"Now How Did This Guy Get Elected" OR Election Reform and Cary NC'S FUTURE

This is a real life twilight zone scenario of an election reform's unintended results. Thanks to Instant Runoff Voting, the worst possible candidate won this Supervisor's election in San Francisco. Is this a peak into the future for Cary North Carolina, who is "experimenting" with IRV this October? Beware of "election reforms" and "experiments" that may help elect a party switching carpet bagger who next faces FBI charges for accepting kick-backs.

Very interesting... San Francisco's recent Supervisor's election shows how an old fashioned traditional runoff election would have been superior to "high-falutin'" instant runoff or IRV. The city of Cary, North Carolina will be experimenting with IRV in their October municipal election.

John Diaz, reporter for the San Francisco Chronicle asks:

Now, how did this guy get elected? June 10, 2007

...So, how did it come to pass that the city's newest supervisor, Ed Jew,apparently did not even live in the Sunset District and was the choice of just 5,125 (or 26.2 percent) of voters? And the FBI is looking into what this "citizen politician"
was doing with $40,000 in cash from tapioca-shop owners who had sought his help with city permits.

It turns out that both of the voter-installed "reforms" -- district elections, instant-runoff voting -- helped make it possible for the flower-shop operator, who once served as vice chairman of the local Republican Party, to get elected in San Francisco on a "grassroots" campaign.

...Dudum led after two rounds, but lacked a majority. Jew ultimately prevailed in the fourth round.

So here we have a winning candidate, Ed Jew - who 73.8 percent of the voters in District 4 did NOT vote for, who had recently switched parties from Republican to Democrat, and did not even live in the district he was running in.

Because of San Francisco's Instant Runoff Voting system, voters didn't get the chance to compare
Ed Jew to Ron Dudum, a lifelong Democrat and actual resident who was trailing by just 53 votes.

The city went through 4 rounds before this "winner" prevailed.

For the record, very few Cary voters may even know that they are expected to rank their municipal candidates this October. The City Council chose not to have a public hearing on the matter.


update from SF: I Can't Believe I Missed Ed Jew's Surrender

Greg Dewar, political consultant in San Francisco blogged this about their new District Supervisor chosen thanks to IRV:

I Can't Believe I Missed Ed Jew's Surrender In Burlingame...

I can't believe I actually missed Ed Jew's surrender in Burlingame, to John Law the week I'm in Burlingame visiting family.

My mom's place is not too far from Casa De Tapioca, and I'd actually walked by the place earlier today. Just goes to show what you miss when you take a week off and go out of town.

Good luck voters of Cary, good luck.

I'm not a fan of IRV

I read a little about it and honestly don't know that much, but it just doesn't appeal to me for some reason. I'm sure there are success stories to balance this one, but this is a pretty scary scenario.

Robin Hayes lied. Nobody died, but thousands of folks lost their jobs.

Vote Democratic! The ass you save may be your own.

Don't believe everything you read in the newspaper

The premise of this San Francisco Chronicle op-ed is that a two-round runoff would have elected a different candidate. This premise is almost certainly false. It is justified in the article only by (a) runner-up Ron Dudum's assertion to that effect, unsupported by any evidence or reasoning, and (b) equally unproven assertions that information would have emerged during the runoff campaign that didn't emerge the first time around.

There are several problems with this.

(1) Evidence about the (real or alleged) bribery couldn't have emerged because the bribery hadn't happened yet.

(2) Questions about Jew's residency in his district had, in fact, been raised -- when he ran for the same office 4 years earlier. By 2006 it was old news.

(3) If negative information about Jew was available but didn't come out in the first round, that's the fault of the press, not the voting system.

Jew may have been flying under the media radar in the first round. But he was not flying under the voters' radar. He was very popular in his community -- ironically, in view of the fact that he may not actually live there. His popularity was built on constituent service rather than endorsements from power brokers. That's why the press was caught with its pants down when he came in first. It had been listening to the organized interest groups rather than rank and file voters.

