An Ice Cream Lover's Guide to Ranked Choice Voting
Our first installment of the Warrant[Ed] series covers Warrant Article 18 and Ranked Choice Voting. "Warrant[Ed]" is a typographical play on "warrant articles" (WA) and my quest to educate myself on the bylaws-to-be facing Brookline.
Let's do a quick mental survey.
Rank the following ice cream flavors with 1 (most preferred) and 4 (least preferred)
❑ Mint Chocolate Chip
❑ I'm not a fan of ice cream :(
As a self-confessed sugary snack specialist, my results would look like this:
Mint Chocolate Chip
I'm not a fan of ice cream :(
Regardless of whether your responses match mine (congrats!) or whether you admitted that you're not an ice cream fan (maybe we can rank vegetables next?), most of us understand the concept of ranking our choices based on preference.
Now, imagine if we applied the same preferential logic to how we elect our government?
While ice cream and politicians both come in a variety of flavors (some more appetizing than others; looking at you horseradish ice cream with red tomato vinaigrette), the two may not be a fair comparison. Nevertheless, the mechanics of Ranked Choice Voting (RCV) offer a user-friendly method for conducting elections.
What does WA 18 have to do with RCV?
Our first installment of the Warrant[Ed] series covers WA 18, which was co-petitioned by five Town Meeting Members along with Select Board Member Raul Fernandez to help study RCV adoption in Brookline town elections (See Article Explanations 2020 STM, p. 32).
WA 18 seeks to explore the logistics and funding necessary to implement RCV and does not directly make RCV the de jure method to conduct elections in Brookline—The latter outcome may involve a home rule petition, another WA, or a combination of different legislative processes.
WA 18 amplifies the RCV movement in Massachusetts, especially as the state gears up for a vote on ballot Question 2, an initiative to bring RCV to all primary and general elections for state executive officials, state legislative offices, federal congressional and senate seats, and certain other offices. Remember to vote Yes on 2 on November 3!
Cool. How does RCV work?
Also known as Single Transferable Voting and Instant Runoff Voting, Ranked Choice Voting serves as a non-partisan, flexible solution that adapts to both single-winner offices (e.g., governor, district attorney, etc.) and multi-winner ballots, such as how Brookline determines Town Meeting membership. RCV resolves the inequalities inherent to current plurality-rule elections that favor candidates with the most votes, even if they receive less than 50% of popular support.
In single-winner elections, RCV empowers voters to rank candidates in order of preference (one, two, three, etc.), ensuring that the people vote for candidates we support rather than simply voting against those we oppose. The candidate who is ranked as the first-choice of more than half of the electorate earns the victory.
If no candidate receives the majority of votes, RCV allows for an instant runoff in which the candidate with the fewest votes in the first round is eliminated and his or her supporters have their ballots count for their second-choice candidate. The instant-runoff calculations continue until one candidate earns 51% of the popular vote and thereby achieves the broadest appeal to the majority of voters.
RCV operates with the same preferential procedures in multi-winner elections. Rather than aiming to get more than 50% of the vote by default, candidates running for offices that have several winners (e.g., city council) simply have to garner a certain share of the electorate in proportion to the number of seats up for grabs. The threshold is calculated as the minimum percentage of total votes cast required to ensure that the number of possible winners equals the number of open seats.
For example, the multi-winner RCV threshold for an election with two open seats on city council would be 33.3% of the total number of votes plus 1 vote. Imagine that three candidates (Joe, Elizabeth, and Don) are running in that election, which had 100 voters participate. Given such a scenario, the threshold is 33 votes, which is one-third (33.3%) of the total votes (100), plus 1 vote. Joe and Elizabeth each receive 34 votes respectively (i.e. 68 votes) and win the two open seats since Don only received 32 votes, thus failing to meet the threshold.
Of course, the math isn't always as clean-cut as in the above example. After voters rank their candidate options, the first-choice candidates who reach the threshold earn an open seat, and any votes in excess of the threshold are then counted for the voters' second-choice and so on until all available seats are filled.
Makes sense. Now why is RCV better than how we do elections today?
RCV promotes reflective representation that benefits diverse groups and provides voters with more choice. The winner-take-all system frequently reduces the people’s electoral power to a choice between the lesser of two evils or splits the vote of like-minded groups (September's Democratic primary in Massachusetts 4th was a good example of a ballot bloated with similar candidates), both of which result in outcomes that expose unjust barriers to people of color, women, and third parties.
