(Maximize the video and set the quality to 720p HD for best viewing. The video is also available for download from here.)
Each senator is marked by a two-letter state code and is color coded by party membership (Democrat, Republican, Independent). The top grey bar is the distance between the mean Republican and the mean Democrat (the standard measure of partisan polarization). The size of each party’s coalition is displayed at the end of the bars. The lower grey bar is the gridlock interval, which is measured as the distance between the filibuster pivots (the 34th and 66th most conservative senators before the Senate rule change in 1975 and the 41st and 60th thereafter). The ‘M’ imposed on top of the bar marks the position of the median senator.
The ideological estimates are constructed by scaling senate voting records with dynamic optimal classification. Keith Poole’s paper on optimal classification can be found here, and the paper which explains how to extend the method to smooth legislator ideal points over time can be found here.
There has been much talk as of late about what has caused politics in this country to become so polarized. The political science literature usually points to changes in procedural rules, regional sorting, a rise in economic inequality, extremism of party activists, and to a lesser extent, redistricting to explain the rise in polarization. However, the word on the street seems to be that Congress polarized because members from opposing parties no longer invite each other over to dinner. At least that is what Evan Bayh—the soon to be former senator from Indiana, who is also a son of a senator with the uncommon experience of taking up their fathers’ profession—had to say about it in an op-ed published in the NYT on Sunday. We political scientists tend to chalk up this account as something akin to pop-psychology, but given the level of genuine animosity apparent in Congress it is hard to see how it has not taken a toll, even if we lack sound methods to measure it.
One way to get a rough gauge of how the decline in bipartisan manners may have contributed to polarization is to look at how periods marked by bitter partisan disputes affected voting behavior of the parties. In order to get a measure of this, I use estimates from a roll call scaling method described here that uses kernel methods to smooth legislator trends over time. This makes it possible to track how polarization changes from one day to the next, in contrast to DW-NOMINATE, the only widely used estimation technique that allows for inter-temporal movement of ideal points, which only recovers a single estimate for each two-year Congress.
The figure above shows that day-by-day polarization trend closely tracks the corresponding DW-NOMINATE trend. Both trends tell a similar story in the long run: polarization reaches a low-point around the 90th Senate (1967-1968) and steadily increases in the following decades. At the same time, the day-by-day trend reveals that political polarization varies considerably between elections.
The next figure magnifies the polarization trend for the period between the 103rd and 109th Senates so that it is possible to view how Senate polarization responds to political events. There are three notable events that affect the polarization trend. The first event occurs during the 104th Congress and corresponds perfectly with the federal government shutdowns that resulted after the Republican controlled Congress failed to pass a budget bill. The second event is Senator Jeffords’ decision to switch parties during the 107th Senate, which was followed by a sharp decline in polarization. The third notable event is the rapid rise in polarization during the run up to and through the early months of the Iraq War. In December 2003, this trend abruptly ceases and then falls, corresponding with the capture of Saddam Hussein.
Some of the events seem to have lasting influence. This is best exemplified by the jump in polarization corresponding with the 1995 government shutdown. Rather than observing a spike in polarization that quickly returns to its previous level, the trend remains at its heightened level, perhaps, as Senator Bayh has suggested, as a result of increased animosity that weakened bipartisan goodwill.
Interestingly, despite my initial expectation that Clinton impeachment trial had increased polarization, in his op-ed Senator Bayh points to the impeachment trial as the one of the two instances (the other being September 11th) where all senators gathered for something other than purely ceremonial occasions and as a highlight of bipartisanship during his twelve years in the Senate:
All of us gathered in the Old Senate Chamber. For several hours we debated how to proceed. Finally, Ted Kennedy and Phil Gramm, ideological opposites, were given the task of forging a compromise. They did, and it was unanimously ratified.
This might offer some context as to why the polarization trend is in decline during that period. It also paints the somewhat paradoxical picture where a Senator best known for his consistently liberal voting record was also one of the most potent de-polarizing forces in the Senate. Although there is no way to know, something tells me that had Ted Kennedy not passed away last summer there would be much less talk at the moment about reforming the filibuster.