Ed Jew was the first choice of only 26% of the voters. But, in a field of four closely matched contenders plus a couple of also-rans, he got more first choice votes than anybody else. Editorial writer John Diaz attempts to imply otherwise, but the fact is that he would have made the runoff.

And then he would have won the runoff. The supporters of the other two major candidates, Jaynry Mak and Doug Chen, preferred Jew over Dudum, almost 60-40. It's that simple.

There are a few situations in which IRV selects a different winner than two-round runoff, but this was not one of them. (As an aside, in those situations, IRV actually does a better job than two-round runoff, although proving that gets a little technical.)

Folks in North Carolina should be paying to close attention to how IRV performs in San Francisco. But they should seek out facts rather than rely on opinions.

Learning about IRV

Start at http://www.fairvote.org/irv, and use the drop-down lists in the left-hand column.

Another place to start is http://www.instantrunoff.com

A word of warning: advocates for several alternative electoral reforms will probably visit this thread soon. Folks who favor a procedure called "range voting" are especially active in the blogosphere right now (although their activities in the real world appear somewhat more limited).

By all means check out range voting, approval voting, the Condorcet principle, the Borda Count, Arrow's (im)possibility theorem, and other esoterica. It's a great intellectual passtime. Just keep in mind what I said about the San Francisco newspaper editorial: you have to evaluate all of the arguments and evidence before you make up your own mind.

the other side of IRV that you won't hear elsewhere

You can go to the IRV proponent website, Fair Vote and they will tell give you lots of pro IRV points.

Or, you can read what others say - political consultants for Greens and Dems in San Francisco, what election officials in San Francisco and other parts of California are saying, what computer scientists and a statistics expert are saying, and you can see why Raleigh and Rocky Mount City Councils turned IRV down, why CALPERS turned it down, and why some say it doesn't meet its political promise.

Download that here, with links to support the comments.

People in Rocky Mount complained that the presentation they received was by proponents only, and that it only gave the pros of IRV, not the cons. Raleigh City Council members asked their consituents and also voting rights organizations before making a decision. In all cases but Cary, the public was part of the decision.
Meanwhile, remember that Fair Vote said that ranked choice voting went well in Scotland for its first try in May, yet 187,000 ballots were spoiled.

Why were 178,000 ballots rejected?

The statement about 178,000 spoiled ballots is a real howler. Please, please, please, follow that link. The article there will tell you that there were two separate elections (local councils and regional parliament) on the same day. In the local council elections, 38,000 ballots were rejected. This was the election using ranked choice voting (called STV in Scotland). The other 140,000 rejected ballots were cast in the parliamentary election, which used a different voting method.

I'm afraid that this attitude toward the facts is typical of NCVoter, and is reflected in his/her/their "report" on San Francisco, which is linked twice in the post above.

Read everything you can get your hands on about IRV. But always evaluate the evidence and reasoning being presented to you.

someone from the SBOE observed that the high ballot spoilage...

...came from mixing voting methods on the same ballot and in the same election.

Johnnie McLean of the NC SBOE wrote that you shouldn't mix voting methods on the same ballot or ideally even in the same election - and you should hold the RCV elections on a different day. Which sort of eliminates the claim of savings from having one election instead of two as FairVote claims.

Dr. Rebecca Mercuri, Statistician, SanDiego task force, CALPERS,

Here is some information you won't find at Fair Vote

What is IRV?

National Advocate for IRV is Fair Vote, known as Center for Voting and Democracy

Raleigh, North Carolina City Counsel - why it tabled IRV.

