What If No One Wins the GOP Presidential Nomination?
Normally, I dread commenting on presidential nomination contests. But as much as I might like to return to the days of short presidential nomination processes (Franklin Roosevelt didn’t declare his intention to seek a third term until the summer of 1940), the reality is that the year-long nomination process is here to stay, and it is time to start writing on it.
But in truth, I’m actually hopeful about this year’s campaign, because I think it could be unlike anything we’ve seen in a very long time. I think the Republican Party really could wind up with a brokered convention – that is, a race where no candidate receives a majority of the delegates by the end of voting. In fact, it might well be the most likely outcome, if only because no particular outcome is particularly probable.
This race is intriguing not just because of one possible outcome. It is interesting because it is difficult even to formulate a workable theory of the race. Charlie Cook uses a brackets metaphor, while Jim Geraghty and Larry Sabato think of the race in terms of tiers, but all of these have problems. Instead, I see a race that is largely chaotic. It is one where an unusually large number of candidates have perfectly plausible paths, if not to the nomination, then at least to lengthy runs deep into the balloting process.
This is because 2016 really is the deepest GOP field in a very, very long time. In fact, it isn’t even close. To be clear, that doesn’t mean that eventual candidate is (or will be) the strongest Republican nominee ever. I think that’s unlikely, and in fact, that is crucial to my analysis. It just means that number eight is unusually strong. In 1996, eighth place in Iowa was businessman Morry Taylor. In 2008, it was Alan Keyes (who placed fourth in 2000). This year, eighth place will probably be a candidate we now see as a legitimate contender for the nomination.
Let’s look at Jonathan Bernstein’s list of potential candidates here, and assume the following candidates get in: Jeb Bush, Mitt Romney, Chris Christie, Rick Perry, John Kasich, Marco Rubio, Bobby Jindal, Scott Walker, Mike Huckabee, Rick Santorum, Rand Paul, Ted Cruz, Ben Carson, Carly Fiorina, John Bolton and Peter King. Some on that list won’t run, but some others probably will (Mike Pence or Rick Snyder being the most obvious contenders).
Let’s rate this field using a points system as follows: 5 points for a sitting veep, 4 for a sitting senator or governor, 3 for a representative, 2 for Cabinet officials, and 1 for “other.” We’ll (somewhat arbitrarily) add a point for “star power,” and deduct one for candidates who haven’t won a race in the past six years. We’ll do this for all the initial fields going back to 1980 (minor note: Harold Stassen receives a 1 even though he was a former governor. An election in 1938 doesn’t have much bearing in 1988).
The total for the prospective 2016 field is 56 points, by far the highest of any field. The next-closest field, from 2008, totals just 39 points. Moreover, the average candidate quality in 2016 is the highest of the bunch: 3.5 points, compared with 3.1 points for 2012 or 3.3 for 2008. Even this doesn’t tell the whole story though, as the 2008 slate is filled with candidates who were much weaker than their ratings suggested (Jim Gilmore, Sam Brownback, Tommy Thompson). Almost all of the candidates on the 2016 list would have been top-flight contenders against the 2012 field, yet many of them will struggle to finish in the top five in a single primary or caucus.
Now to be clear, it is likely that some of these candidates will drop out as we approach actual voting for the usual reasons: they fail to gain traction in the polls, fail to raise money, or are excluded from debates. At the same time, I think that this “early winnowing” effect will be more muted than is usually the case. Most of the candidates on my list tend to draw support from different wings of the party, have different bases of fundraising, and will register at least some support in Iowa. Someone might catch fire, but I think the lack of an overwhelmingly strong candidate means that it is just as likely that the polling remains very tight, with candidates struggling to make it out of the low teens. This keeps even marginal candidates in striking distance and will decrease the incentive to drop out. Our hypothetical field of 16 might be 10 by caucus day, but it will be a very serious group of 10.
The traditional way to analyze the Republican primary is to walk through the early states, gaming out various paths to the nomination. So, we would start with Iowa, which traditionally likes religious conservatives and fellow Midwesterners. This might argue for Scott Walker, who performed well in the state over the weekend, Ted Cruz, who could combine religious conservatives with Tea Partiers, or perhaps for a repeat victory for Huckabee or Santorum.
Next is New Hampshire, where we could see a Chris Christie, Mitt Romney, or Jeb Bush do well. South Carolina traditionally follows (although for now, New York and Utah precede it) and it has long been the establishment firewall. But lately it has been more populist: Huckabee very nearly defeated McCain (in a race whose map eerily paralleled a 1940 anti-prohibition referendum) on the basis of a strong showing in the upcountry, while in 2012 Newt Gingrich beat Mitt Romney in the state by a 12-point margin. So we might label this fertile territory for an insurgent populist, perhaps Ted Cruz or Rick Perry.
That leaves a flurry of caucus states – Nevada, Colorado and Minnesota – to consider before we get into the run up to Super Tuesday. Santorum made a splash by winning some of these in 2012 (though Romney did manage to win Nevada). But look at the second-place finisher in Minnesota and Maine (which held early caucuses in 2012): Ron Paul. Caucuses tend to reward candidates with devoted followings. If Rand Paul inherits his father’s following and builds upon it somewhat as a more reasonable, electable option, one can see him performing well here.
So here you have a perfectly plausible scenario where we exit the early primary phase of the contest with four winners, each of whom is a legitimate presidential contender. What’s more, it’s not entirely clear how they knock each other out. Scott Walker, Jeb Bush, Ted Cruz, and Rand Paul all represent different wings of the party, would draw from different fundraising bases, and would have different demographic appeals. Just as important: None is an obvious choice, but at the same time, unlike 2012, all will have a group of supporters that really likes them; it won’t just be an “anti-Bush” vote trying to coalesce. You can mix up the various winners (Rubio, Christie, Perry, Paul), but the same analysis holds.
Plus, we have states like New York, Utah and North Carolina that have moved up their primaries. We don’t have a good feel for these states, but you could take any one of the above scenarios, add Chris Christie in New York, Mitt Romney in Utah and any number of candidates in North Carolina. Moreover, a strong second-place finisher could decide that he is the next Bill Clinton (who famously won only one of the first 11 primaries in 1992), and try to keep going.
At that point, it really is anybody’s game. No one really has an incentive to drop out, as the RNC’s compressed schedule means the finish line is in sight by the time Super Tuesday rolls around, and all of these candidates can probably win a race here and there to keep the old ball rolling. Money might get tight, but the threshold for winning these contests remains low. It also becomes very difficult for any one candidate to amass a majority of the delegates very, very quickly.
Complicating matters even further, our analyses haven’t fully accounted for the rise of SuperPACs. I suspect 2012 was but a preview of their potential impact. Rick Santorum nearly threw the entire race into chaos in 2012 with a camper and the backing of Foster Friess. Sheldon Adelson helped Newt Gingrich stay in the race through May. Without SuperPACs, they likely would have been out in March, at the latest. What happens if Friess, Adelson, Karl Rove and the Kochs all back different candidates, while a candidate like Paul survives off of grassroots support? That race could go on for a very long time.
But, in fact, the race is even less predictable than the above analysis suggests. To see what I mean, let’s revisit our list of candidates above, putting in a sort of bare minimum for each candidate in Iowa, without any regard for the total vote share. I did this, and I was not generous. Top-flight candidates rarely drop much below 10 percent here, and candidates we today regard as also-also-rans routinely put up strong showings. So when I give two-term governors who have routinely been mentioned as nominee material like Bobby Jindal or Chris Christie 5 percent, or give 11 percent to a candidate like Mike Huckabee, who won 40 percent of the vote the last go-round, I’m being pretty stingy.
The total I came up with was 125 percent. There are two implications to this. First, a lot of objectively strong candidates are going to have to do quite a bit worse than we currently think is possible next January, but we have no real way of knowing just who those candidates are. To be sure, the field will narrow some by Election Day, but I’m already giving the most likely dropouts very small vote shares.
Second, and most importantly, with a deep field such as this that splits the Republican coalition in so many different ways, you really might be able to win Iowa with 12 percent of the vote or so. Alan Keyes surpassed that vote share in 2000, Gary Bauer came within striking distance of it that same year, and Pat Robertson doubled it in 1988.
Given this, almost anyone really can win Iowa this time. Moreover, if we really do have a low-teen threshold for victory in these early races, the types of unpredictable “quantum effects” that political scientists routinely dismiss as irrelevant, like newspaper and gubernatorial endorsements, suddenly become important. Could Chris Christie have a solid debate and shoot from 5 percent to 15 percent in the polls right before the election? I’m basically asking if he can win over two-thirds of Romney voters, so the answer to me is obviously “yes.” Might Terry Brandstad decide it is Kasich's time, and rocket the governor into a surprise third-place finish? If it only takes 10 percent of the vote to do so, why not?
So what happens if, instead of Walker, Bush, Cruz, and Paul, our early winners are, say, Ben Carson, Mitt Romney, Rick Perry, and Paul, with Walker, Bush and Cruz coming in a close second in these states? The result would be seven solid candidates receiving substantial numbers of delegates early on, without an obvious pick for the party. It would quickly become self-perpetuating: The longer candidates continue to rack up delegates, and the longer that the size of the field prevents someone from racking up a huge numbers of delegates, and the longer the field will stay large.
The most credible response to all of this is, in my view, “Haven’t we predicted this before?” This was basically Steve Kornacki’s rejoinder to me in 2012, when I was discussing such a scenario for that year. Our back-and-forth is worth revisiting if you agree with me so far, as Kornacki’s recitation of history is impeccable (as is his wont).
But my rejoinder is basically the same as last time: Past performance is no guarantee of future results. This is especially true since some things really have changed: SuperPAC funding, Internet fundraising (weakening parties), and the size and strength of this field.
Most importantly, we should bear in mind just how close Republicans actually came to a brokered convention in 2012. Had 5,300 Ohio voters changed their mind, and/or 16,200 Michigan voters cast their ballots differently, Romney would have been severely damaged, and that race probably would have gone to the convention. Party elites might even have demanded it.
For that matter, consider 2008 on the Democratic side. John Edwards was a very serious candidate, coming off of a credible run as vice president in 2004. What if he had decided to gut it out for one more week, through Super Tuesday? Let’s say he won only 90 delegates – just 5 percent of the 1,700 delegates awarded that day. If he pulled evenly from Obama and Clinton, this would have been enough eventually to deny Obama an outright majority of the pledged delegates.
Of course, the super delegates would probably have still saved Obama (Edwards would have had to have won about 250 delegates on Super Tuesday to prevent that), but super delegates don’t fit into the Republican calculus to any great degree. More to the point, it only took two equally matched candidates and a tepid effort from a third candidate for the 2008 Democratic race to come dangerously close to a convention. If just four or five evenly matched Republicans making it to Super Tuesday, it’s hard to see how a similar result would be avoided.
This isn’t to say that things necessarily play out this way. A candidate could catch fire and suddenly bring stability to the races, as happened with the Democrats in 2004. A large number of candidates could decide not to run.
Rather, the point is that it really is unknowable at this point what will happen. For now, the race is chaotic and utterly unpredictable. Which makes it fun.