A group of Burlington residents have garnered the necessary 1654 signatures to put a simple question to voters: Should Burlington revert to its old way of electing its mayor?
Instant runoff voting (IRV) — which allows voters to rank candidates as a way to choose a winner on election day rather than holding costly separate runoff elections – was perhaps the most controversial element in a five-way race for mayor in March.
If there is no majority winner (50-plus percent) after all first-place ballots are counted, candidates who statistically can’t win are eliminated, and their second-place votes are distributed among the remaining candidates. This process is repeated until someone with 50 percent of the vote emerges.
If IRV is repealed, then the next election for mayor will revert to the old system: A simple plurality (40 percent) wins the day. If no one has 40 percent on election night, then a special runoff between the top two candidates will be held 30 days later.
A multipartisan group calling itself “One Person, One Vote” held a press conference Tuesday afternoon in a City Hall conference room to declare that IRV was a convoluted system that frustrated voters and may have depressed voter turnout.
“I was an early supporter of IRV,” said Democrat John Ewing. “But I’ve been disappointed in the way it has worked. I think it has proven itself to be a disservice to the voters. I think it’s extremely convoluted and that voters don’t understand how it works.”
If that’s true, it would be a slight turnaround from a poll conducted in 2006 — the first year IRV was used in Burlington to elect a mayor.
In 2006, a University of Vermont professor and his students polled Burlingtonians about their understanding of IRV and found that 63.4 percent of voters said they liked IRV, while 17.9 percent disliked it and 18.7 percent did not have an opinion or didn’t know whether they liked it. Those with a postgraduate degree liked IRV more than those with a high school diploma or less. Only about 8 percent of those polled found the system confusing.
Several of the IRV opponents, including Ewing, backed Republican Kurt Wright for mayor this past March. Another Wright supporter, David Hartnett, also collected signatures for the group and said this was more than a sour grapes campaign.
“This has been a grassroots effort,” noted Hartnett. “We waited to bring in the signatures because we didn’t want this to be about Kurt Wright losing after being ahead, or Andy Montroll who had more first and second place votes and didn’t win. We wanted this to be about IRV. In fact, I don’t think a lot of people understood what they were voting for in 2005 when they approved it.”
Hartnett and other IRV opponents said after two mayoral elections it was time for voters to weigh in — again — on whether the voting system was working.
“There was a lot on the ballot that year, and IRV got kinda lost in the shuffle,” said Hartnett. “In the next two and a half months there will be time to focus on IRV.”
Hartnett and others said that, while the lion’s share of early grassroots energy this year emanated from the New North End, petitioners come from throughout the city and represent all major political parties in the Queen City.
Also joining Hartnett and Ewing at Tuesday’s event was Sam Osborne, a prominent Burlington Democrat, Chuck Saleen, a New North End Democrat, and Linda Chagnon of the New North End. Chagnon alone netted roughly 500 signatures.
Ewing and other IRV opponents said the system has done nothing to increase voter turnout. In fact, only 27 percent of registered voters turned up at the polls in March 2009, which seems pretty dismal.
That can be compared with a voter turnout of slightly more than 30 percent in 2006. That said, 2009 had a better turnout than 2003 (22.3 percent) and 2001 (18.7 percent).
In March 2009, there were five candidates on the ballot, four of whom ran strong campaigns: Progressive Bob Kiss (incumbent), then-Councilor Andy Montroll (Democrat), James Simpson (Green Party), Independent Dan Smith and then-Councilor Wright.
Here’s how the vote broke down on election night: In the first round, Wright had 2951 (33%) votes to Kiss’ 2585 (29%) and Montroll’s 2063 (23%).
In the second round, Wright had 3294 (37%) to Kiss’ 2981 (34%) and Montroll’s 2554 (29%) after Smith’s and Simpson’s votes were eliminated and their voters’ second preferences divvied up among the other three candidates.
In the final round, Montroll was eliminated and 1332 of his preferences went to Kiss, while 767 went to Wright. That put Kiss on top with 4313 (51.5%) and Wright with 4061 (48%).
Hence the inspiration behind this T-shirt, which offered a new definition of IRV: “Keep Counting ’Til Bob Wins.”
In 2006, it also took Kiss more than one round to win a majority of votes, but he was in the lead in both rounds against his closest challenger — Democrat Hinda Miller.
This year, however, the margin was close enough that Wright asked for a recount. Wright called off the hand recount as soon as it was clear that it was not differing from the election night results.
Here was my take in a “Fair Game” column shortly after the election, after some fumed that Wright had the election stolen from him. I mean, really, who would have thought a left-leaning, liberal city would pass on putting a Republican in the mayor’s office? Incroyable!
IRV proponents noted at the time, and will undoubtedly note again in the coming months, that IRV worked as it was designed.
They said the ranking system saves time and money by allowing voters to rank their candidates and take care of the runoff the same night. Also, fewer voters actually take part in separate runoff elections.
But that doesn’t stop folks from claiming hell froze over last March. Vermont GOP Chairman Rob Roper called Kiss’ reelection a “travesty” and claimed the “will of the voters was circumvented.”
Was it true? I decided to look more closely at how the votes broke down, and here’s what I found:
People may not be happy that Kiss was reelected, and certainly at today’s press conference the word “leadership” was tossed around a lot — mostly in a negative sense. As in, Kiss offers none.
Look, it’s no secret that Wards 4 and 7 are more conservative than other parts of the city and that they were going to go overwhelmingly for Wright. And they did.
The question for March 2010, however, is how much of the anti-IRV effort is truly a referendum on IRV, and how much is a referendum on Mayor Kiss. His term isn’t up, technically, until 2012, but a spate of leadership crises related to Burlington Telecom, the Moran Plant redevelopment and the loss of General Dynamics practically put Kiss in the lame duck category. And to think it’s only the guy’s first year of a three-year term.
Sounds like IRV could soon find itself another fossilized relic that anthropologists hundreds of years from now use to define what could be dubbed Burlington’s Progressazoic Era.
This article appears in Dec 30, 2009 – Jan 5, 2010.
Got something to say?
Send a letter to the editor and we'll publish your feedback in print!
Me: “So why isn’t it on the ballot this March.”Jimmy: “Because you didn’t get it on there?”Me? I stirred up as much interest as I could. Wrote papers explaining what’s going on. Even made a lame attempt at a petition with Condorcet. But convenient ignorance seems to prevail over *ostensibly* less convenient knowledge (there is nothing more inconvenient about the Condorcet rules, same ranked-order ballot as IRV, but the counting of the vote does a better job than IRV and that is not inconvenient to anyone).No, that is what my complaint is about. Stupid pro-IRV people who claim that it worked just fine in 2009, just as it’s designed to. And cynical anti-IRV people who feel entitled to rule even when the majority of the electorate is against them.They’re both wrong. But IRV is better than plurality. If Kurt Wright had squeezed out a 40% “victory” with 60% voting against him or squeezed out a “victory” with less than half the electorate returning to the polls on Runoff Day, his election would be far less legitimate than Bob Kiss. We know that Bob Kiss was preferred to Kurt Wright because 252 more voters preferred Kiss over Wright on Election Day. But the problem with IRV is that it didn’t ask the same question regarding Kiss and Montroll. If it had, we would find that 587 more voters preferred Montroll over Kiss. That’s why and how IRV screwed up. It did not screw up by keeping Kurt Wright out of the mayor’s chair. The Repubs don’t want to admit it, but Kurt Wright came in 3rd last March. He was the 3rd most preferred candidate by Burlington voters. Bob Kiss was the 2nd most preferred candidate and Andy Montroll was the most preferred candidate. That is what the ballots said. But IRV had funky rules that prevented Montroll from going up against Kiss.
But Owen, Loyal doesn’t make any sense.Turnout: 2002 Council and Mayoral 7751 2004 Council and Mayoral 8647 2006 Council and Mayoral (IRV) 9865 2009 Council and Mayoral (IRV) 8980turnout is at leas as good (even a little better) with IRV. but the question is what would the turnout be with a delayed runoff (which is what Loyal apparently want to return to)? we don’t have an example for the Mayoral, but how about the 2009 Ward 7 race: Election Day 1573 Runoff Day 787half as much. Vince Dober cannot claim that his election securely reflects the will of Ward 7 voters when he was not the plurality winner on Election Day and squeeked by a thin victory on Runoff Day when only half the previous voters got to the polls. but, it was the law (the rules agreed to in advance) so Vince’s election to Council is legit.People who choose not to vote for secondary or tertiary choices on the ranked ballot are people who choose not to vote in the runoff. But their choice was no inconvenience to them. They *could* have chosen to vote for a back-up candidate (as 2nd choice) on Election Day with no extra effort. But returning to the old system would require people to go through the extra bother and expense of returning to the polls 3 weeks later and we *all* know (if you deny this, I would say you’re untruthful) that far fewer people will show up to the polls. Elections with far less voter participation cannot be considered to be as democratic as elections with greater voter participation. Unless you’re someone opposed to Motor-Voter, we do not want to make voters jump through extra and unnecessary hoops to vote and that is what a delayed runoff does.Even though it did okay in 2006, IRV *did* screw up in 2009. It elected the wrong candidate. It suffered a spoiler problem (but Plurality is even worse) and it will encourage strategic voting by Republicans that liked Kurt and disliked Bob the most. These 1500+ Wright voters have now figured out that their vote for Kurt as first choice kept Andy out of the final IRV round and actually *aided* electing the candidate they least preferred. If IRV survives, these folk have to be thinking about how well their interests are served by being loyal to their favorite candidate. If they had forsaken their favorite candidate (or just stayed home), their lessor disliked candidate (Montroll) would have gotten into the final IRV round and the certainly beaten their most disliked candidate.So IRV encourages strategic voting and that is a bad thing. But Plurality is even worse. Especially for a Dem/Prog/Liberal in Burlington (the majority voter in Burlington). It encourages strategic voting even more. Progs and Dems are gonna have to worry about voting for their first choice and may decide to forsake their first choice just to keep the Republican from winning with 40%.
