Monday, June 04, 2018

RCV and the illusion of choice 2

London Breed

Along with the minor issue of saving money by not paying for run-off elections, the Fair Vote report on Ranked Choice Voting makes a fetish of voter turnout, as if voting itself is the main thing in a democracy, not the quality of the campaign and a sensible voting system.

If voters are participating in a flawed system, the value of their votes is debased. In Russia Vladimir Putin recently got more than 76% of the vote with 67% turnout, the kind of turnout San Francisco gets only in presidential election years.

After a lot of obfuscation, the Fair Vote report admits that the RCV system doesn't actually improve turnout in SF:

Voter turnout across the single round ranked choice voting elections with three or more candidates averaged 42.5% with a median of 48.2% (Figure 8). Turnout was highest (62.9%) in the 2004 Board of Supervisors District 9 race and lowest (18.9%) in the 2013 Board of Supervisors District 4 race. Across only the single round Board of Supervisors elections with three or more candidates, turnout averaged 46.1% with a median of 50.7% (page 15).

Fair Vote makes a remarkable claim about the RCV's success for Asian-American voters:

A FairVote analysis of the 2006 District 4 open seat election reveals how well Asian American voters use the system. District 4 is the only Board of Supervisors district that is majority Asian American. In the 2006 ranked choice voting (RCV) election, an Asian American candidate defeated a non-Asian candidate by 52.5% to 47.5%. If one simulates the election without this candidate, a second Asian candidate again defeats the nonAsian candidate in the final round. If one simulates the election without either of these candidates, a third Asian candidate defeats the non-Asian candidate in the final round (page 5).

That is, Asian-American voters have successfully used the system to consistently choose Asian candidates and reject "non-Asian" candidates! 

How does that differ from plain old tribalism---non-whites voting for non-whites and whites voting for whites? Can anyone think the results would have been different in a traditional run-off system? Are voters in the city's mostly white District One successfully using the system when they elect the very white Mark Farrell?

Fair Vote argues that RCV helps elect people of color in general in San Francisco:

In San Francisco, RCV has helped to elect more people of color. Since the implementation of RCV, people of color have won 22/25 (88%) citywide RCV elections and 27/43 (63%) of the Board of Supervisors RCV elections. In the same period of election cycles prior to the implementation of RCV, people of color only won eight (36%) citywide elections and 14 (37%) Board of Supervisors elections. Overall, 72% of all RCV elections that have occurred in San Francisco were won by a person of color. Prior to RCV, only 37% of the same elections in the same amount of time were won by a person of color (page 21).

Like cartoon knee-jerk, milk monitor liberals, Fair Vote's authors assume that electing "more people of color" is by definition A Good Thing. 

It's more likely that electing more people of color in San Francisco simply reflects the liberalism and lack of racism of San Francisco's voters, not the RCV system, particularly when you consider that the African-American population in SF, for example, has declined from 96,000 in 1970 to 45,000 in 2016

Besides, are we supposed to assume that "people of color" do a better job than whites after they're elected? Of course not. I've argued for years that London Breed has been a terrible supervisor, which has nothing to do with her race (see this and this). I was happy to vote for Willie Brown for mayor, Barack Obama for president, and Kamala Harris for senator. 

Breed was originally elected District 5 Supervisor with 28% of the first-round votes almost by default as the ideal RCV candidate: a blank slate politically, black, a woman, photogenic, and not a homophobe in ultra-liberal District 5.

But if she had to compete with Christina Olague in a run-off election, Olague could have exposed her ignorance on policy. Alas, the RCV system has saddled District 5 with Breed after two essentially issue-lite elections (see this and this).

Fair Vote boxes off a discussion of the bizarre election of Malia Cohen as an "outlier" (page 5). George Wooding's analysis was more succinct:

...District 10 Supervisor Malia Cohen won by receiving only 11.7% of first-place votes cast in her district. After 19 rounds of ballot counting, she finally received 51% of the remaining votes by tallying 2,878 total votes. Less than 50% of District 10 voters even voted for Cohen. Cohen won because she was the best at attracting second-and third-place votes of candidates who were eliminated. Is this representative democracy?

Yes, it is---but it's a bastardized version of democracy. And in an election decided by RCV, less extreme variations of that result are common, since the winner is often elected with a small percentage of voters overall. Even if turnout was often lower in runoff elections, at least the winner got a majority vote. 

The Fair Vote report comes close to self-parody with its conclusion about Cohen's election:

This race represents an outlier because the number of competitive candidates made ballot exhaustion rates high due to San Francisco’s three-choice limit. The exhaustion is a consequence not of the ranked choice ballot, but of voters’ inability to rank more than three candidates (page 6).

In a San Francisco political context, where it's often hard for many to find even one minimally acceptable candidate let alone three, Fair Vote thinks we need more candidates on the ballot to rank:

We also urge San Francisco to allow voters to rank at least six candidates, as required in RCV elections in Maine. San Francisco’s current voting system license is set to expire in 2018, and the California Secretary of State recently certified voting equipment that accommodates up to 10 rankings with a similar grid style ballot design (page 23).

The first post in this series: RCV and the illusion of choice.

More tomorrow.

Labels: , , , , , , ,

Hollywood in 1905

Las Vegas in 1906

Labels: , ,