Mar032010
Burlington rejects instant-runoff voting
Filed under General by david brooks at 9:20 am
My column in the Telegraph today (read its prose poetry here) is about alternative methods of voting, in which I pointed to Burlington, Vt., which uses “instant-runoff voting” in its mayoral race. I lost track of what is happening in Vermont, because yesterday Burlington rejected this voting method, which has been used in two elections, and will return to traditional “one-man, one-vote for one-candidate” rules.
Here’s the Free-Press story, which includes this comment that reflects the problem with IRV:
Mayor Bob Kiss was elected twice through the system. In 2009, Kiss prevailed after three rounds of runoffs. He received just 29 percent of the initial vote in a five-man race but defeated Republican Kurt Wright after three other candidates were eliminated through IRV. “This was an anti-Bob Kiss issue” for the repealers, Keogh said, those unhappy “that somebody can win with 29 percent of the vote.”
March 3rd, 2010 at 12:29 pm
It's instructive that it's easy to pick on a new voting rule when someone wins who a lot of people don't like. Bob Kiss is the only person ever to have won with IRV in Burlington, and now is deeply unpopular. Many saw this vote as a referendum on him. If it had been a straight referendum on him, he would have lost overwhelmingly; as it is, IRV lost 53% to 48%.
There are those who are saying majority rule and IRV are coming back to Burlington. See:
http://7d.blogs.com/blurt/2010/03/councilor-cal...
March 3rd, 2010 at 10:35 pm
IRV EVEN baffleD the elite students of University of Virginia. They just had an IRV election with results none of the candidates could explain.
See:
Marginal mayhem
The University Board of Elections should educate the student body further about its voting methodology Lead Editorial / Opinion March 3, 2010
http://www.cavalierdaily.com/2010/03/03/margina...
In that election, the candidate with the most 1st and 2nd choice votes lost. Even the winner of the UVA's IRV election didn't understand the results.
Consider that UVA “has ranked …among the top 25 nationally since the first U.S. News rankings came out in 1988.”
What more do we need to know to tell us that instant runoff voting is bad for voters?
As one Burlington Voter said: “Being charmed by it ideologically is quite different from experiencing how it twists the results of an election.”
http://repealirv.blogspot.com/2010/03/irv-repea...
March 4th, 2010 at 3:53 am
Joyce…We differ on this issue and the interpretation of what happened in Burlington.
On student elections, what I find encouraging is that so many colleges and universities are using IRV — up to nearly 60, including Dartmouth there in NH. I'm sure it's true that each new class of students has to learn about it, and sometimes the education of student voters about their new system is limited. But see some examples of students reacting to adoption of IRV.
http://archive.fairvote.org/?page=2288
And, as we've discussed before, you keep blogging that Georgetown University got rid of IRV, but in fact it keeps using it. Time for you to drop that from your repertoire.
March 4th, 2010 at 11:49 am
Rebutting Rob Richie. Georgetown U ditched IRV in January of 2009 because of controversy over the bizarre election results and lack of confidence in the election outcome. Let me just refer to “The Hoya”, which is paper for GU:
New Voting System for GUSA
Presidential Election to Feature Plurality System
By Elizabeth Rowe, Kathleen Nahill and Katie Kettle | Jan 29 2009 | GUSA |
On Tuesday night the GUSA Senate voted to change the method of voting in the presidential election from instant runoff voting to a plurality system. This change comes in response to controversy over last year’s election, which resulted in the selection of Pat Dowd (SFS ’09) as Student Association president and James Kelly (COL ’09) as vice president.
Instant runoff voting was adopted by the Election Commission in 2006 after five years in which no GUSA ticket won the majority of student votes.
…
Last year, the senate rejected the results of the presidential election, citing problems with the IRV system. A second election was held with only four of the seven original tickets on the ballot. D.W. Cartier (COL ’09) and Andrew Rugg (COL ’09), who won the first election with 51.2 percent of the vote, were defeated by Dowd and Kelly in the second election.
GUSA Vice Speaker Brian Wood (COL ’09) explained the need for the bylaw change.
“I got a lot of calls [about the last election],” he said. “I have gotten a lot of resistance to instant runoff voting.” Senate Speaker Reggie Greer (COL ’09) said he supports the plurality system, where the ticket that receives the most votes wins the election, regardless of whether or not that ticket receives the majority of the votes.
