Does the Alternative Vote lead to STV?

I don’t think so. The logics of adoption are different. Yet the story of Cleveland (just below) suggests that it is possible… while rare.

AV finds favor where the majority can’t agree on the candidate it wants — but can agree on the one it doesn’t. AV is an agreement to passively form coalitions in elections. I say “passively” because the vote-transfer process does the work, likely with help from elite cues.

STV finds favor where the coalition is to be worked out actively, in a legislature. STV also preserves freedom to break the coalition between elections.

In other words, the STV reform coalition is leery about leaving things to voters — because it thinks that outside forces determine voters’ rankings. It also worries that politicians under AV might not keep the promises they make in return for voters’ second-choice support.

As a first cut at testing this theory, we can compare historical lists of cities with STV and Bucklin voting. Bucklin was the dominant form of single-seat “ranked voting” in the early 20th century. Rather than proceed through sequential candidate elimination, lower rankings were added to higher rankings until the majority winner was found. From the voter’s perspective, however, it was equivalent to AV — a ranked ballot.

There were 24 cities with STV, beginning in 1915. The Center for Election Science lists 39 Bucklin implementations, 1909-17. Here is a spreadsheet of the latter.

Of 39 Bucklin cities, only Cleveland made the switch. Maxey (1922, pp. 84-5) suggests some reasons why:

The party organizations soon discovered that although they could not nominate candidates directly, they could endorse candidates nominated by petition and throw the whole force of the party organization behind such candidates. Furthermore it was discovered that the preferential-choice scheme could be turned to the advantage of the party by passing out the word to all regulars to vote only for a first choice. Thus the alternative votes of the independent voters would tend to build up the aggregate vote of the party candidates, but the regular party voters would contribute nothing to the aggregate vote of the independent candidates.

Maxey’s piece is silent on Bucklin’s partisan implications. A review of Cleveland’s list of mayors, in tandem with the Maxey piece, suggests that Republicans were using Bucklin to manage their “Teddy Roosevelt problem.” The first mayor was a Democrat, but the next three terms went to “standpat” Republicans.

Cleveland’s 1921 election was a big deal, and nationally so. Roosevelt-wing Republicans stopped giving second-choice votes to standpats. They also entered an alliance of convenience with the Democratic Party, switching the city to STV and a council-manager charter. According to the chapter in Barber (1995), the next few city councils were without stable majorities.

To put all this in abstract terms, rapid de-polarization both (a) undermined Bucklin and (b) created an opening for STV. But this trajectory was unique. Only Cleveland pulled it off.