Adam Nagourney reports:
Lawmakers came into office this year representing districts whose lines were drawn by a nonpartisan commission, rather than under the more calculating eye of political leaders. This is the first Legislature chosen under an election system where the top two finishers in a nonpartisan primary run against each other, regardless of party affiliations, an effort to prod candidates to appeal to a wider ideological swath of the electorate.
If you read the article, you will see some early reports that this is working out well, and allowing for less partisan decisions. Focusing on the primaries question alone, I have a few points:
1. If the median voter theorem holds, the nonpartisan primary shouldn’t matter much. That said, the median voter theorem in this context probably does not hold.
2. Imagine you have an expressive voting model, where a big chunk of the electorate will vote for a Democrat no matter what and will not vote for a Republican in a nonpartisan primary. In essence that is like allowing Republicans to vote in the (decisive) Democratic primary. If Republicans vote expressively, they will vote for their own and we are back to this change not mattering. If some Republicans vote strategically, to pull the Democrats closer to the center, they may succeed in doing so. That said, if Democrats observe Republicans voting strategically, the Democrats may respond by voting more strategically themselves, to pull their candidates closer to the Left. Alternatively, Republican candidates may respond by moving to the center, hoping to appeal to the strategic Republican voters. If the more moderate Republicans become “too acceptable,” they may prevent the strategic Republican voters from voting for Democrats and thus influencing the final outcome.
Multiple equilibria, etc., but it is not obvious that, after a few iterations, you end up with a better or a less partisan outcome.
3. Many electoral changes have short-run effects which differ from their long-run effects. In the longer run, there is a Beckerian equilibrium of pressure groups and those groups may manage to master the new procedures and reimpose their will on the proceedings, albeit with a lag. Changes in electoral systems should never be judged too rapidly.
4. #3 is especially true when an electoral system change alters “what happens within parties” vs. “what happens outside parties.” It takes a few iterations for parties to adjust to the new rules and change their nature, to adapt to the new equilibrium.
The bottom line: One virtue of federalism is that such experiments can be tried. I’m all for it. But it is far too soon to think this is a big deal.