There is no situation in which a winner-take-all election can be both competitive and representative.
Ranked Choice Voting, Defined
Ranked-choice voting is a form of proportional representation that allows voters to rank candidates in order of preference. In a single-winner district, it is also known as instant runoff voting. The process is simple:
- Voters rank candidates in order of preference (with their favored candidate ranked first, their second favorite candidate ranked second, and so on).
- The votes are counted. If any candidate received more than half of the first-ranked votes, she wins. If not, voting proceeds to the second round (this is where the instant runoff comes into play).
- The candidate who received the fewest number of first-choice votes is eliminated. The voters who selected that candidate have their votes transferred to their second choice candidate.
- This process continues until one candidate has more than half of the first choice votes.
Instant runoff voting is easy to illustrate. Consider the following hypothetical scenario, in which 100 people must vote for their favorite color (Red, Blue, Orange, Green, or Purple). Each voter would fill out a ballot, which might look something like this (Figure 9):
After the first round of voting, the results might look like this (Figure 10):
In this scenario, Red has the most votes, with 313, but since the threshold is 501, the election isn’t called yet. We have to proceed to the second round (Figure 10). Observe that Purple, with the fewest votes, has been eliminated. Everyone who voted for Purple will have their second choice distributed to the remaining four candidates. In Figure 11, the 109 votes cast for Purple are distributed in the following manner:
After the second round of voting, Red has gained 26 votes, putting it at 339, which is still not enough to push it to the threshold. Orange has gained a significant number of votes, putting it up in the rankings. We move on to the third round of voting (Figure 12), with the candidate with the fewest votes, this time Green, being eliminated. Everyone who voted for Green as their first choice will have their second choice vote distributed to the remaining three candidates.
After Round Three, Blue has been eliminated, but no candidate has met the threshold for victory. We proceed to Round Four (Figure 13).
In the fourth round, we have a winner—Red, with 623 votes. Instant Runoff Voting, for single-office races, is preferable to an extended primary season followed by a lengthy general election period. Consider the recent special election in Georgia’s Sixth Congressional Race. The primary election was held on April 18. No candidate received 50% + 1 of the vote, so a general election had to be called for June 20, 61 full days later. This length of time costs the state election board money, it costs the candidates money, and it drains the attention of the voting public, who have to endure two additional months of campaigning. Had Georgia implemented Instant Runoff Voting (as proposed by Senator Josh McKoon in SB32), the winner would have been declared immediately, saving taxpayers time and money.
Ranked-Choice Voting With Multiple Candidates
In a district with multiple candidates (the idea behind the Fair Representation Act), the process is similar, although somewhat more complicated. It is administered almost identically to the above single-candidate instant runoff, but with more math (this style of voting is also referred to as Single-Transferable Voting).
Single-Transferable Voting (or STV) solves the problem of wasted votes. Under a normal, winner-take-all system, any votes beyond those a candidate needs to win are wasted; that is, the voter could have stayed home and the outcome would not have been affected. For a truly representative democracy, everyone needs to feel that their vote counts, which is where STV comes in.
In this system, multiple candidates can run for multiple seats. In our hypothetical situation (Figure 14), we have six candidates (Washington, Jefferson, Adams, Hamilton, Franklin, and Henry) running for three seats. After the first round of voting, the results come in:
First, we have to determine the threshold for victory. This threshold can be calculated several ways, but the most common (used in elections worldwide), is the Droop Quota.
The Droop Quota (named for Henry Droop, a 19th-century English mathematician), can be expressed as
(T / N + 1) + 1
where T is the total number of valid votes cast (in this case, 1500) and N is the number of seats to be filled (in this case, 3). This formula works in such a way that when there is only one seat to be filled, the threshold is 50% + 1. In the following scenario (with three seats to be filled and 1,500 votes cast), the threshold is expressed as
(1500 / 3 + 1) + 1 = 376
or 25% of the vote. Based on that formula, we know that any candidate who receives more than 376 votes wins a seat. Washington, with 382 votes, wins automatically. But none of the other candidates have met the quota, which means we have to proceed to the second round.
Here is where the strength of the Single-Transferable Vote is demonstrated. The surplus votes (votes cast for a winning candidate beyond the threshold for victory) are distributed down to the voters’ second choices. After the surplus is calculated, the candidate with the fewest votes is eliminated, and voters who voted for that candidate see their vote distributed to their second choice.
