You Make the Call: Are You Being Lied to?

The anti-Measure I ballot statements, fair or foul?

Did the authors of the anti-Measure I ballot statements, in particular, Berkeley City Councilmember Gordon Wozniak, intentionally mislead Berkeley voters regarding Measure I?

Please put aside your opinion of Measure I and IRV and take a moment to evaluate the truthfulness of the arguments against Measure I as put forth by Gordon Wozniak. Click on the highlighted statements and get the full story.

If you conclude, as I have, that Mr. Wozniak has intentionally tried to mislead Berkeley voters, what does that say about his fitness to hold office? Politicians have been impeached and recalled for less.

Comments? Sent them to [email protected]

Mr.Wozniak's Argument Against Measure I

This argument was taken from the official City of Berkeley election website. The argument's misleading statements have been highlighted; clicking on a statement will take you to the critical analysis of that statement. You may use your browser's Back button to return to the argument.

Voting systems should not be changed without very careful study. The methods used to elect representatives are critical to our democracy. Although our present plurality voting system has many faults, it has one overwhelming advantage. It is simple enough for everyone to understand.

Instant Runoff Voting (IRV) is very complicated and confusing. SPOILED BALLOTS will be more common.

IRV systems are currently NOT LEGAL in California.

The California Secretary of State did not allow San Francisco to use IRV in its November 2003 municipal elections.

The Alameda County Registrar of Voters has stated that:
1. He cannot allow Berkeley to consolidate its general municipal election with the statewide election, if it uses an IRV system.

2. Neither current ELECTRONIC VOTING MACHINES nor ABSENTEE BALLOTS can handle both IRV and traditional elections on the same ballot.

3. An election utilizing two methods of voting (traditional and IRV) would result in MASSIVE VOTER CONFUSION.

IRV is MORE EXPENSIVE because Berkeley's municipal election cannot be consolidated with Alameda County.

The recently revealed flaws in electronic voting machines will be magnified with IRV. A paper trail may be impossible with the complicated transfer of votes between candidates.

In most forms of IRV ALL VOTES ARE NOT COUNTED nor IS A MAJORITY REQUIRED TO WIN.
THE SPECIFIC FORM OF IRV SYSTEM IS NOT MENTIONED IN THE BALLOT MEASURE - there are many forms of IRV. Which one are you voting for? YOU DO NOT KNOW.
Let's wait until we know what works and what we are voting on.
Keep our traditional voting system!
VOTE NO on Instant Runoff Voting!
s/MAUDELLE SHIREK, Vice Mayor
s/BETTY OLDS, Councilmember
s/GORDON WOZNIAK, Councilmember
s/WILLIAM WHITE, Chair, Police Review Commission
s/HELEN MEYER, Executive Vice President, Meyer Sound

Wozniak's Rebuttal to the Argument For Measure I

VOTE NO ON MEASURE I. Do Not Give City Council a BLANK CHECK To Choose Any Instant Runoff Voting (IRV) System.

· IRV IS NOT CAMPAIGN REFORM. The Alameda County Registrar of Voters stated "As an election official with nearly twenty years of experience conducting elections, I can assure you that this type of system would result in very high numbers of disqualified ballots and disenfranchised voters."

· IRV IS A STEP BACKWARD. In the 2000 London mayoral election, of the 581,761 first choice votes for eliminated candidates, only 36% were counted in the second round. If IRV is reform, why were 64% of these second choice votes not counted?

· IRV IS NOT SIMPLE. The 2000 Florida fiasco showed that even the "complexity" of a butterfly ballot can be fatal.

· IRV COSTS MORE. If IRV was used in the 2004 municipal elections, Berkeley would have to hold a special election at possibly a different date than the general election, at great cost and substantially lower voter turnout.

· RUNOFFS CAN INCREASE VOTER PARTICIPATION. In San Francisco's mayoral elections, more citizens voted in three of the last four runoffs than in the general elections.

· IRV DOES NOT REQUIRE A MAJORITY. The 2000 London mayor won with a 45% plurality.

· VOTE NO ON INSTANT RUNOFF VOTING - Join Senator Don Perata, Vice-Mayor Shirek, Councilmembers Olds and Wozniak, Police Review Commission Chair William White and businesswoman Helen Meyer.
INSIST THAT ALL VOTES ARE COUNTED! VOTE NO ON MEASURE I!
s/JESSE GABRIEL, ASUC President, 2002 -2003
s/DEAN METZGER, Chair, Transportation Commission/President, Claremont Elmwood Neighborhood Association
s/JANICE THOMAS, President, Panoramic Hill Association
s/SARA MacKUSICK, Chair, Community Environmental Advisory Commission/Vice-Chair, Citizens Budget Review Commission
s/MAUDELLE SHIREK, Vice Mayor

The following critical analysis has been provided by Steven Hill and Caleb Kleppner of The Center for Voting and Democracy; Dr. Shaun Bowler, a professor of political science at the University of California-Riverside, and attorney Rick Young.

