Why the Republicans had to nominate a flip-flopper

(Cross-posted from the Mischiefs of Faction)

With the benefit of hindsight, it's hard to see how Santorum, Gingrich, or anyone else ever had a chance at the Republican nomination this year. But let's not forget -- people were absolutely freaking out about those possibilities just a few months ago. Romney was the troubled front-runner who had a 30% ceiling and was just barely defeating candidates he was outspending 10 to 1. He was also the candidate who allegedly could not be nominated because of his dalliances with moderation or because of his recent flip-flops.

David Karol has an interesting post at the Monkey Cage in which he argues that Romney's "very inconsistency was a necessary, if not sufficient, condition for his success in capturing his party’s presidential nomination this year." But I think it goes further than Karol suggests. It's not just that Romney had to switch his positions to be a credible potential nominee. I would argue that any Republican presidential nominee today would have to be a serious flip-flopper.

One reason Romney's nomination was relatively predictable was that he was running against the sort of people who are simply never nominated for the presidency by the major parties. Gingrich had a notably unsuccessful and short term as Speaker and hadn't held public office in over a decade. Santorum's initial election to the Senate was somewhat of a fluke and his 2006 drubbing in a swing state did not bode well for him. Bachmann was a member of the House. Cain was an eccentric businessman. Parties almost invariably nominate current or recent senators or governors, and of the prospective field, only Pawlenty, Daniels, Christie, Palin, Perry, and Romney fit the bill. Three of those (Daniels, Christie, and Palin) seemed hesitant to fully jump into the contest, and among the three that jumped in enthusiastically, two of them (Pawlenty and Perry) had serious campaigning problems. Once the two of them had functionally dropped, it was hard to see anyone but Romney getting it, unless it was going to be a Really Unusual Year. And of course you never know whether or not you're in a Really Unusual Year until it's over, but by definition, they're really unusual, so the safe bet is that things are happening as usual.

But here's the key point about that: No one taking the stances Romney needed to take to win this year could have had the sort of résumé needed to be a typical major party nominee. The Republican Party has been moving to the right very quickly in recent years. Almost no one taking the stances that Romney is taking now could have been elected as a senator or a governor from most states just a few years ago. So, if you were consistently conservative (like, say, Bachmann or Santorum), you were either doomed to service in the House or to being kicked out of the Senate. If you had a presidential résumé, conversely, it was probably because your views were pretty moderate a few years ago. Arguably, the only person who can get nominated in the current Republican Party is someone who has pivoted to the right rapidly in the past decade. Rapid polarization makes flip-flopping a necessity.

A noble spirit embiggens the smallest party faction

Via Sullivan, here's Mike Tomasky's take on the danger's of a Sarah Palin presidential nomination:

The Republican primary will likely be set up in such a way that she’ll be able to pull the field, and the party, hard to the right, and while that will be fine for Barack Obama come Election Day, it will help cripple any attempt to do anything constructive afterward.

This differs from the situation today how exactly?

Nebraska: The Next Frontier

I've previously mentioned some research by Boris Shor, who managed to collect nearly-complete roll call datasets from every state legislature spanning the past two decades.  This dataset really stands to  revolutionize the study of state politics once it's released.  For a taste of what's in there, check out this working paper by Shor and Nolan McCarty (PDF).  It looks at a whole range of issues, including representation and polarization, from the vantage point of this new dataset.  It's extremely rich.

Shor and McCarty make a few interesting comments about Nebraska, which has the only nonpartisan state legislature in the country. What they find runs a bit counter to what Wright and Schaffner found a few years ago in their APSR paper (gated) comparing Nebraska to Kansas:

When we pool the state’s APRE statistic for the first dimension, we find that it is relatively low at 27%. However, four other states (Arkansas, Louisiana, West Virginia and Wyoming) score lower on this measure of fit. Similar results hold for a two-dimensional model.

We can also use a party-free measure of polarization – the average ideological distance between members – to compare Nebraska to other states. Just like many other states, Nebraska is polarized, and becoming increasingly more so. On average, Nebraska’s Senate is more polarized than 17 other chambers. In fact, it is actually polarizing faster than many other states. By the party-free measure, it polarized faster than 75 other chambers over 1996-2008.

How is it that a nonpartisan legislature is more polarized than 17 partisan ones?  What makes a nonpartisan legislature polarize rapidly?  These strike me as really interesting questions that party scholars should be trying to answer.  I'm on it.

