Sex, Lies, and Audiotape: Why We Need Instant Runoff Voting

Posted on 14 December 2009

Printer-friendly versionPrinter-friendly versionSend by emailSend by email

By Gautam Dutta and Amanda Fowler
New America Foundation

When a lawmaker brags about his sexual exploits, taxpayers beware:  it might end up costing us millions of tax dollars.

The trouble began last summer, when then-Assemblymember Mike Duvall (R-Yorba Linda) boasted about his extramarital affairs to a colleague – over a live microphone.  In the outcry that followed, Duvall was forced to resign his Orange County seat – setting the stage for a costly election to replace him.  The total cost of this “special” election?  Nearly $2 million.

Last month, barely one-sixth of voters went to the polls to decide Duvall’s successor.  But even though taxpayers have already shelled out close to $750,000 for this election, the Duvall saga is not yet over.  Instead, because no one received a majority (50 percent plus 1), the race is headed for an equally expensive runoff.

Making matters worse, the January 12 runoff will be no more than an expensive formality.  Even though Republican candidates won a whopping 70 percent of the vote, the top votegetter from each party will now advance to the January 12 runoff: Republican Chris Norby (37 percent of the vote), Democrat John MacMurry (27 percent), and Green Party Candidate Jane Rands (2.8 percent). In this GOP-heavy district, it doesn’t take a political scientist to figure out who's the odds-on favorite to win.

Unfortunately, this scenario is nothing new. Traditionally, California special runoff elections are marked by high costs, low voter turnout, and high margins of victory.  And because of our perfunctory runoff system, residents of the 72nd District must wait another two months to move past the Duvall controversy and, more importantly, regain representation in the State Assembly.

Between scandals and politicians’ office-hopping, special elections cannot be avoided. Fortunately, there’s a proven, practical way to rescue hapless voters and taxpayers:  Instant Runoff Voting (IRV).

In a nutshell, IRV does away with costly runoff elections, by allowing voters to elect a majority winner in one single election. With IRV, voters rank their choices (1, 2, 3), and their rankings are then used to determine the majority winner.

IRV will relieve voter fatigue, save taxpayer dollars, and shave the amount of time people must go without representation.  But just as important, IRV will also make our leaders more accountable:  by encouraging them to run cleaner, more issue-based campaigns.

Anyone who followed this election knows that negative campaigning quickly became the candidates’ main strategy. Mailers from the top GOP candidates focused more on political and sexual improprieties than on the issues actually affecting voters. This strategy, where candidates try to drive their opponents’ voters away from the polls, has become all too common.

But with IRV, candidates have a strong incentive not to launch personal attacks. Indeed, if IRV had been used last week, Chris Norby would have been less inclined to attack fellow Republican Linda Ackerman, since he would have been vying for the second-choice rankings of her supporters. IRV rewards candidates who reach out to the entire electorate – a positive change that will help elected officials regain public trust and respect.

A growing number of cities are turning to IRV, which is used widely across the country and around the world.  In recent years, Oakland, San Francisco, Memphis, Minneapolis, St. Paul, and Santa Fe have all adopted IRV, and the cities of Los Angeles and Long Beach are seriously considering it.  What’s more, IRV is currently used in Ireland, London, Australia, and New Zealand.

To no surprise, IRV has already attracted broad, bipartisan support:  Barack Obama, John McCain, Los Angeles League of Women Voters, Los Angeles Area Chamber of Commerce, State Controller John Chiang, Secretary of State Debra Bowen, Assemblymember Ted Lieu, California Common Cause, Southwest Voter Registration Education Project, Asian American Action Fund, former Congressmember Tom Campbell, and many others.

A commonsense solution to a persistent problem, IRV will save money and make our democracy stronger.  It’s time for California to stop settling for “politics as usual”.  It’s time to adopt IRV for our special elections.


Gautam Dutta is Political Reform Deputy Director and Amanda Fowler is the Political Reform Associate for the New America Foundation, a nonprofit, nonpartisan public policy institute that invests in new thinkers and new ideas to address the next generation of challenges facing the United States.

IRV is neither proven nor practical. IRV isn't used in all that many places, and where it is used it's neither cheap, easy nor does it make our democracy stronger.

IRV isn't used in many places, and several places that have used it decided either not to use it anymore or get rid of it altogether. Pierce County Wash voted overwhelmingly to dump IRV after one usage - after it failed to deliver a majority winner in 2 out of 3 races where it was used to determine a winner. Those costs have to be paid each and every time IRV is used, versus the costs of rarely used runoff elections. Other states have studied the costs of IRV vs. the problems IRV is trying to solve, and found that IRV just costs too much!

Aspen CO voters recently decided to take a look at IRV again and possibly dump it after many post elections came up about the accuracy and simplicity of an election that was so complicated to administer that they had to hire a private company to do IRV - and they botched it. Voters in Burlington VT and San Francisco CA are also thinking about dumping IRV.

And IRV was used for the first time in Minneapolis, where they had the lowest voter turnout in 100 years! So how does that improve or help democracy when a voting method is so controversial or complicated that voters don't come to the polls?

Many real world studies of IRV elections do show that it's not that easy for voters to understand. These are exit polls and peer-reviewed studies, not informal exit polls done by IRV advocates (who often provide the "education" for voters before they enter the polls). Actual observations by verified voting experts have shown it to be very difficult to tabulate - so how can regular people trust that IRV works as promised?

I observed an IRV tabulation process that deliver an election "win" to a candidate who got 1401 out of 3022 votes - in an election where we were promised IRV ensured a 50% plus one vote majority winner in a single election. And in that election, a traditional runoff stood every chance of electing a different candidate. Even the person who won North Carolina's first and only IRV tabulated election doesn't like IRV and knows he didn't get the majority of votes.

Where IRV has been used, it RARELY ever delivers a true majority winner in a single election - it pretty much just gives the election to the plurality winner in the first round. A real top-two runoff election gives the runoff to the second-place finisher in the general election 33% of the time. Real runoffs do that because they give voters a real chance to chose between two candidates. They can wait and see who the eliminated candidates endorse - something that doesn't usually happen with IRV.

And we need to really take a look at these claims of who endorses IRV. A MN court recently determined that such claims made in support of an IRV initiative were made in violation of MN law. IRV supporters claims that a candidate or other person might have supported IRV in principle without really studying it, or that there support in principle for IRV equals support for a particular IRV initiative were found to be false.

Before California voters consider IRV for any other elections, they really owe it to themselves to take a look at BOTH sides of IRV and find out that it really doesn't deliver on promises made by IRV advocates. Those IRV advocates have admitted in writing that they are pushing IRV to move our government to a system of proportional representation and away from our current system of directly electing our representatives.

I think direct election is more democratic than electing parties that appoint our representatives. I think that party primaries and/or traditional runoff elections are easier, cheaper, and more democratic than IRV.

Howdy Joyce! :) I see your view is still sadly tainted by falsely placing verifiable voting and IRV in opposition to each other. No need for that. Those of us who have studied the Pierce County WA example can see the failure of IRV there has everything to do with the incompetence of their board of elections and nothing to do with any problems inherent in IRV. Your MN example is quite inaccurate. The only claims made against the pro-IRV campaign there were over who had endorsed it, not about anything factual regarding IRV itself. Many exit polls in places where IRV has been used show voters found it easy to understand and liked it just fine (references can be found at Ballot Access News). I could debunk the rest of your examples as well.

How on earth is less than 10% participation in elections good for democracy? Here in NC we often have runoff elections which draw 6-9% of the voters and are extremely expensive. Just last year, we had a statewide runoff for Commissioner of Labor. A grand total of 63,910 people showed up for that, out of about 4 million eligible voters, for a whopping 1.5% turn out. According to the Independent of Durham, "the cost to taxpayers was about $4 million. Given the tiny turnout, it amounted to about $50 a vote." That's simply indefensible.

Elections need to be decided by the largest possible pool of voters. Runoffs in direct elections utterly fail on that score. IRV solves that problem.

This is Chris Telesca - not Joyce McCloy. I have witnessed first hand the many inaccurate claims made by IRV supporters first hand.

Those exit polls you might be thinking of for North Carolina - are you referring to the one where IRV advocate and FairVote employee Dianne Russell admitted in writing to faking a southern accent when she both did voter education and exit polling in the 2007 IRV pilot?

No - you would have a hard time debunking any of my examples. But IRV is more expensive to run than rarely used separate runoff elections. Data for costs per registered voter for IRV vs. traditional runoff elections show that IRV costs more from the get-go and will never break even.

