Thursday, September 20, 2012

Doug Waddell's Question: Bipartisanship in the US Congress

On the AaPS Facebook page this afternoon my friend Professor Doug Waddell of the glorious land of East Tennessee asked me a question:
...which senators would you say have been the most "bipartisan" over the last two years?
My first response would be to point out that, honestly, bipartisanship has never been particularly common, at least since the emergence of the modern parties in the post-War period.  Don't take me at my word: I'd suggest a couple articles if you're interested - including Keith T. Poole's (2007) "Changing Minds? Not in Congress!" in Public Choice.  Poole concludes his study by saying:

An ideology or belief system is ".. .a configuration of ideas and attitudes in which the elements are bound together by some form of constraint". It tells us what is "the 'good' in politics and social intercourse"  . It tells us who gets what, and who should rule. Ideological stability among an elite group of national politicians should not be surprising. How much to tax, how much to spend, how much to redistribute, how much to help people in need, how much to help people to help themselves, and so on, all these questions are fundamental to modern politics. These are questions of what is good. "Beware a wolf in sheep's clothing" is an old saying that points to a fundamental truth - people are suspicious of those who change their minds, especially about something as fundamental as what is good in politics. 
There may be changing minds, but they are not in Congress.
In other words, you may get to Congress by being ideologically loose, but it is unlikely you'll remain there very long because, regardless of what people claim, rarely do they want someone representing them who is, apparently, unpredictable. Furthermore, it seems like bipartisanship is becoming rarer (it isn't just in your head!) - check out the statistics on political polarization at Voteview.com (a summary of Howard Rosenthal and Keith Poole's statistical analyses of political polarization trends - indeed, endangered. Heck, Voteview.com asserts, "Polarization in the House and Senate is now at the highest level since the end of Reconstruction." [italics and blood-red coloration are theirs!]

Don't worry too much though - the differences in ideology often don't make much of a difference - as Thomas Bräuninger (2005), also in Public Opinion, points out in his article "A Partisan Model of Government Expenditure":
Using data on 19 OECD countries from 1971 to 1999, the paper finds support for general partisan hypothesis. The results suggest that the actual spending preferences of parties matter whereas they do not indicate that parties of the left consistently differ from parties of the right in their spending behavior.
In other words, there is a great deal of orthodoxy in the legislatures which are members of the OECD (the Organization for Economic Cooperation and Development) but that orthodoxy doesn't readily translate into orthopraxy - that is to say thought and words may align with platforms, but that doesn't necessarily result in budgets that reflect those platforms.

My second response was, good question - indeed, a very good question, in fact, for at least three reasons.

Reason number one - the constituents of elected officials have a manifest right to know whether or not their elected officials are representing their interests. Which isn't to say that elected officials need always vote in a manner that reflects the general will of their constituency - some folk hold more to the trustee logic of elected representatives, after all (meaning that they believe that their leaders, who of necessity know more than their electors about the issues to be decided as a product of time and inclination, if not native skill, are better prepared to make policy decisions whether or not they align with the preferences of their electors) -  but certainly a real awareness of whether or not their delegates adhere to their party platforms or adhere to their own concerns is important in deciding whether to re-elect (or in some instances, recall) the leader in question.

Reason number two - if we want to be able to predict the outcomes of contentious debates both in committee and on the floor, well, heck, we need to know the proportion of party-liners to pragmatists and, on a related note,

Reason number three - if we want to know which congressional leaders acquire extra-institutional power via serving as negotiators and go-betweens for other actors, well, we need to know which are best at practicing negotiation as a political strategy.

Good enough.  Well then, we find ourselves in another pickle - how do we define and measure bipartisanship.  Hmm.

Well, first thing first - bipartisanship, by its definition, can only exist in a political system in which there are two dominate parties - otherwise we're talking about a more generic term (which encompasses bipartisanship, of course), nonpartisanship.

We could conceptualize bipartisanship in a couple of ways.  First, we could measure each congressman's tendency to vote against one's party platform.  This is problematic in the American system since, frankly, there is very little party discipline within either of the two  major parties and, equally, such discipline as there is often dedicated to factions within the respective parties (e.g. the Tea Party movement within the Republicans).  Secondly, we could measure each congressman's tendency to cosponsor legislation with members of their opposite party - this is a neat idea because it demonstrates a willingness to be very much identified as bipartisan, but again it has flaws - notably that there  is no reason that identification as a bipartisan should necessarily correlate with actual bipartisan behavior overall and, even more importantly, this tendency seems to be largely cancelled out by "bandwagon effect" where influential members of congress find their legislation "jumped on" by other members of their party seeking to be ideologically identified with that legislation (if you're interested in this, check out Rick K. Wilson and Cheryl D. Young's 1997 article  "Cosponsorship in the US Congress" in Legislative Studies Quarterly).

Geez.  Difficult.  Well, I found a measure for this - specifically the Bipartisan Policy Center recently published a document entitled, "Fifty Most Moderate Members of the 112th Congress and Their 2012 Prospects" - it is a simple document that captures some complicated things.  Notably, it ranks congressmen and senators according to their willingness to cooperate in formal votes with members of the other party using the rather complicated DW-NOMINATE scores (with parametrically bootstrapped standard errors - yeah - math) method developed by Rosenthal and Poole (who I mentioned above and whose work is accessible at Voteview.com).  It isn't a perfect measure, but it is a good measure with the added benefit of estimating the odds that these ladies and gentlemen will stay in office come the next election.  You can look at the document yourself for the details, but I'd like to note a couple of generalities from eye-balling the report:

(1) None of the most moderate Democrats are women - only three of the most moderate 25 Republicans are women.
(2) Moderate Democrats seem most likely to come from the Upper South; moderate Republicans tend to come from the upper Mid-Atlantic and Northeast.
(3) Seven of the 25 most moderate Democrats aren't seeking reelection in 2012; three of their Republican peers are not.

I'm not sure this perfectly answers Doug's question, but I hope it is a decent start.