In a study of city councils across more than 7,000 cities, Jessica Trounstine (Princeton University) and Melody E. Valdini (Portland State University) note that the impact of single-winner plurality elections in increasing diversity varies by race and gender: Black and Latina women experience no positive effect to proportional representation in first-past-the-post electoral systems involving single-member and multi-member districts.
RCV also encourages candidates to run more positive, less-polarized campaigns. Instead of mudslinging and feeding into a toxic, hyper-partisan culture, politicians in RCV elections will opt to appeal to a broad base of voters since they will be competing to be the second-choice of even their opponents' first-choice voters.
One Person = One Vote
Such a simple idea claims to be the core of our country’s constitutional calculus, granting each citizen equivalent voting power. In practice, however, the electoral equation fails to add up to its egalitarian promise of fair representation—The winner-take-all system forces us to select between candidates that more often than not fall short of reflecting the wider community.
You do the math: We deserve a better approach to conducting elections, one that accounts for the diversity enriching our communities.
So, is "majority rule" a thing in Brookline?
I was curious myself about whether or not town elections were producing majority-representative outcomes. So, I did what any self-respecting Excel monkey would do: I went bananas on 20 years of electoral data available from the Town Clerk.
I focused primarily on Select Board (SB) elections since the staggered terms of Select Board Members allow us to observe both types of elections in which RCV excels: single-winner and multi-winner, as well as contested (i.e. more candidates than seats) and uncontested races.
During election years in which only one SB seat was up for vote, I examined whether candidates were able to achieve the electoral majority based on single-winner RCV (i.e. receiving more than 50% of the total votes cast).
During election years in which two SB seats were up for vote, I applied the multi-winner RCV logic for proportional electoral majority: >33.3%.
Accounting for these conditions, I observed that Brookline has had:
Three contested, one-open-seat SB elections (2004, 2007, and 2019) and only once has a winner (Jesse Mermell in 2007) gained more than 50% of total votes cast (See Table A below);
Nine contested, two-open-seat Select Board elections and only three instances (2011, 2014, and 2017) in which both winners gained the proportional multi-seat majority of more than 33.3% (See Table B below); and
Three uncontested, two-open-seat elections (2009, 2010, and 2012) in which only a single winner in 2010 (Jesse Mermell) gained >33.3% of the total votes cast—The other unopposed winners in all three elections (including Mermell’s fellow 2010 winner, Andrew Ghobrial) gained less than 33.3% of total votes cast, with blank ballots receiving more of the remaining votes than said winners with the exception of Mermell (See Table D below).
Depending on the election year and available seats, certain candidates for SB win despite opposition from most voters. SB election results have historically failed to represent the town's popular vote—the broadest segment and voice of the community.
Ok, I'm convinced: Change is needed. What's next for WA 18?
Last week, the Personnel Subcommittee of the Advisory Committee (AC), aka Brookline's finance committee, hosted a hearing on WA 18 (yours truly even submitted a written public comment) and received a presentation from the co-petitioners. The subcommittee members in attendance voted unanimously in favor of WA 18; they will then report the results of their review to the full AC. The SB will be conducting its own review of the proposed article next week.
The full AC will consider the concurrent reviews of both its subcommittee and the SB and will hold a vote on WA 18, determining the written recommendation that's presented to the Town Meeting on November 17.
I'm hopeful that Town Meeting will enact WA 18 since it ultimately advances Brookline’s commitment to equity and racial justice. We would be one step closer to joining the ranks of 18 other municipalities and Maine that use RCV for local, state, and/or federal elections.
Equipped with RCV-energized elections, Brookline can finally renew the electoral equation with a more progressive qualifier:
One Person = One Fair Vote
Ranked Choice Voting (RCV) improves upon our current, plurality-style elections by empowering the majority of voters to elect our leaders.
WA 18 is a proposed resolution to study RCV implementation in Brookline town elections.
Make sure to vote Yes on Question 2, a state-wide ballot measure that would bring RCV to specific elections in Massachusetts.
Any thoughts on a topic I should cover? Feel free to send me a note!
Craving analysis of local issues under the lens of data and ethics? Subscribe here.