…
http://www.thehoya.com/news/new-voting-system-f...
March 4th, 2010 at 12:02 pm
Rob Richie says he and I disagree. But NOT SO. Burlington Voters disagree with Rob Richie.
Unpopular mayor Bob Kiss, who owes his seat to IRV, has threatened that the decision of the voters on Mar 2 is not final. He is not satisfied with the 52% vote (thats a 4 point margin by the way). Never mind the campaign to keep IRV was called “50% matters). Never mind that some of the politicians pushing IRV were put into office with less than 50%.
Well, Burlington voters were David, the outside interests and politicians were Goliath, and David won.
Burlington will repulse the outside money interests again, see
“Fact Check
Is this really the end of IRV in Burlington?”
Burlington politicians are examining the best way to move forward in the wake of the vote to repeal instant runoff voting. The main IRV supporters say they accept the outcome and are willing to move on, but Mayor Bob Kiss says this may not be the end of the debate
http://repealirv.blogspot.com/2010/03/fact-chec...
Never underestimate a grassroots campaign that is fighting outside interests. Its their city, it is their vote. Another blog by a Burlington voter:
Tuesday, March 2, 2010
IRV REPEALED
People of Burlington have told the politicians how we want to vote, and it is NOT by instant runoff. We want real runoffs, and will defend the 40% threshold to protect minority rights.
http://repealirv.blogspot.com/2010/03/irv-repea...
The Burlington voters won't let FairVote dictact to them how they will vote.
People will tell the politicians how we want to vote! This is a citizen initiative and we have the power.
…
Ballot Question #5 is a CITIZEN INITIATIVE.
VOTE YES to REPEAL IRV
Politicians, FairVote Maryland, vpirg -
BACK OFF!
…
http://repealirv.blogspot.com/2010/02/city-vote...
March 4th, 2010 at 1:59 pm
I would like to suggest a middle path between the pro- and anti-IRV positions expressed here. As I posted on Ballot Access News blog, ranked choice voting could be a good election reform for the future. But for now RCV seems too complicated for most of the public to feel comfortable with. I have been advocating that reformers consider simple, straight-forward methods like limited voting and cumulative voting as an interim strategy to RCV.
LV and CV are easy to use and understand and can be implemented with minimal or no change on all existing voting equipment. RCV cannot be implemented beyond a single election jurisdiction without a major retrofit of voting equipment. In a statewide or legislative election, every jurisdiction would need to standardize its equipment with every other jurisdiction to accommodate the transfer of ranked votes across jurisdictional lines.
It's also difficult to see how RCV could be implemented on optical scan equipment in a full presidential-year general election. With multiple contests and mutiple candidates per contest–each requiring a ranked choice ballot–voters would have to be given a folder full of ballots to complete their voting.
Limited voting and cumulative voting is a way to introduce the public to multimenber proportional voting. And if they like the concept, it would not be a huge challenge to later “overlay” ranked choice voting.
March 4th, 2010 at 1:59 pm
IRV tends to choose winners that voters do not like as it did in Burlington, VT by eliminating the centrist candidate who the majority of voters favored above other candidates in a pairwise comparison. Burlington demonstrates how IRV does *not* solve the spoiler problem of a nonwinning candidate causing a candidate to win who most voters oppose whenever there are three strong candidates.
Burlington, VT also illustrated IRV's tendency to elect extreme right or left candidates and eliminate centrist candidates due to its fundamentally unfair method of counting rank choice votes that allows the voters supporting the least popular candidates to have the most say (have their 2nd choice votes counted) but the voters supporting the more popular candidates often never have their 2nd choices counted when their 1st choice candidate loses.
March 4th, 2010 at 7:44 pm
Lee, I agree with you that simpler, precinct-summable methods like cumulative voting and limited voting that can be easily audited and treat all voters' votes equally are good reforms, but *not* because they'll lead back to IRV.
RCV or rank choice voting is not bad, and could even be very good if the rank choice votes were counted using a fair counting method like the Condorcet method that treats all voters' ballots exactly alike, unlike IRV that counts the 2nd choices only of some voters whose 1st choice loses and eviscerates voters' rights and does not solve any problems that plurality has but causes additional problems.