Tuesday, January 19, 2010

In the Middle

From the Jonathan Caucus: Chait today has an interesting post claiming that a good part of the reason for what we see in health care reform is that Republicans just haven't bothered to learn the policy basics about the issue.  While I think he may be correct (quick -- who is the conservative counterpart on health care to Jonathan Cohn and Ezra Klein?), there's something else in his response to Steve Bell that I think bears a bit more of a comment.  Bell says:
I believe that such a deal — outlined in very rough form here — would probably get 70-80 votes in the Senate.  The 15 “theological” senators from the Progressive Caucus and the 15 on the obdurate right of the Republican caucus would scream.  That would confirm that such a deal is in the interests of most senators and most citizens.
Now, Chait is probably correct that the plan Bell outlined wouldn't work on policy grounds, and many Democrats believe (with some justification) that for partisan reasons at least 35, and maybe all 41, of the Republicans in the Senate would rather see health care reform die than enact a bill that they actually like.  But there's more to it than that.  Bell (and perhaps Chait, who passes over it without comment) seems to believe that the distribution of votes in the Senate matches the distribution in the population -- roughly a bell curve, with the meaty part in the middle.  If that's the case, it's easy to lop off the ends and find, as Bell suggests, seventy to eighty votes.

But that's not the current Congress is actually like.  Look at this graph, which I've linked to many times: the thin part of the distribution isn't at the ends; it's in the middle.  That's especially true on the Republican side.  There are four outliers to the right (Coburn, Bunning, DeMint, Inhofe), but after that, there's not much difference between the 5th most conservative Republican and, say, the 26th (Wicker).  Then, once you get close to the middle, the Republicans are strung out again, with Snowe.  The same thing is basically true on the Democratic side; there are a handful of moderates, then a larger group of moderate liberals, and then the bulk of the caucus, who are mainstream liberals.  There are no "15 'theological' senators from the Progressive Caucus."  In fact, the 15th most liberal Senator (Schumer) appears to be statistically indistinguishable from the most liberal and also from the 33rd most liberal Senator.  And, really, that sounds about right to me; with the possible exception of Sanders (whose voting record turns out to be to the right of Schumer, for what it's worth), I don't think there are a handful of Senators I'd think of as the "most" liberal.

The other thing to look at is that the distribution of moderates isn't equal.  Two Dems (Nelson and Bayh) are just to the right of center.  But including those two, the 44th least liberal Democrat is about as far from center as is the 16th least conservative Republican.  One way to look at this is that a compromise between, say, the ideal points of the 30th least conservative Republican and the 30th least liberal Democrat is going to be fall far into conservative territory.  There's no way that Democrats would ever go for that deal.

And, it seems, no way that Republicans would accept a compromise that falls near the ideological midpoint (which, as Chait points out, is a fair appraisal of the Senate-passed bill). 


  1. Although this is interesting, I humbly suggest that there is something flawed with a ranking that makes Feingold the third most conservative Democrat and puts Bernie Sanders to the right of 20 Democrats.

  2. Feingold's voting record has been very curious this last year; every measure I've seen has placed him well into the right of the caucus. It's because he's been a very consistent opponent of anything that could even remotely be considered 'pork'.

  3. Pork, but he also has other occasional oddball votes (IIRC, he was the only liberal who voted to confirm Ashcroft).

    It's true that the particular collection of votes doesn't always give the true measure of the Senator -- just to pick one thing, if there had been a vote on single-payer, that would have moved Sanders to the left (since only a handful of Dems would have voted with him). Certainly, I wouldn't just automatically take the votes and declare that Feingold is a very conservative Dem...but what needs to be done with data like these is to explain them, not ignore them.


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

Who links to my website?