Thoughts on IRV

Hostility toward the recent shift to “pick-a-party” primaries in Washington has resulted in a Charter Amendment in Pierce County(a well populated county immediately south of Seattle’s King county, with some conservative rural-flavored outer suburbs and the urban, working class Tacoma) That will end the primary for several local offices, and replace them with one Instant runoff election. At least a few scholars of voting behavior of my acquaintance have long been proponents of IRV. Some of the advantages seem obvious to me; for one thing, it will allow those who fetishize voting their conscience to do so without disastrous results. It also strikes me as a more efficient and accurate aggregator of traditional first-past-the-post elections. The most obvious lines of critique, it seems to me, are 1) Voters won’t understand it, and 2) it removes significant power from the parties. The latter would not be a net negative in the eyes of many; I’m undecided on this front. Had I been a citizen of Pierce County, I’d have thought a lot harder about how I felt about this, but I suspect I would have joined the majority and voted for the Amendment. The former is indeed a problem, but a solvable one and not in itself a reason to avoid such a change.

As an aside, I should note that unlike most of my fellow Washingtonians, I had no problem whatsoever with a switch to the pick-a-party primary; making voters declare a party affiliation for the purposes of participating in the parties primary seems profoundly reasonable to me–it wouldn’t even prevent me from messing with the GOP primaries, as long as I was willing to forgo participating in my own party primary. This seems like a reasonable trade-off to me, but the fetishism of independents reacted strongly against it. If preference aggregation is done through the two party system prior to the general, I’ve got no problem with the parties exercising a little more control than the open primary granted them. But at the end of the day, I’ve got no strong commitment to preference aggregation taking place through the parties first.

It’s not surprising the both major parties opposed this, but this Seattle Weekly article is just weird. It contains a fair amount of alarmist rhetoric about the coming IRV revolution, how Democratic party activists aren’t taking this grave threat seriously enough, etc etc. Perhaps the reason Democratic Party Activists (as opposed to politicians and other insiders) aren’t too concerned about this is because they are democrats as well as Democrats and have interests beyond the party they work for. In effectively one party constituencies like Seattle one can certainly understand the sentiment. But while it’s clear Democratic party officials might not welcome IRV, the article seems to want us to think the rest of us should be worried as well. The only real effort to give a reason for this concern is found in this passage:

But IRV could have undesired results for voters and candidates, says Matt Barreto, an assistant professor of political science at the University of Washington. He points out that in IRV elections with a high number of candidates, a spoiler effect could still prevail.

Imagine a mayoral race with eight candidates where the bottom four candidates’ votes have to be harvested to create a majority. What if the candidate in third tends to be the second choice of those who voted for the bottom four? He gets all those votes. That would mean the person who got the highest number of first-choice rankings could lose, while the guy who initially came in third takes the election.

Missing: an account of why this hypothetical outcome would be worse than the alternative. Is the implied premise that those who choose an unpopular candidate as their first choice shouldn’t have their votes taken into account? We’d abhor that logic if applied to a multi-election primary process, and rightly declare it undemocratic. There strikes me as no a priori reason to prefer a candidate who is strongly preferred by a plurality and disliked by the majority to a candidate who has a broad base of moderate support.

Two final thoughts:

1) I’m strongly opposed to plebiscitarian policy making, for all the obvious reasons, which have been well covered on this blog over the years. But this is one area where plebiscitarian reforms make sense; fair, accountable elections offer a source of legitimacy for legislative outcomes, but how can the rules of voting be legitimate? Voting on voting systems occasionally hardly solves the legitimacy puzzles at the heart of democracy, but it makes a non-trivial effort to address it.

2) I’m siding here with the position of that noted LGM bete noire, The Green Party, as well as other third parties. This is partly because I think their position on this is basically correct, but also because I think they may be dissapointed by their fate under such a system. I expect they’d get more votes, but a number of the unsupported claims third party types are inclined to make regarding the ways in which two party dominance disguises a wellspring of hidden would-be supporters for their views and platform. I strongly suspect this isn’t true, but if it turns out to be the case I’ll admit my mistake. IRV might well bring about put up or shut up time for the Greens, which they and I welcome for very different reasons.