Testimony In Support of
H. 575, H. 576, H. 608, and H. 610
Instant Runoff Voting
Pamela H. Wilmot, Executive Director, Common Cause Massachusetts
Joint Committee on Election Laws
October 19, 2015
Mr. Chairman, thank you for allowing me to testify today. Common Cause Massachusetts supports H. 575, H. 576, H. 608, and H. 610 filed by Representative Kaufman and Representative Story. We thank all of the sponsors for their advocacy for this innovative electoral reform called instant runoff voting, or IRV for short.
IRV was developed at MIT more than a century ago to address the problems of the current system of “plurality” voting and has been given new viability with new voting equipment that allows tabulation of ranked ballots at the same speed as conventional ones. Instant Runoff Voting has been widely used in Europe and in Australia for decades and in several local jurisdictions in the US more recently, including in Hendersonville, North Carolina; San Francisco, Berkeley, Oakland, and San Leandro, California; Minneapolis, Minnesota; Tacoma Park, Maryland; Portland, Maine; St. Paul, Minnesota; and Telluride, Colorado.  It will soon be implemented in Memphis, Tennessee and Santa Fe, New Mexico.
I will describe the mechanics in the next section, but for voters, IRV is as simple as 1, 2, 3. Rather than voting for a single candidate, voters rank their candidates in order of their choice (although they are not required to do so). So that in a field of 4 candidates, a voter would mark their first choice “1” their second choice “2” and so on.
Instant Runoff Voting has a number of benefits, the most important being that it ensures that the candidate who is elected represents the majority viewpoint in the district. We all know of elections where candidates were elected with very small pluralities, sometimes as low as 20%. While a small plurality doesn’t mean that a winning candidate would not have been the choice of the majority, it may mean that in some cases. When multiple candidates run from the same ethnic background, or when candidates with similar ideologies split the vote, the result is to elect a candidate who would not be the choice of the majority. This is anti-democratic and results in what mathematicians call “irrational” results. Our current plurality voting system is the most flawed in this regard. It also results in unfortunate situations where candidates are either asked to withdraw, or do so spontaneously, in order to prevent another candidate from winning. This should not happen, especially in Massachusetts with so few young people willing to run for public office.
Other IRV benefits include allowing citizens to vote for the candidate of their choice without fear of jeopardizing an election (so called “spoiler effects”); potentially increasing voter turn-out since no vote is “wasted”; and, importantly, discouraging negative campaigning, since candidates are looking for second and third place votes as well as first place ones. These are some of the reasons why Common Cause Massachusetts has joined a coalition of citizens, organizations, and reform-minded legislators committed to implementing IRV in Massachusetts elections.
Mechanics of IRV
For voters IRV is as simple as 1, 2, 3. What follows is a series of runoff elections conducted sequentially, but thanks to computers, nearly instantaneously. Here’s how an IRV election would work:
- A voter ranks candidates in order of choice—1, 2, 3….
- If no one candidate receives a majority then the last place candidate is eliminated and his or her votes are assigned to the voters’ second choice. If a candidate now receives a majority of re-tallied first choices, she or he wins. If no candidate has a majority, this process continues until a winner receives a majority.
- If a candidate receives a majority of first choices, she or he wins.
The mechanics of IRV are best understood visually and there is an animated presentation from Fair Vote that demonstrates it well:
The Cambridge System.
Massachusetts has one municipality already using ranked-order balloting: Cambridge. It uses a system of proportional representation or “PR” where city council seats are apportioned according to voter preferences. IRV differs in that it determines the outcome of a single race, not many seats at once. It also is significantly different in its political operation. Proportional representation systems like Cambridge allow a candidate to be elected with a small fraction of the vote citywide. IRV, on the other hand, requires that candidates have the support of a majority of voters in their districts. Thus, while PR is a full representation system, allowing all points on the political spectrum to elect a candidate of their choice, IRV is a majoritarian or centrist one.
But the Cambridge experience is instructive in two ways: 1) it has shown that voters can use a ranked order ballots without much confusion and 2) it has demonstrated equipment compatibility and implementation costs for ranked-order balloting.
Can Current Voting Equipment handle IRV?
As the Cambridge experience shows, the most popular voting equipment in the Commonwealth, the Accu-vote optical scanner, can accommodate ranked-order ballots, including IRV.
So what will IRV cost?
The cost of implementing IRV is difficult to pin down because scale of implementation affects incremental cost, as does equipment compatibility. But given the dominance of the Accuvote equipment, the cost is likely to be quite low. There are three potential areas of cost, not including changing voting equipment, firmware, software, and maintenance.
Firmware: Cambridge paid $40,000 to the manufacturer of Accu-vote, Diebold, to develop firmware (like software, but on a chip) to accommodate their ranked order ballots. This was a one-time cost and programmers have stated that it will also accommodate IRV. The firmware tells the scanner where to look on the ballot to record the relevant information. Even if there is some cost involved to update the program, it will likely be minimal and could be shared by multiple jurisdictions.
Software: The election results recorded by the firmware are then loaded into a computer which then runs a tabulation program. Cambridge paid a one-time cost of $15,000 to develop this software. The programmers have now updated it and will make it available for free to any jurisdiction that implements IRV in the future. They are doing this to increase the affordability of IRV to municipalities and to make the code open to public inspection.
Maintenance: Cambridge pays $15 per machine per year to change the chips from plurality vote for Presidential and statewide elections, to Cambridge PR for city elections. This cost is a nominal one that would be borne for jurisdictions using IRV.
In short, most of Massachusetts has compatible equipment, and the associated costs are not high. When replacing equipment is involved, the cost would be in between $3-5,000 per precinct, which may be offset in certain cases by eliminating nonpartisan runoff elections.
IRV is a reform that is gaining steam across the country. It is supported by individuals and groups from all sides of the political spectrum, from Republicans to Democrats, Libertarians to Greens. It has received prominent public support from President Barack Obama, Governor Howard Dean, and Senator John McCain. Advocates understand that there is a lot of education to do before this reform is ready for widespread acceptance, but several large cities on the west coast have adopted it and we hope that more will follow in their footsteps. I expect that you will be hearing a lot more about it in the next few years and I invite you to learn more. Visit the IRV section of our website at www.commoncause.org, or visit FairVOTE at www.fairvote.org for up-to-date research and policy on IRV and other electoral reforms.