I’m sure we will hear a lot about the RCV measures that just were on the ballot. Only in DC did one pass/cause RCV adoption (or retention).1
I want to float the possibility that these reforms do not scale up. We have heard a lot about their momentum at the local level. Yet some cities are large enough to host two-party politics. Most states certainly are. Two-party politics matters even if a jurisdiction is (or seems) lopsided in its partisanship. The point is that large jurisdictions have competitive, factional politics in the wrapper of an ostensibly two-party system. (We might even see some of these jurisdictions become less lopsided; stay tuned.)
There are many reasons why an anti-party reform might not “scale up.” Elsewhere I have written about governability problems. Here, by “scale up” I mean: be reasonably easy to pass.
A successful anti-party initiative seems to require the backing of one or another major party. Ironic!
Now, if that major party were sufficient to deliver victory on its own, there would be no need for electoral-system reform. Hence the second requirement: support from a defecting faction of the opposing party.
Here are some examples. The x-variable in each is support for an out-party nominee. The y-variable is support for an anti-party reform package. The diagonal in each represents y=x. The key feature of each graph is that data points cluster above y=x. This is suggestive evidence of the hypothesis I’m peddling: out-party plus faction of the other.2
The obvious (to me) question is whether an anti-party reform can win on faction alone. I don’t know. I am not aware of any such case; maybe a reader is. But if I had to develop a theory about why not, it might start with the role of party cues in forming people’s views of these reforms. Negative messages (or perhaps silence) from both major parties is a lot to overcome. Perhaps reformers could overcome such messages (silence?) with arguments about “corruption,” but I suspect those messages would need to resonate with voters’ lived experiences.
- Alaska still has not been called as of this writing.
- See this journal article for the original formulation. The theory got updated in this book chapter. In both, I used a mixed-method approach because ecological inference (which I also used) is notoriously fraught.
Superficially, California’s adoption of top two in 2010 looks like a counter-example. It passed easily with both major party organizations opposed. But the parties didn’t back up their opposition with money and serious campaigning. Voters only heard from the top two side, which had lots of money from business interests and had the support of individual Republicans including the governor and lieutenant governor.
Interesting. Have you seen quantitative analysis of the referendum outcome? I know a little about the case and would be inclined to call it a realigning episode.
Sorry, I don’t recall seeing any quantitative analysis, but it’s been 14 years now. The person to ask about that is Richard Winger of Ballot Access News.
In order to fit top two/final four(five) into your insulating/realigning/polarizing taxonomy, you have to conceive of the factions and coalitions more broadly than party labels and party organizations. Top two happened in California primarily because, leading up to 2010, the business community felt that it had lost control of both major parties and needed a way to make candidates beholden to them directly rather than indirectly through party organizations. I don’t know to what extent this applies to Alaska and the several states that voted down top four(five) this month. But I don’t think the activist advocates for non-partisan elections and weak parties would get anywhere at all without the corporate backing most visibly represented by Katherine Gehl.
In terms of your taxonomy, I think I agree that it makes sense to call this process a realignment.
My taxonomy was designed with party factions in mind. “Faction” is one way to think about the unit of analysis, and I use this for teaching purposes. On-the-ground situations are more messy, even in multiparty systems: most of the Belgian Catholic bloc, for example, but not all; or most of the Brockton (MA) Republican establishment, but not all. I hope this subtle but important point is getting through to the intended audiences.
Then I’m not sure that the taxonomy fits top two/final four(five) all that well. In 2009-2010, the backers of top two in California had the support of an extremely small faction of Republican Party “moderates” (mainly the governor and one state senator), and no Democrats at all. These two Republicans were able to get the proposal on the ballot only by holding hostage a more-contentious-than-usual state budget and forcing the ballot measure down everyone else’s throat.
On the other hand, when it came time to campaign against the proposition, none of the many opponents in both parties did much beyond holding a couple of poorly attended press conferences. (Disclosure: I represented the Peace and Freedom Party at one of these.) I think very little, if any, money was spent on opposition. It’s possible that an agreement not to campaign very hard in opposition was part of the budget deal, but that’s speculation on my part.
So I think it’s fair to say that this reform effort came from outside the political parties. I don’t know what that does to the taxonomy.
It doesn’t hurt the model/taxonomy explanatory power.
I would read a mixed-methods study of how the reform happened.