iowa Democrats caucuses aren't secret ballots like the Iowa Republicans' are. The GOP appears to have an easier system. Has there been talk about simplifying the Democratic caucuses? If not, what reasons have people given for keeping it that way?
Christopher C. Hull: There is talk every four years about simplifying the system, but the Iowa Democratic Party seems to like it this way. Their argument, from my interviews, is that the complexity requires even more organizational skill to handle, which they view as a positive thing. In my book I do recommend eliminating the 15 percent threshold, as I've said, but I do recognize the value in what the IDP says.
Washington: Hello. Thank you for your predictions -- I'm curious about the model you used to make them. What variables do you include? How are they weighted or otherwise adjusted? I'd love to hear more about it, seems like something very different from the usual polls. Thank you!
Christopher C. Hull: Aha! For that, you'll have to buy the book! The models are laid out in detail there...
Okay, okay, I know that was a cheap plug. The short answer is that the Predictive Model is based on four things: The Iowa Poll, results in state-wide straw polls, the Gallup Poll (which has a negative effect in predictions), and the results of my Explanatory Model, which take into account candidate level of effort on a number of fronts.
This prediction is hampered by a lack of data on a number of fronts, including perceptions of electability and ideological crowding. That makes the GOP estimates especially suspect, I think. Huckabee's ideological room on the Right may have a dramatic impact on his performance, pushing him far past Romney, if that's true.
Philadelphia: I notice the names missing from your estimate -- such as Dodd, Hunter, etc. Are they also expected to draw 0 percent?
Christopher C. Hull: Actually, that's just from lack of data. But I don't expect them to do well.
Rolla, Mo.: I assume there is a federal law that mandates that employers have to allow employees to caucus. If people who have to work evenings (which tend to be more blue-collar or lower income) can't show up in proportional numbers to the rest of the population, does this hurt a pure populist like Edwards?
Christopher C. Hull: No, I don't think there is such a law. This process is not official -- just a party meeting. That's part of its infuriating charm.
Yes, I think populists are harmed by the process for exactly the reason you state. And Dick Gephardt apparently suffered from that exact fate in 2004, again according to Jonathan Alter from Newsweek.
Anonymous: There are some obvious striking aspects of your predictions. This morning's analysts on TV were predicting a McCain resurgence, a Huckabee win, and Edwards coming in second. If your predictions come in, this has to hurt Edwards, McCain, and just about any candidate except for the top two in each party (and possibly Giuliani who mostly skipped Iowa). If your predictions are accurate, how do you see this shaping the field, especially in the face of "expectations" that the press seems to be building that candidates such as McCain, Thompson, Paul and Edwards should be doing better than what you predict?
Christopher C. Hull: Oh, I wouldn't put that much stock in these estimates, to tell you the truth. They're interesting because they tell you what the data say -- but I don't know that Huckabee will actually lose, or that McCain will come in fourth behind Paul. Let's just see how folks do, and not put too much stock in these numbers -- yet.
(Note -- in 2004 a version of this model called the top-tier results in order: Kerry, Edwards, Dean, Gephardt...)
Denver: Colorado has caucuses for nominating candidates at state level and the attendance is very poor. I think the attendance in Iowa is also very poor -- what is it, below 10 percent? If there is any tinkering to be done, maybe they should just move to a primary rather than caucus. As for Democrats' 15 percent viability, I think that is better than having just raw percentages. This 15 percent is at the precinct level, so there is plenty of opportunity for voters to choose a second candidate.
Christopher C. Hull: My own view is that the caucus process adds something important to the nomination contest. But I recognize the concern about low turnout.
Silver Spring, Md.: Paul over McCain: If true, would this result mean anything to either candidate outside of Iowa and looking forward? Thanks!
Christopher C. Hull: Yes, I think so. Again, we shouldn't get carried away interpreting these results -- let's see what happens. But certainly losing to Paul wouldn't help McCain any in New Hampshire, especially if Romney does in fact win.
Washington: I think the viability rule is linked loosely to the delegates awarded, so roughly 15 percent is needed to get one delegate to the next level of the process. Without the viability rule, if your candidate gets 14 percent, then you go home. With the rule, you can join another candidate's caucus or, to avoid the my-way-or-the-highway scenario when one candidate dominates, several non-viables can create an uncommitted group after the first round. With so many candidates splitting the vote, though (24, 25, 26, etc.), it is possible for all candidates to fail viability in some areas and then dominate the process in other regions.