Polarized voters in California

Over at Larry Sabato's Crystal Ball, Alan Abramowitz has a nice post about California's recent adoption of the top-two primary, suggesting that it really won't do much to affect legislative polarization.  Why?  Because the state's voters are already incredibly polarized -- more so than the American people in general.  Note this helpful chart:

As the chart suggests, California Democrats are much further left than national Democrats, although California Republicans look pretty much like national Republicans.  Also, Abramowitz notes, liberals and conservatives are geographically concentrated in a few key areas.  If San Francisco liberals are so far left, and there are pretty much no conservatives in the area, then legislative districts in that area will keep electing far-left liberals.

It's interesting to note the slight discontinuity between Abramowitz's voter graph above and Boris Shor's graph of elected officials.  In the latter, California's Republican legislators constitute the most conservative legislative party in the country.  And in general, it's worth remembering that elected officials tend to be more ideologically extreme than their districts.  So the graph above only tells a part of the story about legislative polarization.

(h/t Brendan Nyhan)

Simgov - How are the students doing?

Loyal readers may recall that I've been running a simulation of American government class this quarter.  Basically, most of the students are portraying actual current members of Congress.  They author bills and try to push them through committees and get them passed by the full House.  With the help of clickers, I've been recording their roll call votes.  There are only 30 votes so far, but I've used them to generate ideal points using NOMINATE.  In theory, students can compare their own ideal point with that of the member they're portraying to see how they're doing.  The scatterplot below shows actual member ideal points on the horizontal axis and student ideal points on the vertical axis:

At one point, I thought it might be good to use this method as a grading tool -- maybe deducting points for the ideological distance between students and the members they're portraying.  However, the standard errors generated by only 30 roll call votes are huge, so I don't think it would be proper to base grades off these scores.  That said, the ideal points are suggestive.  For one thing, the students are doing a pretty good job; the scores correlate at .965.  But, of course, that's what you get when all the data are in the extremes, which points to a second inference: my class appears to be more polarized than the actual Congress.  

I'm not sure why they're so partisan.  I mean, the TAs and I try to instruct them in the importance of partisanship and issue warnings when they vote against their party (or district) too much.  But I'm not sure how much of this is us and how much of it is the dynamics of legislative life.  I give the parties time to caucus before floor sessions, and they actually use those times to develop strategies for screwing the other party. They're really quite crafty this way.  I sometimes worry I'm creating a small-scale Stanford prison experiment -- the students really do internalize their roles well -- except that they're still keeping it civil with each other in committee and on the floor and, as far as I can tell, they don't carry their partisan roles outside the classroom.  They're just voting against each other.

Primaries and Partisanship

John Sides has a few posts recently (here and here) in which he cites research by Eric McGhee at the Public Policy Institute of California on the effect of primary systems on the partisanship of elected officials. (Disclosure: Eric and I are working on a conference paper together on this topic.)  His findings suggest that the effect of moving from a closed to an open primary is surprisingly modest.  He further notes that during California's brief experiment with the blanket primary, politicians were no less polarized and the budget was still passed late.

Here's another way to look at this issue.  In the graph below, I've plotted Jerry Wright's data on legislative partisanship in each of the state legislatures during the 1999-2000 session along the vertical axis.  The horizontal axis is the type of primary the state was using at the time, going from most to least restrictive.

Now, notably, the states at the far right (Louisiana with its jungle or top-two primary and Nebraska with its nonpartisan elections) are among the least polarized legislatures.  However, the regression line is nonetheless flat.  We'd expect polarization to decrease as primaries become more open to independents -- they don't.  Indeed, many of the states employing the blanket primary at the time prove relatively partisan.  And Wisconsin and Ohio, which then employed open primaries, have among the most polarized legislatures in the country.  Of course, this is a pretty simple bivariate look at this problem, but one would expect to see more of an effect than this.

So what will happen if California adopts Proposition 14, the top-two primary, this spring?  Well, I suppose that depends on just how representative we think Louisiana is.  Does the Cajun State have a relatively moderate legislature because of its primary system, because of its unusual political culture, or because of other odd institutional rules in place there?  Given that it's hard to separate those out, and given that other evidence about primaries and partisanship suggests little relationship between the two, I doubt the initiative would have anything close to the impact its backers suggest.

I'm wondering if anyone has recent data on Washington's state legislature, the members of whom have been elected through a top-two primary since 2006.  Since Washington has been among the more polarized legislatures in recent years, it will be interesting to see if the chamber has been de-polarizing.