The old left-right ideological model for how Congress votes has lost explanatory power. Something different is afoot in US politics

Over the past few weeks, I’ve written a lot about ideology and polarization. Most of that has focused on the left-right differences. But there’s another shaping theme in politics that is also present: establishment v insurgent.

Presidential campaigns, for example, usually come down to those who are backed by the establishment and those who are not. The candidates backed by the establishment usually win, and the outsiders almost always lose. It’s the main thesis of the great book The Party Decides.

That’s why many political scientists thought that Mitt Romney winning the bulk of congressional endorsements for the Republican nomination in 2012 meant that he would almost certainly capture the nomination. His establishment support came from both left and right of the party, and was actually slightly more conservative than either Gingrich’s or Santorum’s, despite Romney donor profiles indicating that his public was moderate.

Twelve years prior, George W Bush triumphed over the moderate John McCain on the back of establishment support. Likewise, Hillary Clinton saw her 2008 hopes fade because of her inability to lock up support from establishment congressional leaders Harry Reid and Nancy Pelosi.

No one can doubt that this establishment v insurgent dynamic has been present in Congress for a while, but “DW-nominate” scores indicate that this dichotomy is becoming increasingly important. These scores seek to rank congressional members’ ideology in two dimensions, based on their rollcall voting record. In my previous analyses of polarization, I’ve used the first dimension, which usually does a good job at placing members on the liberal-to-conservative spectrum. The second dimension has rarely been used in recent years because it hasn’t seemed to stand for anything.

In the past Congress, however, the second dimension has begun to have more explanatory power – among Republicans, especially. That is, something beyond just being conservative or liberal is beginning to predict voting patterns of congressmen and women. The vote-view folks and I think that this dimension is along an establishment v anti-establishment axis, though one might also argue that it is geographically based – since many insurgent Tea Party members are from the American south.

Consider the debt ceiling debate of 2011. People on both the liberal and conservative side of the aisle supported the bill. Oddly, House Democrats and Senate Republicans were the two groups most likely to oppose the bill.

If the voting had been strictly along partisan lines, we’d expect the best fit line to be straight up and down. That is, the vote could easily be determined as being to the left or to the right. Instead, we have more of a diagonal line that goes from upper left to lower right in the House and upper right to lower left in the Senate. What that means is that there are liberal and conservative elements behind the voting, but there is also something else happening. You might call it a rather strange vote in which the Tea Party caucus, progressive caucus, and black caucus voted all together – though, arguably, it could be explained as establishment House Democrats and establishment House Republicans voting together. In other words, how many times can we expect John Boehner, Eric Cantor, Steny Hoyer, Kevin McCarthy, Nancy Pelosi, and Debbie Wasserman-Schultz all to vote the same way on a relatively divided debate?

The coalition was slightly different with the recent fiscal cliff vote, yet it displayed a similar pattern. Again, there was a left-to-right element, with more Democrats voting in favor this time. But there was also a “second dimension” element.

The lines are, again, diagonal. The differences among Democrats largely disappeared, but they were as present as ever among Republicans. The “establishment” Republicans largely voted in lockstep with one another, while the anti-establishment folks also banded together, voting the other way. The fit is certainly not perfect, with establishment figures like Cantor and McCarthy voting no – thought that actually matches well with the press coverage afterward declaring that Cantor’s vote indicated a break with the establishment leadership, rather than the fact that he’d newly discovered a conservative soul.

This divide between establishment v anti-establishment was present throughout the 112th Congress. Whether it was the 2012 omnibus bill, or the highway and student loan funding bill of 2012, outsiders such as blue-staters Michele Bachmann and Marco Rubio voted alongside southerner Rand Paul and westerner Mike Lee. Others with similar left-to-right rankings, but who were closer to the establishment, such as Senator John Barasso, Representative Cantor, Representative Billy Long of Missouri, and, yes, even very conservative Mitt Romney-backed Representative Steve King of Iowa often voted the opposite way.

It’s necessary to note that while the insurgents seem to be rising, it was the establishment that won in all the situations. My guess is that they will continue to win, even if they need to adjust. That’s why I’m skeptical about whether we’re really likely to see a “Republican civil war” in the coming years, or even the 2016 presidential nomination season.

In practise, the establishment tends to line up behind the eventual winner of the nomination before a war breaks out. That’s why even the hard-fought 2008 Democratic and 2012 Republican nomination winners became quite clear by the end of February in the nomination seasons. Further, if Republicans do well in 2014, which they should given the midterm landscape, then this should placate those currently calling for heads.

The ability of the party establishment to hold onto power might explain why Paul Ryan voted for the fiscal cliff package. Ryan is pretty far to the right in the left-to-right ideological rankings. He’s even been seen as fairly anti-establishment over the long term. Yet, he voted for each and every one of the “establishment” positions on the four key financial bills discussed here. Ryan may be looking to capitalize on the establishment credentials he built up during his run as vice-presidential nominee. We’ll have to see if his future voting record also supports the idea that he’s shifted toward the establishment.

In the meantime, it looks likely that this establishment v insurgent divide will continue through to the next Congress. The votes for and against John Boehner for house speaker did not split according to the liberal-to-conservative spectrum for Republicans, but rather among the second dimension: between establishment v anti-establishment forces. In short, this should looks set to be yet another unproductive and unwieldy Congress.