Painful vote counts under RCV?

One possible bug in ranked-choice voting is the duration of a vote count. This is especially true in the proportional representation (PR) form, since ballots may move around a lot more than in “instant-runoff voting.” Many used to suggest that painful vote counts were a cause of PR’s repeal. This claim resurfaced yesterday in a private exchange about Al Southwick’s piece on PR in Worcester, Mass. Southwick writes:

[…] But the complicated vote count took days to complete. Worcester voters sometimes had to wait a week to learn who had won seats on the city council and school committee.

[…] But despite a number of referendums on changing the Plan E system, the people kept endorsing it. [JS note: Plan E was a council-manager city charter with PR.]

PR was its main weakness. After that first tumultuous election in 1949, interest waned, the number of candidates dropped steadily and voters showed a growing impatience with the long vote counts.

One who felt frustrated was Ayton F. Smith, the brusque, profane managing editor of The Evening Gazette. In 1953, on the Friday after the election, he called reporters Ken Botty and Jim Wheeler into his office and told them to have the complete list of City Council and School Committee members named in the Gazette the next day. When they explained that it couldn’t be done because the vote count and transfers weren’t finished, he growled, “Either we have that story tomorrow or you two will be looking for jobs on Monday!” He wanted to scoop the Sunday Telegram and its managing editor Frank Murphy, his long-time rival.

Thus motivated, Botty and Wheeler went to work on the story, which appeared in the Gazette on Saturday, as Mr. Smith had ordered. They got the winners right, but not in quite the right order. Anyway, it satisfied Ayton Smith.

Over the years, critics of Plan E organized several petitions to dump it. In 1959 one of those petitions came close when Worcester residents voted 35,081 to 31,019 to keep Plan E. But in 1960 they voted 40,873 to 30,386 to dump PR while retaining Plan E.

Southwick notes that candidate entry came drastically downward after that first election in 1949, so it is a stretch to claim that vote counts caused PR’s repeal, even on the basis of Southwick’s own piece. Leon Weaver suggested 30 years ago that the “complicated count” was a way to persuade voters to dump PR rather than a substantive, widely held gripe.

It’s true that the newspaper came to dislike PR by 1960, but this wasn’t due to the counting process. Elsewhere I show that reformist Democrats lost control of the city council median in 1960. This explains the newspaper’s position since the newspaper supported reformist Democrats.

Data from Worcester show how vote counts got easier after 1949. The solid trend is the number of candidates on the ballot. (I have included the effective number of candidates for readers interested in “meaningful” candidate entry and/or party competition.) Numbers below this trend indicate the number of RCV counting rounds needed to fill all nine seats in the council. The slight uptick in 1959 is likely from tension in the local party system — in other words, the actual conditions that produce a repeal — but still below the 1949 level or even the 32 who ran in 1953, which Southwick cites as another frustrating count for reporters.

worcenc

We can repeat the exercise for Cincinnati, which also had a nine-seat council. Again, vote counts were relatively easy by the end of PR. Admittedly there was a recount in 1955, but this probably reflects a tension in the local party system. (Actually, this reflects a war between organized labor and well-heeled liberals.)

cincyenc

What about that spike in 1937? As it turns out, a repeal campaign in 1939 is the only one of five I cannot predict from roll-call voting. Maybe the spike explains it, but it didn’t win.

Just for good measure, here are the figures for New York City. I exclude Staten Island, which is an outlier on district magnitude, candidate entry, et cetera. For reasons of space, this plot doesn’t include the numbers of counting rounds, but you can get a sense from the numbers of candidates.

nycenc

No, painful vote counts do not explain successful opposition to proportional representation (or RCV). Vote counts are a non-issue anyway since computers can do all the work. IBM had developed a PR counting machine by the mid-1930s, but this didn’t save PR in New York City. Cambridge, Mass. has been counting PR elections by computer since the 1990s.

Check out Jeff O’Neill’s OpaVote project for more information on computerized counting.

PR, RCV, race, turnout, and voter error

Does ranked-choice voting (RCV) baffle voters? Our (great-) grandparents used to say so. Better data and methods have led to new evidence, but the popular conversation is no more substantive than one we’d have had in the 1940s. California Gov. Jerry Brown repeated the “complicated” claim in a local-option veto message last September. Some have even suggested to me that RCV is racist.

I think the systematic bias charge is a leap. I think people are talking past each other in the popular RCV conversation. (Witness the number of commas in the title of this post.) I also think there are serious usability problems, but evidence suggests these are not limited to RCV.

First I’ll summarize the work I know. (I will not get into other good work on campaign tone, female and POC candidate entry, etc.) Then I will show my own data from American proportional representation (PR) elections since no modern work speaks to PR. Some of the error rates will be staggering.

My best guess is that high PR error rates resulted when select parties and candidates were mobilizing new voters en masse. If this were true, the actors would not be those who imposed PR in the first place. It also would explain why my error rates spike in some PR elections regarded as positives for people of color.

(more…)