Why Alaskan RCV might survive

It has been fascinating to watch the evolving count of votes in Alaska Measure 2. This would replace jungle primaries followed by an RCV round with a ‘conventional’ system of closed primaries followed by a plurality vote. As of Election Night and the days immediately after it, Measure 2 looked to be going the same way as several other statewide measures: no RCV.¹

Here is why retention in Alaska would not be very surprising despite what might seem like a ‘wave’ against this reform.

I find it useful to keep track of the mode by which some reform happens. By “mode” I mean two things. One is the relationship of the adoption coalition to the incumbent coalition: insulating (reform by incumbent coalition), realigning (reform by groups out of power plus defectors from incumbent coalition), and polarizing (ends against the rest). The second aspect of “mode” is the configuration of party factions and parties-as-parties making up the coalition. Both factors are helpful for understanding choice of reform details (e.g., RCV vs. lists).

The former aspect of ‘reform mode’ seems especially helpful for sorting through the November 2024 measures. Alaska’s 2020 adoption had several signs of being insulating. Much was made of the reform’s potential role in protecting an incumbent U.S. Senator. Early analysts observed that the core of the reform coalition was centered around a bipartisan coalition that already controlled the state’s lower chamber. Subsequent analysis is consistent with the idea that there was realigning activity as well, bringing control of the upper chamber into agreement with that of the other offices.²

I did not hear much about target politicians or coalitions in the non-Alaskan states. Thus I am inclined to believe that the ‘intended mode’ in them (minus Missouri³) was realigning, and this strikes me as a particularly difficult coalition to assemble. The task in Alaska was fundamentally different: preserve the coalition that won in 2020, and offset defections as necessary.⁴

With all of that in mind, and without knowledge of defections from the original coalition, I would have expected either of the following outcomes:

• RCV/jungle survives.
• RCV/jungle gets defeated as part of the general “swing” against Democrats, to the extent that RCV had become associated with the Democratic brand.

It doesn’t look like the latter happened. If anything, the aggregate results suggest there may have been a fair amount of drop-off and/or ticket splitting. Surely we will hear more as analysts dig into the cast-vote record.

A result as close as Alaska’s has precedent. I found several “repealers” (as they used to be known) that lost by just a few hundred votes in big cities. One of these spent at least 14 months in recount, and I could not find in the newspapers that a result ever was announced.

1. cinyc9 on X/Twitter is an excellent follow if you’re looking for real-time updates.
2. This combination of modes is the subject of some work-in-progress.
3. I would begin with the descriptive hypothesis that Missouri was insulating.
4. There is an affinity between the sense in which I use ‘reform mode’ and the sense in which Munck and Leff (1997) used a similar device to explain the fates of democratic transitions.