Steven Hill responds: This statement is not supported by considerable available evidence. In fact, elections in the United States and from all over the world using IRV's ranked ballots reveal that the opposite is true. Ranked ballots have resulted in extremely low rates of ballot spoilage, attesting to the ease of use for voters. When Ann Arbor, Michigan used instant runoff voting in 1975, just a few months after its adoption by voters in November 1974, voter error declined sharply, from 2.3% to 1.2%. Voter error in New York's local school board elections using ranked ballots has been less than the error rate in the city for balloting in the presidential election. Internationally, instant runoff voting is used to elect the president of Ireland and a similar ranked-choice system is used to elect the parliament of Malta. In both elections, the rate of invalid ballots is typically less than 1.0% - well below the national error rate of more than 2% in the American presidential election in 2000.
For 100 years, political scientists have observed literally thousands of elections using ranked ballots in places like Cambridge, MA, Australia, the Republic of Ireland, Great Britain and elsewhere. There is simply no evidence that voters have any difficulty with ranking candidates. According to Dr. Shaun Bowler, a professor of political science at the University of California-Riverside who specializes in voting systems, the idea that ranking candidates is too hard for the voters is, "flat wrong."

Caleb Kleppner adds: This [Wozniak's statement] is false. Electronic voting machines prevent spoiled ballot, so the error rate will not change under IRV. It is no more difficult to fill out a valid absentee IRV ballot than a traditional one: you just have to indicate a valid first choice. However, IRV ballots are even more forgiving than traditional ballots, because if you fail to list a 1st choice but list a 2nd or 3rd choice, your vote will count. As an example, when Ann Arbor, Michigan used IRV to elect its mayor in 1975, the rate of spoiled ballots was 50% lower than the previous election.

B. Mr. Wozniak claims: IRV systems are currently NOT LEGAL in California.

Steven Hill responds: This is false. In fact, the Secretary of State has completed its legal review mentioned in the letter dated March 21, 2002 [in which in which the Chief of Elections reports "we have initiated a legal review of pertinent federal and state statutes, as well as any applicable regulations or procedures, to identify any conflicts or issues, if any, in the use of IRV in the elections process in California."] and found that IRV is legal in California. In recent meetings with Secretary of State Kevin Shelley, I have been a part of a team that is devising the protocols for testing voting equipment that will be used in California for IRV elections. The Secretary fully expects that IRV will be used in San Francisco's elections for the Board of Supervisors in November 2004.

Rick Young adds: There is a big difference between no certification for certain machines and illegality of an entire scheme. For example, the DMV will not certify all cars as fit for use (some fail the smog test while others have faulty brake lights). But that does not mean that cars as a whole are an illegal transportation choice. Ask yourself: would Santa Clara County, San Leandro, Oakland, and San Francisco all pass legislation on IRV if IRV were illegal? Don't you think the Berkeley City Manager or the City Attorney's Office would have said something to the City Council if voters were about to be asked to pass something that were illegal?
On December 12, 2003, I visited Mr. Wozniak in his City Council office and voiced my displeasure with his IRV arguments, in particular, his assertion that IRV is "illegal." After I pointed out that there is nothing illegal about IRV, and it is the machines and software that must by certified - not IRV itself - Gordon admitted that his ballot statement was not exactly accurate. So, when he wrote a letter to the Berkeley Daily Planet, Mr. Wozniak used these words, "There are no forms of IRV that are presently certified by the State of California." Better, but still wrong: IRV itself does not need certification; the machines and software do. But at least he did not call IRV illegal.

C. Mr. Wozniak claims: The California Secretary of State did not allow San Francisco to use IRV in its November 2003 municipal elections.

Caleb Kleppner responds: This is false. The Secretary of State has no authority over whether or not a municipality can use IRV, and the Secretary of State did not prevent San Francisco from using IRV in 2003. The Secretary of State has authority over voting equipment, and the modifications to the voting equipment in San Francisco had not been approved in time for the election, so the Secretary of State forbid the use of particular voting equipment in 2003, not the use of IRV.

