On Electoral Reform in the U.S.: Instant Runoff Yay, PR Nay

There’s an interesting discussion going on in comments here about electoral reform measures. I certainly support instant runoff voting or a similar procedure that would prevent irrational results in multy-party races. This isn’t usually a major problem in American politics because of Duverger’s law (which is actually merely a strong tendency), but it’s a potential problem it’s worth correcting for. Remember that there’s no reason to believe that instant runoff would produce better results from a normative standpoint — a popular vote with instant runoff almost certainly gives us President Al Gore, but it also almost certainly gives us President Stephen Douglas — but there’s no reason to have an electoral system open to producing such irrational results.

Of course, instant runoff is unlikely to lead to robust third parties, since you still need to have a majority coalition to win. Which means some people favor PR precisely because it will lead to a multi-party system rather than a two-party system. On this, I’m much more skeptical for a few reasons, some theoretical and some practical:

• Imagine the Democratic Party is broken into a Progressive Party and a Connecticut for Lieberman Party.  As someone who thinks the idea of voters as atomized consumers is puerile narcissism, I don’t see any added value in voters pretending they aren’t sullying their precious hands with their less savory necessary coalition partners.  On the other hand, I think the downside is clear — it would add yet another de facto veto point to a system that already has far too many. Putting together party coalitions in advance attenuates one of the many barriers to legislative action.  Consider, for example the ACA.  PR wouldn’t change the fact that the median votes in a Senate titled towards conservative interests would have an effective veto over the legislation.  But on the other hand, Bayh, Nelson, Baucus, Landrieu et al were at least part of the Democratic coalition with some small stake in not having the bill crash entirely, which is one reason they voted for a bill almost certainly to the left of their optimal policy preferences.  But if a Progressive Party president had proposed the health care reform, the 15-25 senators from Connecticut for Lieberman Party would have no stake at all in anything passing, so the single payer or robust public option bill proposed by the president would either be negotiated down to some trivial tinkering around the edges by CFL committee chairs or killed entirely.   Both sides would be freer to propose amendments that could blow up the bill. And if the Connecticut for Lieberman Party controlled the White House, there’s zero chance that they would invest their agenda-setting authority in comprehensive health care reform in the first place.   PR is democratically defensible and the filibuster isn’t, but I still think that he burden of proof is on anyone who wants to make the legislative process even more cumbersome and less accountable that it already is.
• A PR system could in theory be created by an act of Congress (and at the state level, but there would be a serious collective action problem inherent in doing it state-by-state, as the strong party in the first major states that went in that direction would be put in a serious disadvantage.)   But assuming that multiple parties would also run at the presidential level, this would mean that virtually every election would be thrown into the House of Representatives with each state delegation getting one vote, a democratic disaster on multiple levels.  So, really, there’s no way to do PR properly without a constitutional amendment, rendering potential reform DOA.
• I also don’t agree that more parties would benefit politics by “getting more ideas” injected into the political system.  The decentralized American system generates a surfeit of political ideas for great to awful.  The problem isn’t the production of ideas.  The problem is getting them implemented by sclerotic American political institutions, particularly if they’re unfavorable to powerful interests.  PR would, if anything, exacerbate this problem rather addressing it.

I have no problem with PR in principle or within a parliamentary system, but in the American context I think it’s a very dubious solution to a misdiagnosed problem.