We got a number of great responses to my posting on polarization. Here’s a response:
A number of respondents suggested making it easier to vote in primaries, including internet-based or mail-in voting. I’ve been suspicious of this in the past, because of fears of concerns about identity-confirmation and a loss of the civic ritual of voting. But the former concern seems primarily technological and relatively easy to overcome, and the latter less important the fewer people who actually show up to vote. It seems highly likely that this would, in fact, increase voting in primaries, and as a consequence decrease the dominance of ideological hard-liners.
Another interesting suggestion was to “form an organization of moderates that recommends candidates in both primaries.” This makes a lot of sense, but it implies that moderates have a sufficient identity as moderates that they would seek out these recommendations, and that there would be a patron willing to put up the resources to fund such an operation. But this seems to have few if any downsides, and at least some upsides, and might not actually cost that much.
One respondent suggested that voters register for whichever party was the majority in their district, which would have the effect of watering down the power of the ideological stalwarts. This might work, but it implies a significant amount of coordination, and would only work in non-competitive districts (that is, in districts where the primary was tantamount to election), but King’s point is that polarization happens as much if not more in competitive districts, so this sort of strategic registration would be a non-starter there.
One idea that was suggested in a couple of posts was scrapping the primary and moving to run-off elections instead. They do this in Louisiana, and I’m not positive that it’s had the intended effects, at least on the Republican side. Can our readers suggest any evidence that run-off elections encourage moderation?
I also want to add that I did NOT mean to suggest that gerrymandering was fine. It’s not. But the reason is not that non-competitive seats lead to more polarization, which King’s evidence suggests is not that case. The real reason is that the House of Representatives is supposed to be sensitive to shifts in voter preferences. That sensitivity could happen even if it meant just swapping pole-huggers from one party for pole-huggers from the other when voter preferences shift. It’s probably also the case that more competitive elections would lead to greater voter knowledge, since more voters be bombarded with information every two years. It’s not clear that this would make them trust their MCs any more than they already do, but I’m also not sure whether this is even an important outcome.
Thanks for all the responses!