Give ’em Hell, Loyal.Repeal IRV
“So why isn’t it on the ballot this March.”Because you didn’t get it on there?
One last thing, if anyone is interested, I have a 6-page paper entitled “The Failure of Instant Run-off Voting to accomplish the very purposes for which it was adopted: An object lesson in Burlington Vermont”. I wrote it last March and disseminated it among a few. But it was ignored in favor or either Dumb (IRV) or Dumber (Plurality).If you want a pdf copy, email me at rbj@audioimagination.com and I’ll send you a copy.
About electing the “least offensive” candidate with a purported small base, I really doubt that is what the case was with Andy Montroll. No one had a majority base of first choice. But Montroll killed either Kiss or Wright (or Smith) with the sum of 1st-choice and 2nd-choice votes. But it was only the 1st-choice (and the transferred or promoted votes from the eliminated Smith and Simpson) that counted in the elimination which is why, under IRV rules, that Montroll didn’t get to the final round (where he would have beaten Kiss or Wright or anyone else).It is true that the Condorcet method favors the *centrist* candidate and that is because for voters on either the left or right fringe, their 2nd-choice candidate is not likely to be the guy at the opposite side fringe. Both the Republicans and the Progs will likely choose the Dem as their 2nd-choice. I don’t see electing a centrist as a bad thing, but that (centrism) is no reason to elect a candidate. The reason the Condorcet winner should be elected is because that is the candidate that the majority of voters prefer to *any* other candidate when asked to choose between the two. That is the only reason (majority rule) and that is good enough.
Why are the anti-IRV folk calling their group ‘one person, one vote’? that seems awfully disingenuous when IRV undeniably provides ONE single transferable vote.If the anti-IRV group had a solid argument for going back to 40% +1 it seems they wouldn’t need to attempt to mislead people about how the system works.This tactic seems really disrespectful to the voters when you consider that IRV was approved just a few years ago by almost a 2/3 majority. If the anti-IRV folks continue telling the voters they were ignorant and didn’t understand what they were approving they’ll likely lose. If they make it about Bob Kiss they might win. (And I’ll lose the 6-pack I bet on this to Haik Bedrosian before the BT mess.)Sad either way.
Well, normally we think that the least offensive candidate (from the POV of the majority of voters) is better than a candidate that is more offensive.The breakdown you ask for is available at rangevoting.org (or you can write your own computer program and get it from the Burlington City Clerk data). It is:1332 M>K>W767 M>W>K455 M2043 K>M>W371 K>W>M568 K1513 W>M>K495 W>K>M1289 WOf the 2008 voters that expressed an interest between the three and picked Wright as their first choice, by a better than 3 to 1 ratio (1513 vs. 495), they preferred Montroll over Kiss. So by voting for Wright as their first choice, they ended up solidifying Kiss’s victory in IRV. And this would not have happened if the ballots were tabulated using Condorcet rules rather than IRV rules.But nothing could have elected Wright except the illegit strategy of a delayed runoff where more Conservatives bother to come back to the polls than the Liberals. That’s the only way they could hope to get Wright elected in a town such as Burlington Vermont. But, at least, the Republicans should have been able to vote for their guy without helping elect the guy they liked the least.
Is that true? Who was the 2nd choice of the Wright voters? Historically speaking, Burlington Republicans have usually gone with the Progs over the Dems. Strange but true. Just look at the number of Kevin Curley’s votes that went to Bob Kiss in 2006.That said. Condorcet does seem interesting. Has it been used anywhere? In such a system I’d worry that someone might be elected who lacks a base of support… What’s to stop a candidate who is likable but doesn’t take strong positions from winning just be being the least offensive?
So why isn’t it on the ballot this March. Why are we choosing between Dumb and Dumber?