Ranked-Choice Voting Does Not Improve Representation

The more factions there are in a political contest, the more likely candidates with at least some positions in common are to band together to form a coalition. This is a principle of participatory democracy.

Imagine there are 15 candidates competing for a seat on a town council. Seemingly, it should be easy to gain the roughly 6.8% needed to become the plurality winner; this is a tiny fraction of the voting public. However, voters have 15 choices, which means each candidate only has a 6.6% chance of earning the vote of any given citizen.

Enter the concept of coalition building. Candidates who have relatively similar ideologies and platforms band together in the hopes of bolstering their electoral chances. In real political terms, this often means progressives ally themselves with their more centrist allies in the Democratic party while conservatives join forces with establishment Republicans.

For the purposes of our scenario, let’s imagine the 15 candidates are whittled down to 5 due to coalition-building: establishment Republican and Democrat contenders, a libertarian, a Green party candidate and an independent. These 5 coalitions emerge because each has a distinct platform and the respective candidates have decided their personal goals are less important than winning the seat and ensuring their ideology is influential in setting public policy, thus bolstering efficacy for their constituents. It is now more difficult to gain a plurality of the vote, which is 20.1% instead of the much lower 6.8% required when more candidates were running. However, because there are fewer divisors in the class of contenders, each candidate now has a 20% chance of being selected by each voter, rather than a significantly slimmer 6.6% chance. What’s more, as when candidates band together they each bring with them the votes of their committed partisans in the voting public, they further bolster their chances of success.

Elections, therefore, when considered through the lens of probability, are easier to win when there are fewer candidates in the race. Political parties know this, which is what leads to the rise of so-called “umbrella parties.” It is the self-interest of parties whose desire is electoral dominance that drives coalition-building between groups whose ideas often only line up in part.

However, when ideologies compete within parties, efficacy is often impeded rather than aided by the fact that parties ostensibly represent greater sectors of the polity. Factions that have joined because they cannot compete on their own, such as progressives within the Democratic party and conservatives within the Republican party, are often left feeling frustrated by the more centrist views the parties, whose focus is more often on retaining power than on legislating the principles in their platform, by whom their groups have been subsumed.

It is under the guise of greater efficacy and representation that new experiments in voting systems are being sold. Among the most popular are ranked-choice voting and approval voting. Both systems allow an individual to vote for as many candidates as they choose. Both ostensibly bolster efficacy through participation. Ranked-choice voting works much like it sound: a voter assigns a rank to the candidates for a given office, from their first choice to their last choice. It is, however, a system that requires a majority vote to win. So, if there is not a clear majority winner in the first round of voting, the candidate with the least amount of votes is eliminated and those votes are redistributed using an algorithm that factors in the ranking unique to each voter. The following video from the state of Maine, which recently implemented ranked-choice voting, explains well how tabulation works under the ranked-choice system:

Approval voting also allows candidates to vote for more than one candidate but is more straightforward. A person can vote for as many people as they want, until all their votes are used up. This is sometimes used in local races where more than one candidate is elected to a body, say to a town council that needs to fill 5 seats. In this situation, each voter is able to vote for 5 individuals. The top 5 vote-getters are considered the winners.

Ostensibly, because voters are freed from the tyrannical cliché of “binary choice” voting by these systems, efficacy is boosted. The guilt stranglehold which dominant parties apply to their voters, warning that a vote not exercised in their favor is effectively a vote for the opposing party, no longer has the same hold. Citizens can vote in a way that more accurately reflects the nuances of ideology, which is often not perfectly encapsulated by public policy. This also means third-parties have a greater chance of electoral success, meaning they’re more likely to field a candidate in more elections. Or, at least, that’s the thinking.

But the problem here is one of numbers. Ranked-choice voting requires a majority vote. Even with a plurality vote, parties are drawn towards coalition-building as a matter of mathematical pragmatism. A candidate likely needs to survive multiple rounds of voting to emerge the victorious majority winner under a ranked-choice system, which means slim margins are not a strategy for success.

In fact, the desire to coalition-build is not done away with by ranked-choice voting. As a nation, Australia has adopted preferential voting (ranked-choice by another name) in its elections, a result which has, according to WorldPost editor Rose O’Hara, led to the rise of centrist candidates who desire to appeal to the widest number of voters. O’Hara writes, “The major Australian parties have become “interest aggregators” in order to target the median voter — a strategy stemming from the idea that all voters can be located somewhere on a bell curve. In this system, there is also an intense period of bargaining for preferences just prior to elections, when parties make deals in order to be listed high on another’s preferences.”

Interest aggregation and representation are not compatible. Interest aggregation involves the same kind of gamesmanship as does the politics of a political structure with a party duopoly. The Republican and Democratic parties form coalitions in the interest of electoral success, meaning they prioritize the pursuit of policy goals based on how well they poll in the electorate. They adopt positions that are popular and noncontroversial in the interest of retaining power, which creates a hierarchy within their ranks. Subgroups, like conservatives and progressives, are often ignored because their positions do not receive as much popular support and polarize the electorate. This creates a paradox: though minority ideologies join major parties in order to have some representation on the national stage, they are alienated by members of the people with whom they have banded together in aid of the mutual goal of winning elections and legislating.

Ranked-choice voting cannot address this issue if the temptation is to appeal to the median voter. The same kind of hierarchical thinking emerges in the push to coalition build in a ranked-choice system as emerges in a first-past-the-post system. There is a specific drive to shape message: whether this is a desire to appeal to the median voter or whether this is the decision of a faction with few followers who bands together with another lesser-group to become a serious force of contention. It is not the neutral landscape, in which voters and parties are free from pressure to express themselves, that proponents of ranked-choice voting like to depict. The underlying limits of mathematics still creates the temptation to prioritize already popular views. Representation is not aided by such a system; minority groups are subject to the same practical pressures to ally themselves with centrists. The polity is constructed in terms of a zero-sum game, in which parties compete for the votes of groups with aggregated interests. The individual does not gain anymore sovereignty of choice; he is subject to the same pressure to vote for the frontrunner candidate rather than waste his vote on a minority candidate who will quickly be eliminated from the ranked-choice tabulation.


Also published on Medium.

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmail
Top
All content protected by copyright. The Politics of Discretion, 2016.
%d bloggers like this: