Instant Runoff Voting is the target of false and misleading attacks in Saint Paul. Perhaps the most egregious alleges that IRV would be too difficult for voters to understand. The implication is that Saint Paul voters aren’t as smart as Minneapolis voters or those elsewhere using IRV.
The weaknesses of our current two-step (primary-general) municipal elections and plurality state partisan elections have become clear in recent years:
Do we really want to continue with a system that draws just 7 percent of the voters to polls for the primary (as we saw in the last St. Paul primary election)? Do we really want to continue with a system in which as few as 37 percent of the voters can elect the winner (the last time a Minnesota gubernatorial candidate received a majority of support was in 1994)?
While not perfect – no voting system is – Instant Runoff Voting addresses these problems in significant ways and produce results that better reflect the will of the voters. Equally importantly, it is a proven and implementable system that has proven easier for voters than two-round runoff systems.
Some people don’t like IRV because it changes the electoral game they’re used to playing. In their efforts to hold on to their advantage, they are using misinformation to scare voters. This backgrounder exposes the fallacies that underpin those efforts.
Reality: The vast majority of voters who use IRV easily understand it – and they like it.
Opponents of IRV try to make it sound complicated. It’s not.
Whether voters fully understand all of the
details of how their ranked ballot is tallied, the record shows repeatedly
that voters have no difficulty using the ranked ballots effectively. For example, fully 99.9% of the IRV ballots cast in
the Burlington, VT elections have been valid ballots. In most cities using IRV, more than 99 percent of the voters properly complete their ranked ballots.
| Election | IRV Valid Ballot Rates |
| San Francisco, 2004 (across all races) | 99.4 percent |
| Burlington, 2006 | 99.9 percent |
| Cambridge, 2007 | 99.4 percent |
| Takoma Park, 2007 | 99.9 percent |
| Aspen, 2009 | 100 percent in mayor's race; <1% in the council race (1) |
(1) More ballots had errors among absentee voters, a difference which can easily be addressed with additional voter education. In most jurisdictions, ballot error rates are low regardless of the amount of machine error
notification provided to voters at the polls. Some cities use equipment
at the polls, some don't. More than a third of voters in San Francisco, for example, now vote by mail, but error rates in IRV races remain extremely low.
Strong
voter education and good ballot design are the most important means of
minimizing voter error. These standards have been implemented across the United States and exit polls clearly show that voters overwhelmingly understand IRV and prefer it over the old systems.
| First-time IRV voters who understand IRV | |
| San Francisco (CA) | 86 percent |
| Burlington (VT) | 89 percent |
| Takoma Park (MD) | 88 percent |
| Cary (NC) | 95 percent |
| Hedersonville (NC) | 86percent |
In a 2008 survey, 60 percent of respondents in Cary, NC (not all of whom were voters and familiar with the system) indicated they understood IRV and 69 percent said they preferred it over the old runoff system.
African Americans (62.5 percent) and seniors aged 65 and older (66.6 percent) were the most likely to understand IRV. And they were the most likely to say they support the city using it: 74 percent of African Americans and 79.4 percent of seniors. Lower income groups were also more likely to support it, with the group most likely to support IRV earning between $20,000-$30,000.
Every exit poll taken during the first-time use of IRV in US cities are clear in showing that voters overwhelmingly prefer IRV over the old system.
| Favor IRV to city’s prior system | |
| San Francisco | 82 percent |
| Burlington | 78 percent |
| Takoma Park | 89 percent |
| Cary | 72 percent |
| Hendersonville | 71 percent |
While a new voting method may be unfamiliar to some voters, the record – as opposed to opponents’ rhetoric – shows that the change need not translate into confusion and that there’s no valid reason to doubt Saint Paul voters’ ability to adapt to a new way of voting. The evidence very clearly shows that the vast majority of voters in cities using IRV understand how to use it.
Sources:
Reality: One election is better than two.
One election gives voters more choice and increases voter participation
Some IRV opponents argue that voting is more difficult when voters are offered more than two candidates two chose among. They also suggest that voters who don't show up for primaries prefer that the primary field be vetted by a small group of “informed” voters before November. Or, they assert that primaries give voters time to re-evaluate the top two choices on the November ballot.
But none of these theories hold up under even the most casual examination. None of the exit polls conducted in cities using IRV show that voters don’t like choice or that they find it confusing or that it scares them away from voting. In fact, the studies show just the opposite: that most voters understand and like IRV, and they tend to rank two or more candidates on the ballot.
Under the current system, voters have the option to get to know as many or as few candidates as they wish in order to decide which ONE candidate they will vote for. It is no different under IRV, except that voters are able to rank more than one choice.
These arguments also gloss over a serious problem with the current system: only a small number of voters are fully participating in the current primary-and-general-election system. The majority of voters don't and, as a result, elections are not representative of the community at large. The Saint Paul voter turnout data illustrates how this problem has grown since 1993:
|
|
1993 |
1995 |
1997 |
1999 |
2001 |
2003 |
2005 |
2007 |
|
Primary turnout |
35,883 |
18,550 |
31,668 |
16,928 |
37,994 |
19,226 |
25,303 |
5,606 |
|
General turnout |
63,915 |
43,690 |
61,362 |
76,326 |
59,864 |
32,652 |
59,509 |
30,620 |
|
Primary/General |
56% |
42% |
52% |
22% |
63% |
59% |
43% |
18% |
These numbers demonstrate that primaries don’t foster greater voter participation. While some may prefer the winnowing that a primary provides, the barriers to participation significantly outweigh any advantages. In addition, primaries are costly and – by prematurely narrowing the field of candidates – needlessly truncate the discussion of issues of concern to voters and become a recipe for negative campaigning.
Opponents also make a three-cushion argument that primaries are necessary to winnow a candidate field to a “manageable” size for debates and candidate forums. In support of this assertion, they point to the 2001 16-candidate mayoral race in Saint Paul but, of course, this number of candidates is the exception not the norm. In the past four election cycles, city council races in Saint Paul (and Minneapolis) typically had 3 to 6 candidates. The 2005 Saint Paul mayoral primary was a contest among 8 candidates while the 2009 primary was a contest among 4 candidates.
One election leads to less expensive elections and campaigns
Holding two elections also costs a lot more, both for taxpayers and for candidates.
Reduced election costs
Reduced campaign costs
There is also growing evidence that IRV helps reduce the costs of campaigns and mitigates the disparity in fundraising among candidates. Using a two election system gives an added advantage to candidates with ready access to big money. IRV can reduce the influence of big money in campaigns as indicated in recent IRV mayoral elections in Burlington, VT and Aspen, CO where the winners were significantly outspent by losing candidates.
These experiences show that not only does IRV reduce campaign spending by eliminating the cost of a second election it reduces the need to spend money on negative campaign ads and has the ability to trump campaign spending as a measure for winning campaigns.
Reality: IRV significantly increases voter participation.
IRV has the immediate and desirable affect of increasing voter participation when it replaces two-round runoff systems with single elections.
In San Francisco, the increase in effective voter participation has been significant, due to the elimination of low turnout runoffs. Under the former runoff system, the November voters would narrow the field and then a much smaller number of voters would return in December to make the final decision. Under IRV, all voters help determine the winners in the decisive November election. With much greater participation in the November elections, this means that more voters participated in the final round of the IRV election than in the December runoff. For example, the runoff turnout in 2000 ranged from 53 to 64 percent of the November turnout. In 2004, the final round of IRV turnout ranged from 75 to 90 percent of first round turnout, meaning winning candidates received more votes under IRV than the former runoff system (http://www.fairvote.org/sf/SF_EvaluationRCVsuccess.pdf)
As an example in one Supervisory District: The last time a December runoff was conducted in this race in 2000, voter turnout in the December election was 59 percent of the turnout in the general election a month earlier. Under the first IRV contest in this District in 2006, 88 percent of the people who voted in that race particpated in the final and decisive round.
And, the most dramatic rise in participation has been in predominantly ethnic minority neighborhoods. Before IRV, San Francisco’s most diverse neighborhoods had the lowest turnout rates in the runoffs. Under IRV, voter participation in the city’s six most socio-economically diverse neighborhoods has increased the most dramatically--by up to 300 percent in some districts.
Click here for more information.
In Burlington (VT), both the 2006 and 2009 mayoral IRV elections saw both higher turnout than the two previous non-IRV mayoral elections as well as significantly higher effective voter participation rates. Burlington uses IRV for its mayoral elections and eliminates the need for a separate runoff if no candidate receives sufficient (40 percent) first-choice votes. In the 2009 election, roughly 7 percent of voters did not rank either of the two finalists and, thus, effectively “sat out” the final round of the “instant” runoff in the mayor’s race. This means that the effective participation rate in the election was 93.3 percent of voters in the final round. This compares to a drop off of nearly 50 percent of voters that went to a runoff a few weeks later to decide the outcome in the Ward 7 council race as IRV is only used for mayoral contests.
Click here for more information.
In Aspen (CO), a record 2,544
voters turned out for the first-time use of IRV elections in 2009, electing
winners in mayoral and council races in a single decisive election. In this race, 95.5 percent of voters cast a vote in the final round of the IRV election This 4.5 percent fall-off from the initial round compares to a fall-off rate
of 56 percent in the 2005 runoff election, before IRV was adopted. This year, just 42 percent of general election voters retruned for the runoff in one of the city council races, which are not conducted using IRV.
Aspen began using separate runoffs in 2001 to ensure majority winner elections but, like in other cities with runoffs, the city experienced expensive and low turnout runoffs and switched to IRV for the mayoral elections. Under the consolidated runoff system, a significantly more voters help elect the city’s representatives.
Click here for more information.
The same is true, but in reverse, for municipal top-two primary systems like those used in Pierce County, Minneapolis and Saint Paul. Voter participation is very low and unrepresentative in primary elections (as low as 5 to 7 percent in Saint Paul) in comparison to the general election. IRV allows the greatest and most diverse number of voters to choose among the greatest number of candidates in the general election.
In Pierce County (WA), a full 96 percent of voters had their vote counted in the final round of the 2008 County Executive race, and 90 percent of voters had their vote counted in the County Council race. Compare these participation rates to the 23.4 percent of voters who turned out in the recent 2009 primary elections for county races that are not yet covered under IRV.
Click here for more information.
The reason for the low turnout primaries? Because there's almost no attention paid to them–-the media hardly cover them, there's no public notice and candidates tend to only target "regular" primary voters. So the vast majority of voters aren’t even aware a primary election is underway. Primary voters are not representative of the larger community, especially communities of color who are disproportionately underrepresented in primaries. Candidates who can't mobilize a base of new supporters during a low turnout primary have a difficult time making it through this step in the process.
Bottom line: Significantely more voters participate in November general elections and, under IRV, more voters elect the winners.
IRV does not eliminate negative campaigning altogether, but based on the experience in cities using IRV to date in San Francisco, Burlington, Aspen and now Minneapolis, there is a great deal of anecdotal evidence that IRV leads to more civil campaign tactics and substantive discourse.
This is because under
IRV, more candidates compete in the decisive general election, reducing
the number of head-to-head campaigns where negative campaigning can be
effective in winning votes.
Under IRV in multiple-candidate races, candidates have a new incentive to both differentiate themselves from other candidates while also appealing to supporters of other candidates. Negative attacks can be counter-productive and penalize offending candidates at the polls. For this reason, IRV encourages more positive, substantive campaigns in which candidates try to earn first-choice support while remaining attractive to other candidates’ supporters.
Here has been the experience to date in cities using IRV:
San Francisco
The change in the nature of campaigning during the first use of IRV elections in San Francisco in 2004 was noticeable enough to merit a headline in the New York Times: "New Runoff System in San Francisco Has the Rival Candidates Cooperating". More recently, in a 2008 race, the editor of the San Francisco Bay Guardian wrote that in the highly contested open seat race for the Board of Supervisors:
[The winner] will probably be the one who gets the most second-place [rankings]. So it's in everyone's interest not to go negative. If Sanchez, say, started to attack Quezada, the Quezada backers would get mad and leave Sanchez off their ballots — and that would hurt Sanchez when the second-place votes are counted. So everyone has been pretty well behaved in [District 9]. I've heard a few whispers here and there, and a few people have tossed off a few nasty comments, but overall the candidates and their supporters recognize that it's better to stay positive.
Burlington
A similar dynamic occurred in the highly competitive 2009 mayoral race in Burlington.
The Burlington election certainly was a model of clean, open debate. Democratic city councilor Bill Keogh, a past skeptic of IRV, told the local daily newspaper, “This campaign has been very, very good.” The four leading candidates had been “as forthright as they can be with their views,” he said. “This is the most respectful and informative campaign in Burlington in a long time.”
Aspen
The Aspen Times commented after the 2009 Aspen city election that: "[We] have been impressed with the professionalism displayed…[C]andidates have treated each other respectfully during these stressful times.”
Minneapolis
In Minneapolis, the usually highly contentious 5th ward council race has taken on a very different tone this year. Candidates are engaging in less personal attacks and staying more focused on the issues. According to the the Minnesota Indendent:
Ward Five has a history of poisonous politics. The contest four years ago
got particularly acrimonious, with Samuels being parodied as an Uncle Tom. And
there have been some signs that the race will not be without fireworks this
year.
Despite some heated rhetoric, the incumbent believes that the
specter of instant-runoff voting, in which candidates will rank their favored
candidates, has had a cleansing affect on the race so far.
“I think it’s
much less stressful this year,” he says. “There’s a lot less personal attacking
going on.”
In Ward 6, a similar dynamic is taking place:
Given the slim margin of victory four years ago, it’s not particularly
surprising that numerous candidates are eying Lilligren’s post this election
cycle. The Democrat has attracted five challengers more than any other
incumbent in the city. The introduction of instant-runoff voting, in which
voters will rank their preferred candidate, further complicates
matters.
Lilligren was a strong supporter of adopting ranked-choice
voting. He’s visited Cambridge, Mass., where they’ve long utilized such a
system, and believes it encourages more constructive, issue-oriented
campaigns.
“It’s a much more collegial, positive campaign environment,”
he says. “I just think it’s better for the city if council candidates aren’t
clawing each other apart.”
“Three of Lilligrens challengers Andy Exley,
Laura Jean and Michael Tupper have come up with a unique strategy: they’re
campaigning together. The three candidates have been holding shared
meet-and-greet events and are refraining from criticizing each other.
Also, see KSTP news report on the Ward 6 race.
The purpose of IRV is to consolidate a two-election runoff into a more efficient, cheaper and participatory single-election runoff. Some voters will choose to rank one candidate and elect not to participate in the runoff should that candidate not make it through the first round or two. IRV elects the winner of a majority of voters continuing in the final round and does so in the election with the highest and most diverse turnout.
As we saw in the previous page, the fall off in voter participation in the final round of an IRV election is no where near the fall off in participation in a primary or a runoff under a two-election system. Effective voter participation is much higher under IRV.
Some
jurisdictions, such as Minneapolis are
using a system that limits voters to three choices, due to limitations of
current voting equipment. When it comes time to purchase new machines, ballots can be designed to provide voters with the opportunity to rank more than three choices if they wish.
Reality: IRV is counted by machines in most jurisdictions in the US and abroad.
IRV-capable equipment is used across the country – in San Francisco, Pierce County, Cambridge (MA), Burlington (VT) and Aspen (CO) and in Scotland, New Zealand and other countries – to tally ballots.
IRV-capable machines will not be used in Minneapolis
in 2009 because they are not certified for use in Minnesota. Minneapolis anticipates that certified IRV-capable machines
will be available for purchase by the next scheduled municipal election in
2013. According to elections director Joe Mansky, Ramsey County
anticipates buying new machines by 2011 and they will have the opportunity to purchase
IRV-capable machines for this election. The next generation of voting equipment is IRV-capable.
IRV is a relatively simple and doable change that can have a potentially big impact on voter participation, cost of elections and campaigning, the way campaigns are conducted and – ultimately – on representation.
IRV does not “throw out” incumbents as some fear. Nor does it benefit them any more than the current system. Incumbency is the most likely predictor of winning an election under traditional as well as IRV elections. But this doesn’t make the campaigns that challenge the incumbent irrelevant. Candidates often do not win their first race, but the experience and exposure gained from that campaign is critical to the success of a future campaign. Under IRV, all of the candidates benefit from competing for votes in one decisive election and are not weeded out prematurely by a small sliver of voters who turnout for primary elections.
Also, because concerns about "spoiler" candidacies disappear with IRV, worthy candidates who might have refrained from running in a plurality election can offer themselves to voters, and voters can vote for the candidate they think is best without worrying about wasting their vote.
And more important than the candidates, the people win. They vote once, their votes are equal, they get to chose from the larger group instead of a winnowed group from a small electorate, and it saves their city money.
IRV is especially valuable in open seat races where 3 or more candidates often compete without an incumbent advantage. Over time, the composition of an elected position or board can change to better mirror the population it serves. This is being seen in San Francisco now, after six years of IRV elections. The San Francisco Board of Supervisors is more diverse than it has been in previous years: it now has 7 out of 11 persons of color, which is the most ever, and three women, which is more than in prior years. Click here for more information.
Additional factors may contribute to the growing diversity in representation, but IRV is likely a key influence, as was the change from at-large to district-based elections in the 1990s in creating a more diverse board. With regard to political diversity, San Francisco is a predominantly Democratic city so that party affiliation is still strong with 10 Democrats and one Green Party member.
Two other examples demonstrate how IRV impacts representation and boosts the electoral chances for candidates representing communities of color and women.
Finally, in Cambridge, where a multi-seat form of IRV has been used for over half a decade in multi-seat elections, elections have consistently resulted in strong representation of ethnic minorities and women on the city council and school board. Click here for more information.
Challenges to IRV – in the courts or on the ballot – are not new and are mounted because IRV does work. Repeal efforts are typically led by a losing candidate or their supporters who prefer the old status quo. IRV was repealed, for example, in Ann Arbor, Michigan, after it resulted in the election of the city’s first African-American Mayor in 1974. The system worked the way it was supposed to – electing the candidate preferred by the majority of voters. The race unseated an entrenched base of power, which first challenged the system in the courts. When it lost, it sought to repeal it and succeeded in low turnout election.
So, what are the current repeal efforts about?
Cary, North Carolina
Cary was not the site of a repeal effort but rather a decision by the city council to opt out of the pilot program this year due to uncertainty about how the ballots would be tallied. IRV worked well in Cary and is preferred by the voters: the exit poll taken at election time showed that 72 percent of the voters liked IRV and a 2008 town survey affirmed this support, showing that 69 percent of residents favored IRV, and 51 percent strongly favored IRV. Hendersonville joined Cary in being the initial cities to try IRV under North Carolina’s IRV pilot program and is continuing to use it.
Aspen, Colorado
Despite the success of the first IRV election in Aspen, advocates of the old status quo runoff elections – led by a defeated mayoral candidate in the recent election – are seeking to repeal the new system. The city council has voted to put an advisory measure on the ballot about whether to keep IRV. If this measure passes, it will require another vote of the people to revert back to the traditional two-round runoff system.
The opponents are boosted in their efforts by the Minnesota Voters Alliance, the organization that failed in its legal challenge against IRV in Minneapolis when the Minnesota Supreme Court ruled unanimously that IRV is constitutional.
The Aspen IRV election was very successful overall. Voter turnout was high, the cost was low and majority winners were elected without a second runoff election. It was also an example of transparency and verifiability in the way IRV ballots are counted.
While there seems to have been some voter confusion in the novel multiple seat races, this issue can be easily addressed through voter education and ballot design for future elections.
Burlington, Vermont
As in Ann Arbor in 1974, the defeated candidate for mayor who lost in the 2009 mayoral election has been the inspiration for efforts to repeal IRV there. Read full story at http://www.fairvote.org/lessons-from-burlington.
Pierce County, Washington
In an editorial, the Seattle Times called the IRV repeal effort in Pierce County a "reflex action by politicians who didn't like it responding to voters who weren't used to it." It went on to say that opponents "can't say it didn't work" and that it's a "new and promising system, and it has hardly had a chance to prove itself”.
Political opponents in Pierce County are working to overturn IRV despite its overall successful first-time used.
In a recent commentary, IRV expert Professor Richard Connolly, observed: "Whose interests will be served if they succeed in fooling voters into overturning our voting reform? Behind all the false rhetoric is their fear of an unstated truth: IRV elections are simply too fair and too competitive for the incumbents in the two parties?"
The 2008 implementation of IRV in Pierce County
was more challenging than some of the other implementations due to the overall size of
the election during a presidential election year. Despite these challenges, however, the elections were successful and the system
had the potential to save the county a significant amount of money once the start-up
costs are recouped and if it is expanded to other county elections. Click here to see the independent evaluation report.
Those working to repeal the system – including the newly elected county executive who was responsible for implementing the new system as county auditor – and who was elected under IRV – point to the cost, delays and voter confusion caused by IRV.
Proponents of IRV – in particular the local League of Women Voters – have refuted these claims and point to the various ways that the election officials have incorrectly inflated the cost of IRV implementation, including: 1) attributing the upfront investment in voting equipment to the first year of use instead of properly amortizing this cost over the life of the equipment; 2) assigning costs to IRV that would have occurred without the new system; and 3) not factoring in future savings gained from eliminated primary elections.
There were delays in counting ballots and reporting results in the 2008 election, but these had nothing to do with IRV. They were due primarily to the large volume of mailed in ballots, which were still arriving days following the election because Pierce County, where most voters vote by mail, allows ballots to be postmarked on election day.
IRV is not the cure-all for our election and campaign ills. But what IRV has shown to do is effectively and fairly combine two elections into one, elect majority winners in a single election, increase voter participation, provide voters with more choice and reduce negative campaigning.
No voting system is perfect. The question we pose is: can we do better than we do now? The answer is yes. IRV is a tested and accepted system that is in use in countries around the world and now used in eight US cities. More than a dozen US cities are slated to implement IRV in the near future. This is not an unchartered experiment that Saint Paul will be embarking on if the ballot measure is successful. It will be one of many cities and countries, as well as many organizations that use this voting system. And we believe that the city has the capability and competence to implement IRV, and we believe the voters care enough about IRV to learn and use the system.
IRV is legal and constitutional. IRV will eliminate municipal primaries, saving the cities money and increasing voter participation. IRV will result in candidates affirmed by a majority of voters without costly two-round elections. IRV opens the doors of opportunity for more candidates. IRV will make campaigns more civil. And, yes, IRV will change the way our current campaign system works. To some this change may be scary. For those of us who want to see our system improved, it is a welcome opportunity we are proud to promote. It is one of the ways we believe we can enhance the democratic process at the local level to make elections more competitive, participatory and meaningful.