By Howie Fain

Rank the Vote is proud to publish this historical overview of ranked choice voting in the US, authored by Howie Fain. A retired public school teacher from Worcester, Massachusetts, Howie was a co-founder of FairVote in 1992. Find more of Howie’s work at

This is the fifth of nine installments on the history of RCV. We hope the history that unfolds in these posts is as revelatory and inspiring for you as it was for us. Let the curtains rise and the knowledge drop!

Ranked Choice Voting in the US, Act II, 1915-1947

Ohio was ground zero as the PR League honed in on municipal reform, with a complete package that included: 1. STV in a small (usually, 7- 9 members) council; 2. council-manager governance; and 3. officially non-partisan local elections. It also solidified the practice of referring to this particular election system by the generic term,  “proportional representation (PR).”

Ashtabula was first up, adopting and then holding its first election for a seven-member council, all in 1915. Other Ohio adoptions followed in Cleveland (1921); Cincinnati (1924); Hamilton (1926); and Toledo (1935). 

Actually, there were four other cities, in four other dispersed states, in quick succession between Ashtabula and the rest of the Ohio pack. Three were bit players who barely had time to say any lines before exiting the stage; evidently, there were immediate second thoughts and near-immediate repeals in Kalamazoo, MI; Sacramento, CA; and West Hartford, CT. The fourth was the longevity exception: Boulder, CO held its first PR election in 1918, and its last in 1950, making it all the way to Act III.

While there is so much more that will be said in this and Act III about the political difficulties PR faced in holding on, once adopted, its relative “complexity” and newness, and lengthy hand count (up to one week) cannot be ignored, especially as a factor in those cities that folded up the PR tent so quickly.

But right from the opening curtain, the proportional ideal—accurately representing the diverse interests of the electorate—was a conspicuous success, upending both winner-take-all results and ethos. When the 1915 results of Ashtabula’s first PR election were in, the local newspapers assessed the results. One wrote: “The drys and the wets are represented; the Protestants and Catholics; the business, professional and laboring men; the Republicans, Democrats, and Socialists; the English, Swedes, and Italians all are represented. It would be hard to select a more representative council in any other way.” And this in just a seven-member City Council! The other: “It is generally conceded that it has given Ashtabula a broadly representative council, probably the most representative in the city’s history, and that is the real aim of the [PR] system.”  

But political representation is often the embodiment of Newton’s third law: every action creates an equal and opposite reaction. Amid increasing anti-Catholic bigotry and KKK activity by those of “old American stock,” in Ashtabula, PR was targeted and then repealed in 1929 a decade and a half in.

Cleveland was the next Ohio jurisdiction, voting to adopt PR in 1921. As a far bigger city, the reformers conceded that the size of the Council could not shrink to as few as nine and still be representative, so the machine-dominated 33 single-district ward seats were replaced by a total of 25 Council seats, apportioned in four multi-member, proportionally elected districts (electing 5, 6, 7 and 7 members respectively). This was an important development for STV-PR, offering a model for geographic based multi-member election units within a single jurisdiction that is still relevant today, not just for a city, but for state and federal legislative bodies as well 

In 1920, two-thirds of Cleveland’s population was immigrant or first generation. Just four percent was black, which grew to eight percent over the next decade. PR in Cleveland was promoted in part to “safeguard against racial and religious prejudices” by ensuring racial and religious minorities political representation on city council.

The PR council was typically and broadly representative, with three African-American Councilors by 1927. At least one woman was elected to every PR Council, a streak that immediately ended when Cleveland reverted to the 33 separate ward seats in 1933 following repeal. PR in Cleveland was under persistent attack from its inception: five repeal votes were held in the same time period in which five PR elections were held.

Continuing to advance in Ohio, Cincinnati, Hamilton and Toledo followed Cleveland’s 1921 adoption shortly thereafter. Cincinnati’s 30+ year story is a major part of the near-final collapse of PR in Act III. Apart from the significant focus on racial diversity sustained under the system, Cincinnati was heralded by many reformers for its “good government” 3rd party model (“Charter” party in these officially non-partisan elections).

If 1915 was the official start of the play for PR, set in America’s heartland, 1937 represents the crown jewel making its grand entrance, the shining star that was going to carry STV-PR to full glory in the United States: New York City’s first PR election for City Council.

Proportional Representation in New York City replaced a 65 single-district representation system dominated by Tammany Hall Democrats (in the prior decade, winning an average of 91 percent of all seats, and in 1931 winning all but one of them) with a 25-member Council elected in different numbers in five different borough-based multi-member districts, the specific number to be elected from each being variable by turnout. 

NYC was also the only U.S. jurisdiction to selectively pull just PR from the usual reform package: it maintained partisan elections, and a directly elected strong mayor when adopting PR for the Council—it does not appear that IRV or any other voting reform proposal for elected single-winner positions in the city was even on the table alongside using PR for the Council.

No surprise, the combination of real choices in a deliberately fair system greatly opened up this partisan election process, and it wasn’t just a few Republicans who could now finally claim some of the available seats. Given the political interests of New Yorkers in this period, Communists and others across the spectrum also won seats, in fair proportion to the votes they won. But New York City is also a good reminder that ensuring proper representation of the majority is the essential flip side of ensuring proper representation of the minority: Democrats won an average of 60.3 percent of all seats during this period; it’s just that now, other parties, and even an occasional independent could be heard in legitimate policy debate. And, Tammany Hall had been decisively broken with the advent of PR.

As PR advocates hoped, the New York City prize immediately produced advances elsewhere, such as in 1938 when the Massachusetts legislature included the “PR” package as a new option for local governance (“Plan E”). Cambridge, engulfed in city hall scandal, immediately signed on, then Lowell and Worcester followed later. Four other smaller cities in the state also joined in, but unlike these three, withdrew quite quickly. By then, it was already early Act III, and the Massachusetts Legislature had itself changed course in dramatic and aggressive fashion. As elsewhere, the system was attacked by the displaced machines, but not only did Cambridge weather all the attacks, the 1941 opinion of the Massachusetts Supreme Judicial Court (Moore) upholding PR remains a great and relevant read to this day for RCV advocates, as it dissects both the ranking and the deliberately proportional elements of representation.

In Massachusetts, the Plan E enabling law and related election administration statutes included provisions relative to use of “proportional representation or preferential voting.” The latter almost certainly refers to single-winner RCV, but we are unaware of any single positions in any of these cities being directly elected (as Plan E included council-manager governance), let alone being elected with instant runoff voting.

Other cities around the nation (for a total of 22) rounded out the roster in Act II. The very last of these adoptions actually took place in Act III (1948, in Oak Ridge, TN). 

Even though the PR League of the United States had long since narrowed its focus and strategy to local governance, its 1932 merger with the National Municipal League effectively and formally shut the lid on PR advocacy for state and federal elections in the U.S.. That didn’t mean that winner-take-all state legislative and congressional elections were any less problematic—just that there was now no voice in the United States actively promoting a proportional alternative for those bodies, even nominally. As the principle of PR had become even more widespread globally in this time, specifically for state and national level elections, that was… strange.

On the international stage during this Act, the newly independent Republic of Ireland wrote STV-PR into its constitution for its national legislative elections. Though it uses parliamentary governance, Ireland also established direct election of a mostly ceremonial office of President, and chose to use IRV to determine the single-winner.

Stay tuned for the next installment,Ranked Choice Voting in the US, Act III, 1947-1960

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of Rank the Vote, its members, supporters, funders, or affiliates.