Thursday, October 1, 2009

Pawlenty's Charge

Tim Pawlenty is kicking his presidential campaign into high gear today.

It's worth noting just how rare a Pawlenty nomination would be. Here are the results of open GOP nominations (leaving out, that is, nominations of sitting presidents):

1968 Nixon 2nd try (former nominee and VP)
1980 Reagan 3rd try
1988 Bush 2nd try (and VP)
1996 Dole 3rd try (and VP nominee)
2000 Bush 1st try
2008 McCain 2nd try

McCain, Dole, G. H.W. Bush, and Reagan had all been not only candidates, but runners-up, in the previous open nomination.

The big exception, of course, is George W. Bush, but just as obviously he was the son of a Republican president (and governor of a much bigger state). Bush defeated one previous VP (Quayle) and one previous candidate (Forbes) on his way to the nomination.

That's not to say that Pawlenty can't do it; it's a pattern, but there's been no study showing a causal connection between previous candidacy and nominations on the GOP side. That is, we don't know why it happens, including the possibility that it's a chance result, so I'd be very cautious about using the pattern to make predictions. And Democrats nominate first-time candidates frequently (Obama, Kerry, Clinton, Dukakis, Carter). It's tempting to think that there's something different about the two parties that allows Obama to defeat Clinton, Edwards, and Biden on the one hand, but that results in McCain prevailing over Huckabee and Romney on the other side -- but we don't actually know that.

Remember, it's only 27 months to the Iowa Caucuses. Ready yet?


  1. Interesting that you start your historical chart at 1968. Of course, Goldwater was an upstart in '64. No, Pawlenty is no Goldwater. I just mean to say that it's not impossible for untested national candidates to get the GOP nomination, just rare.

  2. Yes, and I should have been explicit about it. The reason I started in '68 is that before 1972 there's a significant data problem; it's a lot harder to know who really counts as a candidate. Then I figured that since Nixon was obviously a candidate in 1960, I might as well include him too, but I should have explained that in the original post.

    BTW, Goldwater was nominated in 1960, and received the only 10 votes Nixon didn't receive on the first ballot. He also received a handful of votes in several primaries, peaking at 1.3% in Nebraska (I found the results at So he wasn't a brand new candidate, exactly...that's why it's hard to do pre-1972 (I can't find a quick reference that says any more about Goldwater...White doesn't mention him as a candidate). Ike, of course, had never ran before...neither had Willkie. But Dewey had been a candidate for the nomination in 1940 before winning it in '44 and '48. So beginning in 1944, it's just Ike, W., and mostly Goldwater.

    So it's happened before (and I would never say that it's impossible), but it certainly is rare.

  3. This is a pattern that Ed Kilgore attempted to shoot down earlier in the summer in a post at FiveThirtyEight. What he called the "next in line myth," though, I found to be too narrowly defined and begged for an examination of the differences in patterns across the two parties.

    You pick up on some of that here, Jonathan (first-timers on the Democratic side vs. rehashes on the Republican side). Some of what is missing in what could be a be a better specified model is a focus on the elites within the parties and the rules that each party sets for presidential nominations. Though it may be changing, Republican elites were more homogeneous in the past (relative to the Democratic elite) In addition, the GOP's allowance for winner-take-all and loophole contests certainly (not to mention an increasingly frontloaded primary calendar) created an environment where known quantities with connections within the party were set up to do the best.

  4. Josh,

    Thanks for the comment. I'm interested in hearing more from you on particular, do you think it's voter side (elite Republicans preferring the obvious next-in-line candidate), or candidate side (candidates with experience as national candidates learning things that are useful for the next run) that's driving the results? Some of both? Or is it some other factor?

  5. Jonathan,
    I think it's a little bit of both. The Cohen, et al. book that John Sides mentioned in his Beck/Limbaugh post today has really guided a lot of my thinking on this. The parties and the elite donors within the parties have a significant say in who emerges in the invisible primary.

    But the candidates aren't powerless here. I don't know that they are learning anything out on the trail per se that helps them in subsequent runs, but they do see who does well within other campaigns and go after them later (campaign staff and donors). McCain surrounded himself with some of the Bush people that tore him down in South Carolina in 2000. Pawlenty is already looking to some of that Bush campaign talent. (Of course, Pawlenty's already got Andrew Sullivan seeing Bush II (or III, I guess.).)


Note: Only a member of this blog may post a comment.

Who links to my website?