Printer Friendly

Success for instant runoff voting in San Francisco.

In November 2004, San Francisco held its first election with instant runoff voting (IRV), an important innovation in democracy that could spread to cities and states around the country. IRV was used to elect seven seats on the city's Board of Supervisors and will be used for citywide offices next year. Although hotly contested--one election in fact drew a remarkable twenty-two candidates--exit polls showed that city voters overwhelmingly liked IRV and found it easy to use, a finding that crossed racial and ethnic lines. All winners were decided either on election night or shortly thereafter. Observers noted how candidates reached out to more voters and engaged in fewer of the negative attacks that plague American politics.

IRV simulates a series of runoffs within a single election. Voters rank candidates in order of choice: first, second, third. These rankings indicate which candidate has support from a popular majority. In each runoff round of counting, the weakest candidate is eliminated. If a voter's first choice gets eliminated, his or her vote goes to that voter's second "runoff" choice. San Franciscans are thus able to vote for the candidates they really like without concern for the dilemma of a "spoiler" candidacy.

Recommended as an option in the Eighth Edition of the National Civic League's Model City Charter, instant runoff voting has been used for decades for national and local elections in Ireland and Australia and recently was adopted for mayoral elections in London and Wellington, New Zealand. Before switching to IRV, San Francisco used to decide the majority winner in a December runoff election, an expensive procedure that the city had to perform each year from 1998 to 2003. With citywide runoffs costing the city more than $3 million, IRV will save millions of tax dollars. Candidates don't have to raise more money for a second election, so independent expenditures have dropped significantly. Winners are now determined in the general election, when voter turnout tends to be highest, receiving far more votes than winners in past city council runoffs. Cities electing leaders in multiple elections would see similar advantages by using the instant runoff instead of a "delayed runoff." It's no wonder that a San Francisco State University exit poll commissioned by the city found that only 13 percent of voters preferred the old runoff system. (1)

These advantages aren't the only reason the nation is watching. To understand the national implications of instant runoff voting, think back to the 2000 presidential election. If the nearly hundred thousand Ralph Nader voters in Florida could have ranked a second candidate as their runoff choice, Al Gore might be president today. Democrats must have wished many times throughout the 2004 presidential campaign that Florida and other battleground states were using IRV. Similarly, Republicans could have responded to Ross Perot's candidacy in the 1990s simply by trying to get as many first- and second-choice votes as they could.

Advocates of IRV say the system is more accommodating to independent-minded candidates, who can run and introduce fresh ideas into electoral debate without spoiling the outcome for another, more popular candidate. These candidates can push important issues that get ignored in this era of poll-tested campaign sound bites and bland appeals to undecided swing voters. Voters are liberated to vote for these candidates, knowing that even if their first choice can't win, their vote will go to a front-running candidate as their second or third choice.

IRV also offers something for those tired of polarized politics and mudslinging campaigns. IRV encourages coalition building among candidates. Because winners may need to attract the second or third rankings from supporters of other candidates, we saw less mudslinging and more coalition building and issue-based campaigning in most of San Francisco's seven council races. A New York Times profile of the campaign was headlined "New Runoff System in San Francisco Has the Rival Candidates Cooperating." (2)

Drawing bipartisan support from reform-minded Republicans and Democrats such as John McCain and Howard Dean, legislative bills for IRV have been introduced in twenty-two states. Ballot measures supporting IRV passed by a margin of two-to-one in all three cities where it was on the ballot this year: Berkeley, California; Burlington, Vermont; and Ferndale, Michigan. All three cities are now on a clear path to enact IRV in the coming years. Officials in bigger cities such as New York, Los Angeles, and Seattle are watching San Francisco's successful implementation closely.

San Francisco is leading the United States with modern democratic methods, but we expect more to follow. The main hurdle for many cities has been voting equipment that cannot support IRV. This is starting to change, and whenever the debate is focused on its merits relative to a two-round election or first-past-the-post plurality race, IRV wins.

Did IRV live up to its promise? Let's examine several suppositions posed by advocates of the system.

No More December Runoff Elections

Supposition: San Francisco voters and election officials will not have to worry about December elections. The Department of Elections will run one fewer election per year, providing more time to prepare for the next election. This will help the department run better elections.

Assessment: Supposition correct. San Francisco will not hold a December runoff election, for the first time since 1998. Winners were definitively determined in three races on election night, and in the four remaining races on November 5. Winners likely will be identified within twenty-four hours in future races.

Significant Tax Savings

Supposition: The city will save significant money in eliminating runoff elections. According to figures released by the Elections Commission in 2003, it cost San Francisco taxpayers at least $3 million to administer citywide runoff elections. Administering a runoff election in any one of the eleven supervisorial races cost a prorated amount. By eliminating the delayed runoff, the system will save taxpayers the costs of public financing for supervisor races. Public financing makes available up to $17,000 in public funds to candidates in a runoff election, so the savings is up to $34,000 per supervisor runoff as well as the administrative costs of running the program.

Assessment: Supposition correct. San Francisco avoided December runoff elections in the four supervisor races in which the candidate did not win an outright majority, so the city saved at least $1.2 million in administration and public financing costs this year alone. Avoiding future runoff elections in the city's annual November elections will quickly repay the one-time costs of implementing IRV ($1.6 million to modify the hardware and software of voting equipment modifications, and $800,000 for voter education), leading to substantial ongoing savings to San Francisco taxpayers.