Rocky Mount, North Carolina City Counsel opposition to IRV

Chuck Herrin, IT Certification specialist and White Hat Hacker – on auditability and security

Dr, Rebecca Mercuri, e-voting expert– potential "gaming" of the ballot set

Kathy Dopp, MS mathematics, President of US Count Votes - more mathematical analysis is needed

San Francisco Director of Elections Jim Arntz on costs and labor for voter education and for recounts

Tony Bernhard, former Yolo County clerk/recorder about fairness of the system and whether the system is needed

Greg Dewar, Political Consultant in San Francisco – how IRV has affected elections in San Francisco

John Dunbar writer for "Beyond Chron" a daily internet newspaper in San Francisco. Instant Runoff Voting Not Meeting Expectations

The San Diego Elections Task Force Report and Decision not to use IRV

CALPERS, the California Public Employees’ Retirement System trustees decision not to use IRV

Assessment of Ranked-Choice Voting in the San Francisco 2004 Election – Pros and Cons,data.

Assessment of Ranked-Choice Voting in the San Francisco 2005 Election - Pros and Cons, data.

Kevin McGuire, an associate professor of political science at UNC Chapel Hill about voter --- participation tied to knowledge of candidates in judicial contests.

NC Coalition for Verified Voting – concerns about impact on election integrity.

NC Law on IRV “Pilot Program” and IRV Impact on North Carolina and Counties
Improving Ballot access as best way to help third parties in North Carolina.

thank you

for the info.

my only worry with the cary elections using IRV is that there wont be enough candidates to actually test the system. unopposed is unopposed. and in a 2 way race its all the same.

Draft Brad Miller-- NC Sen ActBlue

"Keep the Faith"

That's the best reason to be wary,

IRV is only effective there are enough candidates and that's not been our problem here in NC that I can tell. When we get the third party ballot access issues resolved then maybe it would be worthwhile to take on a complication on the ballots.

As it is, IRV only keeps the BOE from having to hold another election in a very close race as I understand it.

Cary elections

are non partisan. So its not a question of access.

Draft Brad Miller-- NC Sen ActBlue

"Keep the Faith"

The SBOE knew in March 2007....

...that IRV poses a risk if used in the May 2008 election. So why is anyone claiming that it would have kept us from having a runoff in June of this year? We should be lucky we didn't use it - we might be looking at a Florida style election meltdown if we had used it.

Candidate turnout

We (advocates for IRV) always make the argument that it improves voter turnout, and we have evidence to support that. I think it also improves candidate turnout. We don't mention that as often, perhaps because it will take a while for the evidence to accumulate.

It takes time for political newcomers to gain experience and overcome the advantages of incumbency. IRV lets candidates test the political waters and gain campaign experience without fear of helping elect an opponent through vote splitting.

See also this brief article on multi-seat elections.

What evidence shows IRV increases voter or candidate turnout?

Either you have evidence of IRV improving voter turnout, or you do not. Saying that it will take a while for evidence to accumulate implies that you don't have that evidence yet in hand.

In San Francisco, turnout is down since the first IRV election in 2004. And costs are way up! After using illegal uncertified software for the first three years, San Francisco was required to hand tabulate all the IRV races - and it took teams of paid workers a month of 16 hour days to do just that!

The Washington Post did an article about the recent IRV elections in Takoma Park MD that complained that they couldn't get enough candidates to run in the races - something IRV was supposed to fix.

And here in Cary, two candidates dropped out of the At-Large council race. One cited IRV as the reason she dropped out and supported the eventual winner. So maybe IRV works to instantly run off candidates before the election?

And one more thing - this whole business about IRV seems to be about being election training wheels for candidates. It doesn't mean that any of those candidates have a better idea.

It also causes much wasting of time for volunteers who have to decide early on which candidate to work for and donate money to. Too many candidates means too little volunteer time for each candidate when it is most important.

And in the general elections which IRV supporters are also trying to use IRV for, more than one candidate from a party on a ballot means that political parties can't put their support behind one candidate. SO the political party that can support a particular candidate in an election has an advantage over a party that can't do that.

73.8 pct preferred someone else, "winner" won 4th round

The question is answered in the OP -

Now, how did this guy get elected?

San Francisco's system of electing supervisors did not require him to win over very many of the city's voters -- and he did not. ...

On election day, 73.8 percent of the voters in District 4 expressed a preference for someone other than Ed Jew. ...

Dudum led after two rounds, but lacked a majority. Jew ultimately prevailed in the fourth round.

Progressive citizens of Minneapolis Minnesota are trying to get rid of IRV in their city, and have formed an alliance.

They state at their website that:

Instant Runoff Voting should be stopped:

IRV is undemocratic in several ways:

It eliminates the primaries which play a vital role in the electoral process; It creates false majorities and suppresses minority viewpoints and its structure makes it susceptible to strategic manipulation.

IRV is impractical because it easily leads to unreasonably large fields of candidates making it nearly impossible for voters to identify them.

IRV is a confusing format that will alienate voters, thus consolidating power for the elite ruling class.

He who decides a case without hearing the other side cannot be considered just.”



why did you use the quote saying the one guy was ahead after round two, without mentioning the fact that the other guy was ahead after round one?

The simple fact is that the loser, Mr Dudum never had a majority of votes. So should the guy who didnt win have been elected over another guy who "didnt win"?

This whole scenario doesnt really scare me at all.

Draft Brad Miller-- NC Sen ActBlue

"Keep the Faith"

San Francisco voters understood IRV /preferred it over old syste

NCVoter gets it very wrong in the points above.

> IRV is a confusing format that will alienate voters, thus consolidating power for the elite ruling class.

First, it is a completely bogus argument to make to say that IRV is a confusing format that will alienate voters.

If NCVoter did his or her homework and looked at the results of actual usage in SF, he/she would have learned that:

1) 87% of voters in San Francisco understood IRV perfectly well or fairly well, with only a very small percent experiencing any lack of understanding.

2) Voters in SF preferred IRV by a ratio of 3-to-1 over their old two-round runoff system

3) Voters in SF thought that IRV was "easy" by almost a 3-to-1 ration vs. those that thought it difficult

4) By a margin of greater than two to one (37% to 15%), voters perceived the Ranked-Choice Voting system as more fair than the runoff system.

5) A majority of voters in SF ranked the maximum number of candidates possible, demonstrating that they were able to use the format effectively.


Nearly $1 Million in Voter Ed, and 700 Voter Outreach Events

Voter education has to be repeated each year – many ballots rejected at polling places, less voters prepared to "rank" their choices in 2nd year of IRV

the proportion of voters expecting to rank choices was lower in 2005 (54%) than in the 2004 election for Board of Supervisors (67%).

San Francisco spent about $1.83 per registered voter on IRV specific voter education, and held over 700 voter outreach events. Even then, and in spite of the fact that IRV had been approved for well over a year, many voters came to the polls not prepared for IRV.

Since it is so close to the election, because IRV is so complex, and no one knows anything about it, Cary should spend at least twice as much on voter education than did San Francisco. Candidates also should provide additional education to their supporters to reduce confusion.

Oct 5, 2004 Ranked-choice voting a matter of picking 1, 2, 3.

Despite outreach efforts, many S.F. residents who came to the polls had not heard of ranked choice voting..

…a number of voters interviewed after handing in their ballot in the basement of City Hall Monday had not heard of ranked-choice voting -- despite the $776,000 spent on education -- and therefore hadn't studied more candidates…."It was pretty much a guess," District 7 voter Mack Shaver said of his second and third picks..

Mailers to educate the voters didn’t reach voters in time

…Part of the problem, Shaver speculated, was that those voting Monday hadn't received the bulk of informational campaign brochures that are bound to show up in mailboxes in coming weeks

IRV 2005 - Voters Ballots Frequently Rejected - & Voters Less Prepared 2nd Year.

“Director Arntz (of San Francisco) said that voters were concerned that their ballots were frequently rejected by the Eagle machines because they did not complete all three columns. This is partially attributed to the Department not having enough funding to conduct pre-election outreach to voters. Community outreach would have enhanced voters awareness of how the ballots needed to be marked in order to be counted efficiently. Therefore, voters were less prepared to vote in this, the City’s second ranked choice election. Also during the development of RCV in San Francisco, the majority of public comments wanted voters to be alerted if all three columns were not marked. People were concerned RCV would not be successful if voters did not have every opportunity to mark all three columns.

