Gerrymandering and Partisanship

Matthew Yglesias doesn’t share the common view that gerrymandering has caused congress to be more partisan. He notes a recent paper (pdf here) by Princeton’s Nolan McCarty, UCSD’s Keith Poole, and NYU’s Howard Rosenthal and quotes a bit and I quote a bit of that bit

Both pundits and scholars have blamed increasing levels of partisan conflict and polarization in Congress on the effects of partisan gerrymandering. We assess whether there is a strong causal relationship between congressional districting and polarization. We find very little evidence for such a link. First, we show that congressional polarization is primarily a function of the differences in how Democrats and Republicans represent the same districts rather than a function of which districts each party represents or the distribution of constituency preferences.

I comment on what I guess the paper says after the jump. I don’t see how it can be convincing (now maybe I would if I read it).

I’d guess that you McCarty, Poole, and Rosenthal are probably right, but I question your methodology. You and they seem to assume that representatives are free agents who decide based on their own policy preferences and those of their constituents. In fact, party leaders (in particular former majority leader DeLay) can impose party discipline. The leadership has actual power, and, in particular, has the power to decide who to appoint to the conference committees which actually write the final bills.

A possible explanation of polarization is that the Republican leadership used its great power to impose party discipline. Now how did that happen ? Part of it is that current and recent past leaders are hard line ideologues who are very sure that they are right and, especially, that Democrats are always wrong. Partly it is that they are partisan, eager to beat Democrats aside from the policy substance. Partly it is that they are unscrupulous.

So how did such extreme leaders (extreme especially in the means they are willing to use to impose discipline) get elected ? I’d say a likely explanation is that the median Republican representative moved right, became more ideological and partisan. That could be caused by gerrymandering.

I don’t see any way to test and reject the hypothesis that gerrymandering caused the median Republican to be further right which caused a hard line leadership to be elected which caused increased party discipline which appears as polarization. Of course the same thing could have happened to the Democrats. I happen not to have a sense of increased discipline based on the use of extreme means by the leadership, but I might be wrong.

Other representatives districts can affect Shays if they cause other representatives to vote for DeLay in the caucus.

The example of Cao seems to support this argument. He praised the stimulus and then voted against it. That sure looks like effective party discipline as opposed to ideological polarization with causes confined to New Orleans.
Or Connie Morella (R MD). Her votes are hard to reconcile not only with what she said as a representative from a bluish district but also with what she said before and anfter serving in congress. Looks like effective discipline not ideological polarization to me.

To try to sum up, I’d say that the causes of increased polarization prominently include the election of Gingrich, DeLay, Boehner and Cantor by the Republican caucus. That was caused by the orientation of the roughly median Republican representative and not on Chris Shays.

In the end I have no idea what role gerrymandering played. My argument is that you can’t determine the role by assuming that a congressman is affected only by his or her own district, and, I think, that assumption is necessary to your logic and the empirical analysis McCarty, Poole, and Rosenthal.