Increased Votes Cast in the Decisive Election When Winners Are Chosen

Supposition: More votes will be cast in the decisive election, and winners will receive more votes than winners in the old delayed runoff system. In the previous two non-IRV supervisor elections in 2000 and 2002, relatively high voter turnout elections in November were followed by a sharp decrease in voter turnout in December. The average decline in voter turnout from November to December was 50.5 percent in runoffs in December 2000 and 31.4 percent in runoffs in December 2002.

Winning candidates received a majority of the low turnout December electorate in those two runoff elections. But compared to the total voters who participated in that supervisorial election in November, winning candidates in December received a low of 25 percent and a high of 41 percent of the November turnout, with most races in the low end of this range (see Tables 1 and 2).

Assessment: Supposition correct. In any runoff system, the winning candidate must win a majority of valid ballots cast in the final round of counting between the top candidates (for example, the "continuing ballots") rather than a majority of total ballots that might have been cast for the race. In this year's IRV elections, all winning candidates won a greater share of the first-choice election turnout than any winning candidate in the December 2000 runoffs.

Here are the numbers for 2004. Every winning candidate in runoffs had a larger vote total and percentage than winners in 2000. All of the winners were ahead after counting first choices, but in several cases their share of the vote was small--meaning, without an instant runoff we would not have known if they had broad support; in District 5, for example, Mirkarimi received only 28 percent of voters' first choices among the twenty-two candidates. A plurality election can break down this way whenever more than two candidates run, thereby allowing a candidate to win a key office with far less than 50 percent support--including unrepresentative, polarizing candidates who never could have won with a majority.

Comparing the runoff elections in 2000 to the IRV election in 2004, no December winner in 2000 had more than 36 percent of the November total. In contrast, every winner in 2004 had more than 38 percent of the first count total--and all but one (Mirkarimi in District 5) had a percentage greater than 44 percent. Support for winners was significantly higher if determined by the number of people who ranked the winner with at least one of their three rankings. District 5 had twenty-two candidates, but Mirkarimi was ranked by 47 percent of voters. Every other winner drew at least one ranking from at least 53 percent of voters.

In sum, San Francisco's first IRV election was a resounding success. Voter understanding will only increase from its current high level. We applaud election officials, city leaders, candidates, and of course the voters for showing how this sensible refinement of plurality and runoff elections makes sense--and showing that it has never been more timely.
Table 1. Board of Supervisor Races, 2000
                                                     (Winner's Votes
                                                     in December
              November Election   December Runoff    Compared to Total
District      (Total Votes)       (Winner's Votes)   Votes in November)

District 1    28,194                   7,486         26
District 2    38,206                 No runoff       NA
District 3    24,860                   7,202         29
District 4    27,407                   8,453         31
District 5    36,115                   10,384        28
District 6    23,425                   8,472         36
District 7    33,867                   9,333         27
District 8    38,791                   9,578         25
District 9    23,765                 No runoff       NA
District 10   23,884                   5,887         25
District 11   25,023                   8,345         33
Supervisor Races, 2002
District 4    20,452                   8,289         41
District 8    31,902                   11,096        35

Table 2. San Francisco Supervisor Races with
Runoffs (Votes Counted as of November 10, 2004)

                        Total Valid   Votes in      Counted
District   Winner       IRV Votes     Final Round   for Winner

1          McGoldrick   28,011        25,253        13,615
5          Mirkarimi    34,413        25,681        12,999
7          Elsbernd     30,838        23,768        13,538
11         Sandoval     22,789        18,039        10,513

                                                  Percentage of
                        Winning Votes   Ballots   All Ballots
                        as Percentage   Ranking   Ranking
District   Runoff (%)   of All Votes    Winner    Winner

1          53.9         48.6            16,908    60.4
5          50.6         37.8            16,244    47.2
7          57.0         43.9            16,331    53.0
11         58.3         46.1            12,184    53.5

Measures of Success

* The City of San Francisco eliminated December runoff elections and determined clear winners shortly after the November election.

* The city saved $1.2 million by not having to administer four runoff elections. It will cover all costs associated with moving to IRV the first time it prevents the need for a citywide runoff, as took place in 1999, 2001, and 2003.

* Winners received significantly more votes and overall support than winners in traditional runoffs and winners in conventional plurality voting elections.


(1.) Neely, F., Blash, L., and Cook, C. "An Assessment of Ranked-Choice Voting in the San Francisco 2004 Election: Preliminary Report." San Francisco: Public Research Institute, San Francisco State University, 2004 (; retrieved Dec. 14, 2004).

(2.) Murphy, D. E. "New Runoff System in San Francisco Has the Rival Candidates Cooperating." New York Times, Sept. 30, 2004, p. A16.

Steven Hill is the Irvine Senior Fellow for the New America Foundation and author of Fixing Elections: The Failure of America's Winner-Take-All Politics (New York: Routledge Press, 2003; Robert Richie is the executive director of the Center for Voting and Democracy.
COPYRIGHT 2005 National Civic League, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hill, Steven; Richie, Robert
Publication:National Civic Review
Date:Mar 22, 2005
Previous Article:Networks in a new "gateway".
Next Article:HAVA or havoc?

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters