My Turn: City voters fortunate to have IRV
Vee Gordon
// Published March 2, 2009
in
Burlington Free Press
The upcoming election of a mayor in Burlington offers a stellar example
of the value of instant runoff voting (IRV). With five candidates on
the ballot, Burlington voters have a wide choice. Under plurality
election rules common in the United States, having so many viable
candidates is likely to generate a “spoiler” dynamic. That is where
similar candidates, splitting the majority, might allow a candidate not
favored by a majority of the voters to win with less than 50 percent.
Many jurisdictions use runoff elections to avoid the “spoiler” problem and defend majority rule. But Burlington uses an even better method, also recently adopted by San Francisco, Minneapolis and Memphis, that avoids the cost and often lower voter turnout of traditional runoff elections. That better method is instant runoff voting. The League of Women Voters of Vermont has been advocating for IRV statewide for the past decade.
With IRV, Burlington voters are allowed to rank as many candidates as they wish in order of preference. If no candidate is the first choice of a majority (over 50 percent), the candidate with the fewest votes is declared defeated and voters who supported an eliminated candidate then have their ballots counted for their next choice among the remaining candidates. This process continues until there are only two candidates in the running. In the final runoff round, the finalist preferred by the majority is elected. IRV thus simulates a series of runoff elections, but with a single election.
Many people, however, are accustomed to the notion that the candidate with the highest initial vote count ought to win. When there are only two candidates the winner, by definition, has a majority of the votes. But in a multi-candidate race, with any runoff system, the leader with the most votes in the first round is not necessarily the rightful winner. The race isn’t over yet. Only after the field of candidates has been reduced to two finalists can it be determined which is the most preferred choice.
In Burlington’s 2006 mayoral race the ultimate winner of the instant runoff happened to be the leader in the first round as well, so the issue of a “come from behind” victory did not arise. With five candidates in 2009, the “top vote getter” in the first round could mathematically have just 21 percent support, while the absolute majority of 79 percent agree that they do not want that candidate. It is natural for supporters of a losing candidate, who had a lead in the first round, to try and blame the voting method, rather than accepting the reality that between the two finalists, more voters preferred the other candidate.
Vermont has a long history of requiring majority winners. Settling for mere plurality winners is a relatively recent compromise to avoid repeated voting. For most of Vermont’s history, when no candidate topped 50 percent for most offices, rather than giving the seat to the plurality leader, repeated votes were held until some candidate won an absolute majority. Although most people are unaware of this, that is still the state law for local elections in Vermont conducted at town meeting. Section 2660 of Title 17 of state law requires majority winners at town meeting. With repeated voting, the candidates with the fewest votes are dropped one at a time until some candidate achieves a majority — which is how Burlington’s IRV law works.
In the recent election of the new Republican National Committee chairman, in a field of five candidates, Michael Steele was in second place behind Mike Duncan in the first round. But rather than declaring the “top vote getter” elected, the Republican Party requires a majority winner. As candidates dropped out one at a time in subsequent rounds, Steele came from behind to win with a majority. IRV accomplishes the same thing, but without the cost or lower voter turnout typical of public runoff elections.
As a final note, I would like to observe that John McCain, Howard Dean and President Barack Obama have championed instant runoff voting. Perhaps its time to take IRV national.
Vee Gordon of Essex Junction is legislative chairwoman of the League of Women Voters of Vermont.
Many jurisdictions use runoff elections to avoid the “spoiler” problem and defend majority rule. But Burlington uses an even better method, also recently adopted by San Francisco, Minneapolis and Memphis, that avoids the cost and often lower voter turnout of traditional runoff elections. That better method is instant runoff voting. The League of Women Voters of Vermont has been advocating for IRV statewide for the past decade.
With IRV, Burlington voters are allowed to rank as many candidates as they wish in order of preference. If no candidate is the first choice of a majority (over 50 percent), the candidate with the fewest votes is declared defeated and voters who supported an eliminated candidate then have their ballots counted for their next choice among the remaining candidates. This process continues until there are only two candidates in the running. In the final runoff round, the finalist preferred by the majority is elected. IRV thus simulates a series of runoff elections, but with a single election.
Many people, however, are accustomed to the notion that the candidate with the highest initial vote count ought to win. When there are only two candidates the winner, by definition, has a majority of the votes. But in a multi-candidate race, with any runoff system, the leader with the most votes in the first round is not necessarily the rightful winner. The race isn’t over yet. Only after the field of candidates has been reduced to two finalists can it be determined which is the most preferred choice.
In Burlington’s 2006 mayoral race the ultimate winner of the instant runoff happened to be the leader in the first round as well, so the issue of a “come from behind” victory did not arise. With five candidates in 2009, the “top vote getter” in the first round could mathematically have just 21 percent support, while the absolute majority of 79 percent agree that they do not want that candidate. It is natural for supporters of a losing candidate, who had a lead in the first round, to try and blame the voting method, rather than accepting the reality that between the two finalists, more voters preferred the other candidate.
Vermont has a long history of requiring majority winners. Settling for mere plurality winners is a relatively recent compromise to avoid repeated voting. For most of Vermont’s history, when no candidate topped 50 percent for most offices, rather than giving the seat to the plurality leader, repeated votes were held until some candidate won an absolute majority. Although most people are unaware of this, that is still the state law for local elections in Vermont conducted at town meeting. Section 2660 of Title 17 of state law requires majority winners at town meeting. With repeated voting, the candidates with the fewest votes are dropped one at a time until some candidate achieves a majority — which is how Burlington’s IRV law works.
In the recent election of the new Republican National Committee chairman, in a field of five candidates, Michael Steele was in second place behind Mike Duncan in the first round. But rather than declaring the “top vote getter” elected, the Republican Party requires a majority winner. As candidates dropped out one at a time in subsequent rounds, Steele came from behind to win with a majority. IRV accomplishes the same thing, but without the cost or lower voter turnout typical of public runoff elections.
As a final note, I would like to observe that John McCain, Howard Dean and President Barack Obama have championed instant runoff voting. Perhaps its time to take IRV national.
Vee Gordon of Essex Junction is legislative chairwoman of the League of Women Voters of Vermont.