Christopher C. Hull: Hm. I don't think so, actually -- the delegates awarded are linked to population and Democratic vote, I think, and so don't correspond with that 15 percent figure. But I could be wrong, as that's pretty far into the arcana.
Annandale, Va.: Do you think the Orange Bowl will have any affect on the turnout? Will it help Hillary since the woman might caucus while the men watch football? I am a political junkie and would stay home to watch the game rather than spend three hours trying to vote.
Christopher C. Hull: Great question! Sure, it's too bad that this process is also discriminating against college ball fans -- though of course personally I avoid what I call "amateur football." I'm sure that will start an entirely new, violent debate here.
Washington: You expect results at "9 p.m.-10 p.m. for Republicans and 11 p.m.-midnight for Democrats." Is that local or D.C. time?
Christopher C. Hull: Central.
Alpharetta, Ga.: Do you think caucus results might raise questions about whether McCain or Rudy should have competed more aggressively in the state?
Christopher C. Hull: Hell yes -- Giuliani especially. I'm not sure why his campaign continues to stick with this late-state strategy, while his polls erode not just in early states, not just nationally but in the very states where he's ounting on winning.
Albany, N.Y.: One of the major issues among other polling wonks has been how to identify "probable caucusees" among respondents. Can you say a little bit about how you identified these folks and how your approach differs from other estimates, such as the recent Des Moines Register poll that has caused something of a flap?
Christopher C. Hull: Oh, my estimates aren't polls at all, remember. They use the Iowa Poll -- the Des Moines Register poll you mention -- which has the best caucus attendee filter in the business, in my view, though.
What's the flap over the Iowa Poll, by the way?
Philadelphia: Are Republicans allowed to vote in the Iowa Democratic Caucus, and may Democrats vote in the Iowa Republican Caucus? The press is telling of Republicans voting for Obama. Is this permitted, or would the Republicans have to switch their registration to being Democrats? If they have to switch, how far in advance of the caucus would they have to have switched in order to vote today?
Christopher C. Hull: Yes, if a voter is willing to change parties at the caucus site, they may. So yes, Republicans (and Independents) may well have some impact on Obama's results.
Bremerton, Wash.: It's the A-caucu-lypse! Can you tell me if the "15 percent viability" rule came from Bygone Days, or were they put in relatively recently? And I'll predict Edwards and Romney, with Richardson and Paul doing better than expected.
Christopher C. Hull: I'm actually not sure. My understanding is that the viability rule was in place in the mid-1970s -- but it may have come about from the Democratic Primary reforms after 1968. Good question to which I don't have the answer.
Anonymous: Did Ohio Gov. Strickland's bashing of the Iowa caucus the day after he stumped for Hillary in Iowa resonate with Iowans?
Christopher C. Hull: I hadn't heard about it, actually. Have to pass on that one, too.
Atlanta: What role do you think endorsements from local people in government and past officials play in winning support for candidates?
Christopher C. Hull: I think there is some. My colleagues Steger, Adkins and Dowdle have done a lot more work on that than I, though -- Google their journal articles and you'll get an excellent answer to your question.
Boston: Ugh, I am so sick of people saying they want to vote for the candidate they would most want to have a beer with. That is the most ridiculous criteria for electing arguably the most powerful person in the world. They are not supposed to be just like us -- they are supposed to smarter than the average person, better educated than the average person, more driven than the average person, basically not like you and me. Vote for who you think can solve the country's problems the best, not who you think you would be friends with!
Christopher C. Hull: Fair point.
Washington: Do you suggest any good international sites or blogs for assessing the interest in and impact of our elections overseas?
Christopher C. Hull: No, I'm not sure I know of a good place to direct you. I love RealClearPolitics.com, but I'm not sure that's what you're asking about.
Christopher C. Hull: Well, thanks, all for your terrific questions. I have to sign off now. Please consider picking up my book -- again, "Grassroots Rules: How the Iowa Caucus Helps Elect American Presidents" from Stanford University Press.
Editor's Note: washingtonpost.com moderators retain editorial control over Discussions and choose the most relevant questions for guests and hosts; guests and hosts can decline to answer questions. washingtonpost.com is not responsible for any content posted by third parties.