The example of the 2008 Democratic Party Commissioner of Labor race is a good one. Based on the rather conservative costs per registered voter for doing IRV determined by the Maryland Legislative Research department (when studying whether or not to implement IRV in 2008 and 2006) found the cost per registered voter to be something like $3.10 to $3.50 for starters and 50 cents per voter each year after that for voter education (a figure I found laughably low). For NC's 6 million voters, that would be a nearly $20 million start up cost and $3 million per year after that. And in this Democratic primary - there were not 4 million voters eligible to vote in May 2008 - get your facts correct.

All that to save NC 4 million dollars ever 4 years? Sorry - this is one instance where IRV math doesn't work out.

And you can't be intellectually honest and use a number determined by $$ per vote. You should be working to increase interest to increase turnout - not use funny math and complicated and confusing tabulation procedures to merely get a slightly larger plurality win.

And your examples of the tiny turnout leave much unsaid. For example - the turnout was very low in those precincts and counties where there was only that one race on the ballot - and interestingly enough, Mary Fant Donnan won in this precincts. But in the precincts where there were other races on the runoff ballot, turnout was much higher and John Brooks won those precincts. In fact, it has been shown that the very things we do to decrease the need for runoff elections also drive down turnout for those few runoff elections we do have.

But what it comes right down to is which is more democratic - traditional runoff elections or IRV? If you look at how IRV does nothing but deliver the same plurality winner with s slightly larger number of votes the majority of the time, at a much higher cost in dollars and in public confidence in elections, IRV costs more. But in traditional runoff elections, voters give more votes to the person who came in second in the initial election 33% of the time - a % that does not happen in IRV. So if all IRV does is give the illusion of a majority win, and does not allow for voters to consider different information AFTER the first round of the election - it can hardly be said that IRV is more democratic. That's why so many communities that used IRV one time have ditched it (like Cary NC, Pierce County, and possibly Aspen) or are thinking about doing so.

And no matter what kind of convoluted math you want to use - 1401 out of 3022 votes in the Cary 2007 District B race was not a 50% plus one vote majority win. That is just one more promise IRV advocates make that they cannot keep.

NO one has ever asked those people who claim on exit polls that they like IRV and find it easy to understand have ever taken those same people back to the Board of Elections and had them watch how complicated it is to do an IRV tabulation vs. a straight counting of votes on ballots for traditional elections. At the 2007 Cary IRV tabulations, many observers - including board members of the NC Center for Voter Education and a Cary Town Council person who in her real job does governance issues and who voted FOR IRV saw just how complicated and confusing the IRV tabulation process really is. We saw the Wake BOE not be able to follow simple procedures, and make mistakes that should not have been made. And after a secret vote recount, they came up with different results than those witnessed the day before.

Do you really think the people that Dianne Russell questioned with a fake southern accent know how complicated IRV is on the back end? If they had any idea, they - like the rest of the observers - would opt for traditional elections and runoffs. Which is exactly what Cary did in 2009 when they were asked to be lab rats once again for IRV. They said "no".

Hendersonville took part in their second IRV elections. Just like before in 2007, they didn't need to do IRV tabulation because winners were determined in the first round. In other words - IRV was money wasted a second time. But had they needed to do IRV tabulation to determine a winner, they would have gotten in to trouble. The IRV tabulation procedures for their DRE touchscreen machines were kept secret from the public so observers couldn't follow along to see if the BOE was doing the job right - so much for open government and democracy when IRV comes to town! If they would have had to do a hand to eye recount of those IRV ballots from the thermal paper trail (that fails to print out 9% of the time) they would have been so screwed!

So please sell IRV selective wonderfulness somewhere else - North Carolina knows that IRV doesn't deliver on the promises made. That's why only ONE out of over 530 municipalities wanted to take part in the pilot for 2009 - down from 2 in 2007. What does that tell you?

And I notice that you don't disagree that the judge in MN found that the Better Ballot campaign violated election law regarding "celebrity" endorsements IRV - and didn't really care that they were knowingly violating the law? Do you think that those false endorsements played a role in getting people in St. Paul voters to narrowly approve IRV? And that if voters had known the truth, that perhaps they might not have approved IRV in the first place? Or doesn't the truth and obeying the law matter when something holy and sacred like IRV is involved?

That would be like saying that Chris Telesca endorses IRV today because he had an open mind about IRV and for two weeks in 2002 or so thought it might be a good idea. Then you fail to tell people the truth - that Chris Telesca took a good long hard look at all those claims made by Rob Richie and others about IRV and found them wanting. So you need to clarity ANY claims of endorsements by Obama and others about IRV. And I would love to ask Obama and McCain if they have ever actually voted in an IRV election, or ever counted votes in an IRV election? If the answer is no, and if they had never actually studied an IRV election like the one in Cary from top to bottom or read the SF Grand Jury Report from last year - I'd say they might want to withhold their endorsement of IRV.

Chris Telesca

Sorry to confuse you with Joyce. Actually I was referring to a study done by the NC State Board of Elections and reported by Dr. Michael Cobb of NC State

Your cherry-picking and name-calling pretty much make all my counterarguments for me. I'm perfectly willing to discuss the pros and cons rationally. As for the MN example, I won't say supporters of IRV there are saints by any means, but simply that your charges have absolutely nothing to do with the pros or cons of IRV itself.

The study you refer to was not done by the NC State Board of Elections. The study was done by volunteers and paid IRV activists from DemocracyNC and from FairVote. Dianne Russell, former director of IRV America for FairVote, was down in Cary not only providing voter education as voters went into the polls, but asking them about IRV on the way out.

I don't know what you mean by cherry-picking and name-calling. You asked for examples, and I provided them. Who did I call names? I made reference to a written admission by Dianne Russell that she deviated from the instructions she was given by the Wake County Board of Elections on how to provide voter education on the new IRV voter method in order to get more positive results in the exit poll she was also taking at the same sites. She also admitted faking a southern accent when she did exit interviews. How is that name calling?

I have discussed IRV cons (and the few pros) rationally - but the cons outweigh the few pros. But I have to ask - why do any IRV advocates have to exaggerate the benefits or even the supporters of IRV to convince people of the benefits? The folks in MN not only were found guilty of breaking state election laws by making claims that people at one time supported IRV - they were supposed to get written statements from the people they quoted and they were told that many times - but they made a point out of not following the law. What does that tell you if you have to break an election law to get people to cast a vote for you or for your particular election reform?

Why do people have to lie or break the law to win an election? That isn't the sort of position I expect to see election reform advocates taking.

I think the reason why the law requires such written permissions is because people can and do change their minds about IRV and other matters. Just because you were quoted supporting a particular position in 2003 doesn't mean you support it now. Barack Obama is a perfect case in point - he was on record as supporting "Single Payer" for health care reform. He doesn't support it now. Are you telling the truth if you were to post that he supports single payer now because he supported it back in 2003? Or are you being intellectually honest when you state that he supported it in 2003 that somehow infers he supports it now?

IRV would vastly improve elections in California.

The crazy debate here about North Carolina isn’t particularly helpful. It makes me doubt SAS, the statistical analysis software from Cary, however. The 3022 “votes” were almost certainly “ballots”, some of which showed no vote for any candidate still in the election during runoff rounds (after last-place candidates were eliminated), and ballots without valid votes shouldn’t be counted as votes.

Anyway, I’ll be in (Western) North Carolina for Xmas and would be happy to meet in Asheville with doubters and reporters to learn about and address North Carolina concerns.

SAS software wasn't used to tabulate the IRV ballots - it was all done by hand after the first column votes showed no majority winner.

The 3022 "votes" were votes counted in the 1st round. The SBOE won't share the raw voting data with verified voting advocates so we can't see exactly how well or poorly voters understood IRV - we only have to take the word of the IRV advocates who came (with fake accents) to educate voters about IRV and to do exit polls with them.

Basically, we were promised that IRV would ensure a majority 50% plus one vote winner in a single election. There was nothing said about how this would be a "preferential majority" with s shifting threshold based on how many voters cast votes in the 2nd or 3rd column. That's sort of like ENRON vote counting, isn't it? Either the threshold stays the same throughout the entire vote counting process, or all IRV does is give you larger plurality wins - which are not the same thing as a true majority win. In fact, IRV has a big failure that is being called "majority failure". Votes should be counted at votes - and you can't start lowering the threshold be removing votes from the total number of votes counted for a candidate in the first round.

IF you are an LA voter who hasn't yet voted with IRV - what possible use could you be to voters or reporters in Western NC?