Steven Hill adds: This [Wozniak's statement] is incorrect. San Francisco did not use instant runoff voting for its November 2003 elections for one simple reason: the voting equipment being modified by the city's vendor, Election Systems and Software, was not yet ready to perform the task of counting IRV's ranked ballots. The vendor was still in the middle of upgrading the equipment and completing the application for certification of its IRV-ready equipment.
Legally speaking, the Secretary of State has no jurisdiction over deciding whether a charter city like San Francisco may use an electoral system like instant runoff voting, cumulative voting, at-large elections, district elections, or any other electoral system. That is something only a court may decide, and only after a lawsuit has been filed to prevent implementation. California election code says that no "voting system" may be used in California without the prior approval of the Secretary of State. (Elections Code Sections 19200 - 19201), but many have misconstrued this passage. The Legislature defined "voting system" as "any mechanical, electromechanical, or electronic system and its software, or any combination of these, used to cast or tabulate votes, or both." Elections Code Section 362. For instance, the Secretary defines "voting system" using the Elections Code's definition, and adds: "Examples: DataVote, Mark-a-Vote, Votomatic; Pollstar, Optech, etc." See Procedures for Approving, Certifying, Reviewing, Modifying, and Decertifying Voting Systems, Vote Tabulating Systems, Election Observer Panel Plans, and Auxiliary Equipment, Materials, and Procedures, at Section1858.
In addition, charter cities like Berkeley and San Francisco have broad discretion under California law to chart their own course in terms of matters like how it conducts its elections. For example, the City Attorney of San Francisco issued an opinion dated February 20, 2003: "The California Constitution provides that charter cities such as San Francisco may govern their municipal affairs, including the manner and method of electing local officeholders. Cal. Const., Art. XI, § 5. When a charter city adopts a law governing municipal elections, and that law poses a genuine conflict with State law, the local law prevails unless the State law advances a matter of statewide concern. Johnson v. Bradley, 4 Cal.4th 389, 398 (1992), discussing California Fed. Savings & Loan Assn. v. City of Los Angeles, 54 Cal.3d 1, 17 (1991)."
Read properly, the plain language of Division 19 ("Approval of Voting Systems") of the Elections Code restricts the Secretary's responsibility to approve or deny certification to only machinery and equipment used in elections, and the procedures used with that equipment. The Secretary has no authority under Division 19 of the Code to certify an electoral method such as instant runoff voting added to a municipal charter.

D. Mr. Wozniak claims: The Alameda County Registrar of Voters has stated that [h]e cannot allow Berkeley to consolidate its general municipal election with the statewide election, if it uses an IRV system.

Rick Young responds: Former Oakland City Councilmember Danny Wan wrote to the Alameda County Registrar of Voters, Mr. Bradley Clark, concerning the Registrar's reluctance to implement IRV in Oakland. Mr. Wan first noted that because Oakland is a charter city, it can establish its own election procedures. He then asked Mr. Clark if there were any legal constraints on IRV that he was not aware of. Mr. Clark replied "As Registrar I have made a policy decision for my office not to conduct any instant run-off or preferential election until there are State laws to regulate."
It is, therefore, false to say that the Registrar "cannot" allow it. Rather, it is simply Mr. Clark's policy decision: he simply chooses not to. As explained above/below, Mr.Clark is both legally and technically able to conduct such an election.

E. Mr. Wozniak claims: The Alameda County Register of Voters has stated that [n]either current ELECTRONIC VOTING MACHINES nor ABSENTEE BALLOTS can handle both IRV and traditional elections on the same ballot.

Steven Hill responds: Broadly speaking, this statement is incorrect. In fact, San Francisco intends to handle both IRV and traditional elections on the same ballot in November 2004, when it will elect its Board of Supervisors using IRV in the same election in which San Francisco voters will vote for president, Congress, and other offices. San Francisco typically has about 30 percent of its ballots cast by absentee voters, and the absentee ballots also will handle both IRV and traditional elections.

F. Mr. Wozniak claims: The Alameda County Register of Voters has stated that [a]n election utilizing two methods of voting (traditional and IRV) would result in MASSIVE VOTER CONFUSION.

Caleb Kleppner responds: [T]he assertion about massive voter confusion is false. Current elections involve more than one method of voting: single winner and at-large elections. In fact, some ballots involve 3 types of elections: single winner plurality, single winner runoff, and at-large plurality. Unless the Registrar thinks "massive voter confusion" currently reigns, it is a false statement that massive voter confusion would result from the combination of an IRV and a non-IRV ballot. The 1975 Ann Arbor ballot combined IRV with a vote-for-one ballot, and the rate of voter error went down by 50%.

Rick Young adds: Voters will have an opportunity to familiarize themselves with any changes. Measure I states "The city clerk shall conduct voter and community education to familiarize voters with instant runoff voting."

G. Mr. Wozniak claims: IRV is MORE EXPENSIVE because Berkeley's municipal election cannot be consolidated with Alameda County.

Rick Young responds: Measure I clearly states that IRV can be implemented if and only if certain conditions are met. While some ballot measures are full of legalese, the text of Measure I is, for the most part, written in plain English. Check out this key section:

"Notwithstanding any section of this charter to the contrary, upon a determination by the City Council of all of the following, that:

a) the voting equipment and procedures are technically ready to handle instant runoff voting in municipal elections;

b) instant runoff voting will not preclude the City from consolidating its municipal elections with the County; and

c) instant runoff elections will not result in additional City election costs, the council may by ordinance establish a system of instant runoff voting for the offices of mayor, city council, and auditor, in any manner permitted by the State of California Elections Code."

Just in case you think that there is some trick in the wording you don't see, check out the analysis by the Berkeley City Attorney:

Under this proposed charter amendment, before the council may adopt a system of instant runoff voting it must make three findings: that: 1) voting equipment and procedures are technically capable of handling instant runoff voting; 2) that instant runoff voting will not preclude consolidation of City elections with Alameda County; and 3) that there will be no increase in City election costs as a result of instant runoff voting.

Allow me to say it in even more simple terms: under Measure I, Berkeley cannot use IRV until the City Council determines that IRV works, that the city can continue to consolidate elections with the county, and that it is not more expensive than the existing system.
And all indications are that IRV will save the City a lot of money. The analysis by the Berkeley City Attorney's office has found that under IRV, "the savings from avoiding a runoff election ranges between $100,000 for a runoff election in a council district, to $300,000 for a citywide runoff election for mayor or auditor. These cost savings may be offset in part by some costs associated with establishing a system of instant runoff voting."(emphasis added)

So why does Mr. Wozniak keep talking about Berkeley having its own elections? Under Measure I, that simply cannot happen. What information does Mr. Wozniak have that no one else has?

H. Mr. Wozniak claims: The recently revealed flaws in electronic voting machines will be magnified with IRV. A paper trail may be impossible with the complicated transfer of votes between candidates.

Steven Hill responds: This is incorrect. San Francisco's upcoming IRV election in November 2004 will have a paper trail. The presence of a voter-verified paper trail is independent of IRV.

Rick Young adds: In fact, Secretary of State Kevin Shelly is requiring all Direct Recording Electronic (DRE) voting machines in California to include a voter verified paper trail (new machines by July 1, 2005 and existing machines by July 1, 2006). As for concerns about electronic voting machines, we already use them in Berkeley. If there are flaws with electronic machines, they will be there with or without IRV.

I. Mr. Wozniak claims: In most forms of IRV ALL VOTES ARE NOT COUNTED.

Caleb Kleppner responds: This is simply false. In every IRV system, each voter has one vote, and every vote is counted, just like in any other election. Not all voters choose to indicate a preference between the top two candidates, but that is a very different statement than, "all votes are not counted."

Steven Hill adds: In his court papers, Mr. Wozniak relies on an example of the mayoral election in London to bolster his statement. He cites information on the London election from the web site of my organization, the Center for Voting and Democracy. Yet Mr. Wozniak deleted the specific information in which our web site makes clear that the London mayoral election did NOT use instant runoff voting, but rather another electoral method called the "supplementary vote." This distinction, which is clearly stated on our web site, is missing from Mr. Wozniak's statement. With the supplementary vote, voters are limited to two rankings. Consequently, the supplementary vote can lead to a higher number of voters whose rankings run out and therefore do not participate in the final runoff round.

J. Mr. Wozniak claims: nor IS A MAJORITY REQUIRED TO WIN.

Caleb Kleppner responds: This is false. In every IRV system that is or has been used in public elections, winning required a majority of the valid votes. The system always reduces the field to the strongest two candidates, and the winner is the candidate who receives a majority of the valid votes in the final round of calculation.

Steven Hill adds: This statement is incorrect, though it depends greatly on how one defines "majority." In IRV, the winner must have a majority of what is defined as "continuing ballots." A ballot is deemed "continuing" if all of the candidate rankings have not been eliminated or there are still more choices indicated on the ballot. But this is also true of a traditional "two round" runoff like the type currently used in Berkeley. Consider an election in which 100,000 voters vote in the first election, but in the subsequent runoff voter turnout drops by 30% to only 70,000 voters. Suppose the winning candidate in the runoff has 60% of the vote, or 42,000 votes. That winning total is only 42% of the original 100,000 voters in the first election (declines in voter turnout in runoff elections is not uncommon. In fact, turnout in Berkeley's December runoffs since 1986 is 28% lower than in November general elections). Thus, one could just as easily define two round runoffs in a similar way as IRV:

"The winner is the candidate who has a majority of votes from the 'continuing ballots,' with continuing ballots defined as those cast by voters who participate in the second election ONLY."

Rick Young adds: A strict majority (50% of the votes plus one) is not even needed in Berkeley right now. Under the current system, candidates for the offices of mayor, auditor, and councilmembers of the City only need 45% the votes cast for each such office. (If voters approve Measure H, which will also be on the March 2, 2004 ballot, the required percentage would be lowered to 40%.) It is important to note that if Measure I passes, under an IRV system, the winner would still need a majority of the valid votes in the final round of calculation.

K. Mr. Wozniak claims: IRV IS NOT CAMPAIGN REFORM. The Alameda County Registrar of Voters stated "As an election official with nearly twenty years of experience conducting elections, I can assure you that this type of system would result in very high numbers of disqualified ballots and disenfranchised voters."

Rick Young responds: Mr. Wozniak has accurately quoted from a letter written by the Alameda County Registrar of Voters. But one small problem: Mr. Wozniak has omitted the previous sentence which gives the reader the correct context. That sentence reads "Under a full preferential system such as is used in the Australian State of Victoria, a voter MUST mark a preference for every candidate or the ballot is considered spoiled and not counted." Reading the sentences together, it becomes very clear that the Registrar of Voters was discussing one particular type of IVR, not IRV in general as Mr. Wozniak would lead the reader to believe. Ask yourself why Mr. Wozniak lifted a sentence completely out-of-context. Was it an attempt to deliberately mislead the voters?
(For the full text of the letter, see page 32 of the City Council packet)

L. Mr. Wozniak claims:IRV IS NOT SIMPLE.

Caleb Kleppner respons: This is false. Voters simply rank candidates in order of choice. Second graders can do this. The evidence from places that actually use IRV shows that it is not complicated and confusing to voters.

Dr. Shaun Bowler adds: Preferential voting is a pretty natural act - especially these days. Every day consumer choice means choosing over a range of options where people are quite ready to express preferences ("if restaurant A is crowded let's go to B" or - in a restaurant - "if you don't have apple pie left I'll have berry"). Pretty much every day we express a preference ranking.. which is what IRV implies. (i.e "If I can't have candidate Apple I'll have candidate Berry"). So ranking choices is certainly not hard or unfamiliar to any of us these days.

M. Mr. Wozniak claims: IRV COSTS MORE. If IRV was used in the 2004 municipal elections, Berkeley would have to hold a special election at possibly a different date than the general election, at great cost and substantially lower voter turnout.

Rick Young responds: This is false and misleading. Apparently Mr. Wozniak believes if he repeats his statements enough, people will begin to believe them. As discussed above in Statement G, Measure I requires that Berkeley cannot use IRV until the city council determines that it is cheaper than the existing system and the city council certifies that the city will continue to be able to consolidate elections with the county.

N. Mr. Wozniak claims:RUNOFFS CAN INCREASE VOTER PARTICIPATION. In San Francisco's mayoral elections, more citizens voted in three of the last four runoffs than in the general elections.

Rick Young responds: Mr. Wozniak again selectively edits his information to cast the most positive light on his arguments. If we were to examine all December run-off elections - not just mayoral elections - in San Francisco since 1975, we would find that turnout declined in 7 out of 10 runoffs elections.
But why does Mr. Wozniak examine San Francisco when the most relevant place is Berkeley itself? Perhaps because turnout in Berkeley's December runoffs has declined for ALL EIGHT runoffs since 1986 by an average of 28%.

Steven Hill adds: This [Mr. Wozniak's claim] is extremely misleading. For instance, one study by my organization, the Center for Voting and Democracy, revealed that voter turnout declined in 82 of 84 runoff races for all federal primary elections between 1994 and 2002. The average turnout decline per election race was 35 percent. The overwhelming weight of evidence from both local and national examples reveals that runoff elections typically lead to lower voter turnout.

Last updated 2/13/04. Mail your comments on formatting, broken links or browser issues to the webmaster.