…Director Arntz asked the Commission to review the implementation of programming ballot accepting equipment so that voters will be able to make only one selection of a candidate for an office without the tabulating equipment automatically rejecting that ballot as being under-voted, thus requiring Poll Workers to ask the voter if this was, indeed, the voter’s intention, and therefore, perhaps, invading a voter’s privacy. ” Nov 8, 2005

It's not bogus to make the argument that IRV is confusing

There were two types of elections in 2004 in San Francisco. The non-IRV ballots had a normal rate of invalid ballots (spoiled, undervoting and overvoting) - the IRV ballots had a rate 7 times higher.

Asking people if they understand something is much different than designing an experiment where you can test people to see if they really do understand something. The proof is in the invalid ballot rates - 7 times higher than non-IRV ballots.

Preferring something doesn't mean that it's a good thing. Some studies show that people prefer touchscreen voting machines - would you seriously recommend election administrators implement those preferences nationwide?

33% of voters showed up to San Francisco's first IRV election not knowing they would be expected to rank their choices - after that community spent approximately $1.80 per registered voter on voter education. That percentage went up to 45% the second year of IRV. Since those voters didn't know that they would be expected to rank their choices, they either weren't able to rank more choices, or if they did chances are good those were just "donkey" votes cast to fill in the blanks.

How do voters know how "fair" a more complex method of voting is to all people in their community might be? How many people would think that a "literacy test" is fair to administer to voters as long as they think they could pass it?

Here in NC, we already have a form of a literacy test - it's requiring voters to cast a separate vote for president and not let them cast a straight party vote that also includes the presidential race. The statewide level of people who either under or overvote their ballots is right at the national average. But in some counties that have very low literacy rates, the rate of under and over votes is right at the highest in the nation.

So if those people already have a hard time voting for president, you are going to disenfranchise them further by demanding they rank their choices and that their ballots will be exhausted sooner if they don't. And when they don't - it's like decreasing voter turnout for that group of people.

IRV empowers / enfranchises all voters, incl. those not in rulin

IRV is a confusing format that will alienate voters, thus consolidating power for the elite ruling class.

Again, NCVoter tries to mislead us. What a preposterous statement that s/he makes above.

In fact, in San Francisco when they used a two-round runoff system, voter turnout would plummet after the initial election, and that drop in turnout would disproportionately hurt low-income and minority voters.

By using IRV and electing the winners in a single high-turnout election, San Francisco dramatically increased effective voter turnout, especially among those groups that suffered the most under the old system. In an analysis of voter participation for a citywide assessor-recorder race, voter turnout increased by an estimated 2.7 times by using IRV, and over a quadrupling of turnout for some of the poorest and most racially diverse neighborhoods.

So, it can be said that NCVoter is completely wrong, and in fact the opposite is true to what s/he says. IRV helped to empower and enfranchise those NOT in the ruling class by increasing their effective participation at the polls.


"disenfranchised those who are already the most powerless"

Final Count - 187,578 spoiled ballots in Scotland

Rejected ballots

The Holyrood vote used a different system of proportional representation - the additional member system - using two crosses.

The two systems involved the same people voting at the same time and the votes were counted or rejected by the same machines.

The figures suggest that voters seemed to understand the new system for the councils much better than the existing system, for Holyrood.

The new figures show that the total of rejected ballots across both elections now stand at 187,578.

This is more than in the infamous presidential election in Florida when fewer than 180,000 ballots were rejected.

The poorer the voters were, the less educated, the more likely to spoil their ballots

May 7, 2007 Not so much an election as a national humiliationScotland’s voters were treated with arrogance and contemptMelanie Reid
As one man commented caustically on the BBC website, it wasn’t rocket science. No, but neither was it exactly straightforward. In the polling station last Thursday morning, confronted with two ballot papers, three different columns of names and the requirement that I use crosses and numbers in different places, I readily confess that the new Scottish voting system made me pause in my tracks.

....More than 100,000 people – around one in 20 of those who voted – had their ballot papers rejected in the election: a figure so scandalous that analogies with hanging chads don’t really begin to describe it. Their votes were rejected because the forms were too confusing for them (let’s leave aside the tiny minority who spoilt their papers as a form of political protest). What is now crystal clear is that the poorer and more ill-educated the voters were, the more likely they were to put the wrong marks in the wrong places, and unwittingly invalidate their forms.
...Naturally, the reverse holds true. In areas of greater affluence, where people are wealthier, healthier and better educated, the trend was reversed...

...The conclusion is unavoidable: the ballot forms were simply too much for the country’s least favoured citizens.... The ill-designed ballot papers disgracefully disenfranchised those who are already the most powerless and voiceless in society, leading one to suggest that, if the much touted “best small country in the world” aspires to be a model for democracy, then it urgently needs to find a way that its people – all its people – can express their wishes.

.... Bewildering? Imagine how Scotland’s thousands of elderly and barely literate felt.

IRV experience in Scotland -- let's get this right

The refutation of NCVoter's (mis)interpretation of the election in Scotland is contained in the very article quoted from above. "The figures suggest that voters seemed to understand the new system for the councils much better than the existing system, for Holyrood."

In this sentence, the phrase "new system for the councils" refers to IRV. The phrase "the existing system, for Holyrood" refers to a completely different voting method, which they call the "alternative member system" (and Germany and New Zealand call "mixed member proportional"). There's nothing wrong with it in and of itself. But election administrators had introduced a new and, as it turns out, very confusing ballot layout.

I will try to make this completely idiot proof. The confusing ballot layout and at least 140,000 invalid ballots were in an election that was not an IRV election.

There were some other problems with the Scottish elections, and the whole story is a little complicated. (We're not the only country that has problems with voting machines.) But the introduction of IRV for local council elections went very well.

No - the Scottish elections didn't go well

The Election Commission report on the election disaster pretty much stated that the confusion between the two ballot styles at the same time was a cause for the problem.

Clearly, if voters had not used different styles at the same time, they wouldn't have had such problems.

Also, handicapped voters in Scotland claimed STV made it harder for them to understand what was going on.

NC's own Johnnie McLean observed that the way around that mess was to have the two different election styles at two different times - which sort of takes the wind out of the sails that IRV does the work of two elections in one. If you have IRV and non-IRV elections on the same ballot, you are inviting trouble like you had in Scotland. So if you have the IRV elections at a different time than you have the regular elections - you don't save money or time.

increaing "effective voter turnout"?

Isn't that like doing ENRON vote counting?

Who did this analysis you write of? By saying that you elected a candidate in one election vs two (and that last one might have had a lower turnout) and claiming the higher initial turnout is an increase in turnout is ENRON accounting.

2.7 times the voters didn't show up for IRV than for a non-IRV race. And IRV did not result in 4 times the number of voters in the poorest and more racially diverse neighborhoods in 2004's IRV election over an non-IRV election.

In fact, turnout among eligible voters in the 2007 IRV election is down from the same election 4 years earlier that did not use IRV.

What you do in IRV is you deny voters the chance to learn the differences in the top two candidates - something that you can't do in IRV because you might have 12 candidates crowding the stage and not enough time for them to develop their message. A traditional runoff election allows a candidate to talk about the real issues and what differentiates them from the other candidate - which is why Robert's Rules of Order recommends traditional elections and separate runoffs over preferential voting (AKA IRV).

Here in NC, we had two traditional runoff elections won by African American candidates that they both feel they could not have one using IRV. One race in Rocky Mount had greater turnout in the runoff than in the general election. IRV doesn't allow for candidates to generate the kind of heat they could have the potential to do in a traditional runoff election.

Increasing effective participation is not the same thing as actually getting more people out to vote. A real majority is different than a manufactured or "preferential" majority.