This process, while complicated at first glance, is both intuitive and reasonable. Voters can vote their conscience, confident that even if they vote for a candidate who they do not believe can win, their vote will still “count.” In the same way, no voter will “waste” a vote by honestly ranking a popular candidate.
Now we proceed to the second round (Figure 15), where votes for the winner (Washington) are distributed down.
In the second round, Washington sees 2% of his vote (the amount he exceeded the quota) distributed to the second choice candidates of his voters. Two of his voters had listed Jefferson as their second choice; and one each listed Adams, Hamilton, Franklin, and Henry as their second choice. Their votes are distributed accordingly. Henry, with the fewest number of votes, is eliminated; his 153 votes are then distributed to the second choices of his voters (Figure 14).
After Henry’s votes are distributed, Hamilton is left with the fewest votes, but no one has managed to reach the 376-vote threshold. Hamilton, therefore, will be eliminated, and his 248 votes will be transferred to the second choice of his voters (Figure 16).
In Round Four, Jefferson passes the threshold, receiving 412 votes. Since he only needs 376 votes to win a seat, his surplus votes will be passed down to the remaining candidates (Figure 17).
Neither of the remaining candidates have reached the 376-vote threshold, but Adams, with the fewest number of votes, is eliminated, leaving Franklin the winner of the third seat.
The advantages of Fair Representation Voting over single-district, winner-take-all elections are clear. In a normal, single-district, winner-take-all election, the most mainstream candidate (Washington) would have won with 25.46% of the vote, leaving almost three-fourths of the electorate represented by a candidate for whom they did not vote. In a multi-winner district where the candidates who received a plurality of the votes were elected, 60% of electors could point to a candidate they helped elect. But with Fair Representation Voting, a full 75% of voters are represented.
Fair Representation In Georgia
Enacting Fair Representation Voting in Georgia would require consolidating the 180 General Assembly districts into a much smaller number of multi-winner districts (such as 60 3-winner districts or 36 5-winner districts), as well as consolidating the 56 Senate districts into multi-winner districts (such as 10 5-winner districts and 1 6-member district, or 17 3-winner districts and 1 4-member district). According to Burden, et al, Ranked Choice Voting in 5-Member Districts and Ranked Choice Voting in 3-Member Districts “create better systems of representation for general election voters.” That same report found that Ranked Choice Voting in 3- and 5-winner districts would provide better outcomes in the areas of Legislative Functionality, Electoral Accountability, Voter Engagement, and the Openness of the Voting Process.
It is difficult to overstate the potential benefits of Fair Representation Reform. Some possible outcomes include “serious potential for more third parties, independents, and independent-minded major party candidates (including ones with new geographic bases, such as rural Democrats and urban Republicans) to hold traditional major party candidates accountable and potentially win seats.” Additionally, “scholars judged that the combination of RCV with multi-winner districts would improve legislative functionality, increase the influence of third party candidates, increase independent and third party voter turnout, and increase the number of women and minorities. They thought that the shared nature of representation in multi-winner districts had the potential to change legislator behavior for the better and result in the passage of more representative legislation.”
Can It Be Done?
The easiest answer to this question is “yes.” Ranked Choice Voting (in the form of Instant Choice Voting) is used at all levels of government all over the world. In the United States, states from Colorado to Minnesota to California to Maine use Ranked Choice Voting in state and local elections. In Arkansas, Alabama, Louisiana, South Carolina, and Illinois, Ranked Choice Voting is used for overseas ballots. And this is far from a new phenomenon; Cambridge, Massachusetts has used Ranked Choice Voting in its full multi-winner district form to elect its city council since 1940.
Internationally, Ranked Choice Voting in multi-winner districts is used to varying degrees to great effect in elections in countries like Ireland, Australia, India, New Zealand, Hong Kong, and Scotland. There are slight variations in the techniques used—for example, Australia uses a different quota for determining the victory threshold—but the premise is identical.
Quite simply, Ranked Choice Voting (in either Instant Runoff or Single Transferable Vote) is an electoral winner. The advantages, in terms of administration, politics, and policy, are undeniable.