Exhausted choices (not to be confused with Exhausted ballots) are ballot rankings that are not considered in rounds of an election using some variation of Single Transferable Vote (STV) due to elimination in prior rounds. The ballot rankings get eliminated because all candidates who appear in the particular ballot/ranking were eliminated from the election's prior rounds of tallying. The practice of eliminating ballots from consideration is sometimes referred to as ballot exhaustion. Single-winner STV is sometimes referred to as "instant-runoff voting" or "ranked-choice voting".
Terminology[edit | edit source]
Single-winner STV is sometimes referred to as "instant-runoff voting" because of the way the ballot count simulates a series of runoffs, similar to an exhaustive ballot system, except that voter preferences do not change between rounds. It is also known as the alternative vote, transferable vote, ranked-choice voting (RCV), single-seat ranked-choice voting, or preferential vote.
A ballot becomes exhausted when a voter:
- Exhausted choice: a voter can list their preferences such that when applied to a runoff round it is for a candidate who is already eliminated - the vote is taken out of this round of the election
- Spoiled choices: if a voter cast a spoiled ballot which included an overvote or undervote as their top remaining choice
- Overvotes - example: voter accidentally ranks two candidates for the choice being evaluated
- Undervotes - example: voter ranks only one candidate on their ballot and that candidate is eliminated from the contest before the final round
"Ballot exhaustion" occurs when rankings on a voter’s ballot prevent their vote from being counted and determining the election’s end result. The selection in this particular race is then discarded, and does not influence the final outcome.
Exhausted ballot and exhausted ranking[edit | edit source]
"Ballot exhaustion" or "ranking exhaustion" occurs when a ballot/ranked-list is no longer countable in a tally as all of the candidates marked on the ballot are no longer in the contest. An "exhausted ballot" or "exhausted ranking" occurs when a voter ranks only candidates that are eliminated from a race.
Ballot exhaustion almost certainly occurs for some voters when all the candidates a voter ranked have lost, even though two or more other candidates remain in the race being tallied. This might happen because a voter undervoted (chosing not to rank all or many candidates) or because a voter ranked as many candidates as allowed on the ballot paper. Since such a ballot/ranking contains no rankings of a candidate still in the race, it is allowed to exhaust and is no longer included in the tally for winner. As such, it may be possible for the winner to have a majority of all the non-exhausted votes, but not a majority of total votes cast in the election.
For example, it's possible for 10% of rankings are exhausted in the first round of a tally, but the winner to only beat the loser by only 5% of the remaining votes If voters who had their ballots/rankings exhaust in this example election were permitted to choose again (e.g. in a top-two runoff election) a different winner might emerge with a clear majority of votes cast in the runoff. Advocates of RCV argue that voters should provide as complete of a ranking as possible to avoid this uncomfortable situation.
Spoiled ballots and spoiled choices[edit | edit source]
Because the ballot marking is more complex, there can be an increase in spoiled ballots.
According to FairVote, an "inactive" or "exhausted" ballot counts for candidates in the first round but not in the final round. Ballots become inactive for the following reasons:
- Election administrators limit voters to a certain number of rankings, such as three, and all of their ranked candidates are eliminated during the round-by-round count. This is known as "involuntary exhaustion".
- A partially-spoiled ballot due to one of two conditions:
- Overvote: The ballot is disqualified due to error, such as giving multiple candidates the same ranking.
- Undervote: The voter doesn’t rank all candidates, and all of their ranked candidates are eliminated during the round-by-round count. Also known as "voluntary abstention", this is the most common source of inactive votes.
In some jurisdictions, election officials ignore overvotes and undervotes in the first round of tabulation, skipping to the voter's second preference during the tabulation process. Overvotes and undervotes in the second and subsequent rounds of tabulation may also be ignored rather than treating them as spoiled ballots and discarding them entirely. In cases where a voter has ranked only candidates that did not make it to the final round of counting, it is only then that the voter's ballot is said to have been exhausted.
Other terminology[edit | edit source]
- inactive choices
- disqualified ballots
- discarded ballots
- spent ballots
- wasted votes
Example[edit | edit source]
Voters in an STV elections such the "instant-runoff voting (IRV)" elections (or "ranked-choice voting (RCV)" elections) held in Burlington, Vermont a few years ago rank candidates on a preferential ballot. The city of held a second mayoral election using instant-runoff voting on March 3, 2009, which was the second mayoral election since the city's 2005 approval of instant-runoff voting (IRV). A candidate from the Vermont Progressive Party (Bob Kiss) had won the first election under the system in 2006. In 2009, he was running for reelection. The official results of the 2009 election were as follows:
First round[edit | edit source]
In the first round, Dan Smith and Jason Simpson were eliminated, as well as the all of the write-in candidates. There were four ballots that did not have preferences listed, so they were placed in the "exhausted pile" to indicate that those ballots would play no active role in later rounds. Ballots for Smith, Simpson and the write-in candidates were transferred to the second preference on those ballots for the second round (or were moved into the "exhausted pile" in the second round).
|Kurt Wright||Republican Party||2,951||32.9%||Ballots preferring Kiss, Wright, and Montroll advance to the second round.|
|Bob Kiss||Progressive Party||2,585||28.8%|
|Andy Montroll||Democratic Party||
|Dan Smith||(independent)||1,306||14.5%||Ballots preferring Smith, Simpson and the write-in candidate were distributed to Kiss, Wright, and Montroll in the second round according to the first remaining preference on these ballots.|
|James Simpson||Green Party||35||0.4%|
Second round[edit | edit source]
In the second round, Simpson's, Smith's and the write-in candidates' ballots were transferred to Kiss, Wright, or Montroll (depending on the voter's greatest remaining preference). Montroll had the fewest first-remaining preferences, and thus was eliminated prior to the third round, with Montroll's ballots being distributed to Kiss and Wright according to the final remaining preference on ballots preferring Montroll.
|Kurt Wright||Republican Party||+343||3,294||36.7%||Ballots preferring Kiss and Wright advanced to the third round.|
|Bob Kiss||Progressive Party||+396||2,981||33.2%|
|Andy Montroll||Democratic Party||+491||2,554||28.4%||Ballots preferring Montroll were distributed to Kiss and Wright in the third round according to final remaining preference on these ballots.|
Third round[edit | edit source]
In the third round, Montroll's ballots from the second round were distributed to Kiss and Wright. Since many Montroll voters supported Kiss rather than Wright as their final remaining preference, Kiss pulled into the lead in the third round. Because 6.7% of voters didn't express a preference between Kiss and Wright (with their preferred candidate eliminated in prior rounds, and their ballots placed in the "exhausted pile"), Kiss prevailed over Wright with a plurality of 48.0%.
|Bob Kiss||Progressive Party||+1332||4,313||48.0%|
|Kurt Wright||Republican Party||+767||4,061||45.2%|
Background[edit | edit source]
STV-based systems in use in different countries vary both as to ballot design and to whether or not voters are obliged to provide a full list of preferences. In jurisdictions such as the Republic of Ireland and Northern Ireland voters are permitted to rank as many or as few candidates as they wish. Consequently, voters sometimes, for example, rank only the candidates of a single party, or of their most preferred parties. A minority of voters, especially if they do not fully understand the system, might even "bullet vote", only expressing a first preference. Allowing voters to rank only as many candidates as they wish grants them greater freedom, but can also lead to some voters ranking so few candidates that their vote eventually becomes "exhausted"; that is, at a certain point during the count it can no longer be transferred and therefore loses an opportunity to influence the result. (In First Past the Post elections, many, sometimes most, votes are disregarded, as there is no opportunity to mark back-up preferences. To the extent that voters mark back-up preferences and the back-up preferences consulted - many are not consulted even if marked - the portion of votes ignored under STV is less than under First Past The Post. Back-up preferences are not consulted if the vote is cast at the start for a candidate who wins in the end as the last seat is filled, or cast for a candidate who is eliminated at the end. They are also not used if they are marked for a candidate who has already been elected or eliminated.)
In Australia, voters are required to write a number beside every candidate, and the rate of spoiled ballots can be five times higher than plurality voting elections. Since Australia has compulsory voting, however, it is difficult to tell how many ballots are deliberately spoiled. Where complete rankings are not required, a ballot may become inactive if none of the ranked choices on that ballot advance to the next round.
Most jurisdictions with IRV do not require complete rankings and may use columns to indicate preference instead of numbers. In American elections with IRV, more than 99% of voters typically cast a valid ballot.
A 2015 study of four local U.S. elections that used IRV found that inactive ballots occurred often enough in each of them that the winner of each election did not receive a majority of votes cast in the first round. The rate of inactive ballots in each election ranged from a low of 9.6% to a high of 27.1%. As one point of comparison, the number of votes cast in the 190 regularly scheduled primary runoff elections for the U.S. House and U.S. Senate from 1994 to 2016 decreased from the initial primary on average by 39%, according to a 2016 study by FairVote.
Commentary[edit | edit source]
Opponents of ranked-choice voting (RCV) bring the concept of discarded ballots (both exhausted ballots and spoiled ballots) as an important drawback in RCV. In RCV, a voter's lesser-preferred candidates may be eliminated before their first choice, so that by the time their favorite is eliminated the vote may have nowhere to transfer to. On average in competitive RCV elections over 10% of ballots are exhausted or spoiled. In some cases, the eliminated candidate may have actually been the candidate preferred over all others, but because RCV doesn't count most of the rankings voters put down, it can fail to elect the most popular candidate.
RCV's algorithm doesn't count most of the rankings voters put down on their ballots. In ranked-choice voting, which of the voter's rankings will be counted and which will not depends on the order of elimination. As a result, it may not be safe for voters to vote for their favorite in RCV, just like with traditional first-past-the-post/choose-one voting. Worse, in some cases, voting your conscience can actually backfire, resulting in a worse outcome than if you hadn't voted at all in RCV. Opponents to RCV believe the system should count your vote and it should be able to make a difference, but in ranked-choice voting that's not necessarily the case.
STAR voting[edit | edit source]
What's the difference between an exhausted ballot in RCV and a vote of lesser-preference in STAR? An exhausted ballot in RCV is not counted in the deciding round, even if it could have made a difference. A vote of lesser preference for one of the two finalists (or no preference between candidates) is counted in STAR Voting's automatic runoff round is counted and the voter intent to support or oppose both finalists equally, was respected.
Voters in any system can choose to vote in a way that's not as effective as it could have been, but the STAR Voting system won't waste the vote of a voter who showed up and voted their conscience and your vote will never backfire. It's also next to impossible for an inexperienced voter to accidentally waste their vote in STAR Voting.
Order of elimination and incomplete tabulation. Compare these uncounted exhausted ballots in RCV to a vote of no-preference in STAR Voting, where a voter explicitly chose to score both finalists equally. These votes are counted and do make a difference to help advance the candidates who were more preferred. Allowing voters to give equal scores in STAR is the key to preventing spoiled ballots, and it's also key for eliminating vote-splitting between similar candidates and maintaining election accuracy in larger fields of candidates.
Proponents of RCV - Exhausted Choices[edit | edit source]
These occur when all the choices a voter has marked are eventually eliminated and their ballot has no active choices remaining. We should note that because of inactive ballots, the "majority" 50% in RCV can refer to a majority of active ballots, and not necessarily to 50% of the original number of ballots cast. We believe RCV works best when voters complete all of the available ranks and communicate their complete priorities. But voting is voluntary in the United States, and if a voter does not wish to make a rank, they are free not to - even if that means denying themselves a chance to make their 2nd, 3rd, 4th, etc. choice known and possibly decisive in an election.
A voter may not wish to provide any support to candidates whose policies they strongly object to - even when that support is only relevant once all of the candidates that voter prefers have been eliminated from the race. We disagree that an inactive ballot is necessarily a sign of a voter not understanding how to take advantage of all that RCV has to offer. If a local election is about an incumbent doing a good job or not, some voters might be highly invested in the incumbent, and may not have strong feelings about any of the challengers.
It's important to explain that not all RCV jurisdictions allow a rank for every candidate. Minneapolis, for example, often has 15 or more candidates for Mayor, but allows voters to mark only their 1st, 2nd and 3rd choices. New York City's Democratic mayoral primary in 2021 had 13 candidates but allowed voters only five ranks. So RCV elections in these locations are certain to have a higher proportion of inactive ballots compared to places where the RCV ballot offers a rank for every candidate. They are completely valid uses of RCV.
It's true that an exhausted ballot no longer makes a difference in an election. However, it's important to point out that the same thing also happens quite often in conventional elections.
Imagine three candidates in a conventional election. The polls closed at 7 pm and the first precincts reporting show the race with two candidates at 40% each, and one candidate is way behind with just 20%. The counting for the 40%/40% race will probably go long into the night, and the voters who supported the 20% candidate realize their candidate can't win, and their vote is not relevant to the ongoing count. Whatever criticisms there are for exhausted or inactive RCV ballots, they should also be applied to the situation for conventional plurality elections we just described. In any race with more than two candidates, not every vote can matter right up until the finish line.
The percentage of inactive/exhausted ballots can depend on how close an election is, how many candidates compete, and how many rounds of counting are necessary for a winner to be declared. An example of a high number of inactive ballots is the very large and competitive 2014 race for Oakland, CA Mayor. Sixteen candidates competed, and it took all 16 possible rounds of counting for a winner to be declared. After three rounds of counting, the number of inactive ballots was 0.01%. By round 10, they reached about 1%, 7% in round 14, 14% in round 15, and then jumped to 24% in the final and 16th round of counting. It was in that final round the incumbent mayor was eliminated, and it seems only half of her large number of (possibly overconfident) supporters filled in a 2nd rank. A more typical situation is the 2018 election for Mayor of San Francisco. Nine candidates competed, and the election required nine rounds of counting. After five rounds of counting, just 0.01% of ballots had become inactive/exhausted. After the final and 9th round, 8.4 % of the original set of ballots had run out of ranks.
On one hand, exhausted, inactive ballots are a fact of RCV.
But there is another side to that coin: RCV is designed to be as inclusive as possible in how it incorporates 2nd, 3rd, 4th choices, etc. of the supporters of defeated candidates. On balance, we believe RCV is much, much, more inclusive of voter preferences than it is exclusive.
See also[edit | edit source]
Links[edit | edit source]
- "Ballot (and voter) “exhaustion” under Instant Runoff Voting: An examination of four ranked-choice elections"
- Craig M. Burnett, Vladimir Kogan
- Journal: Electoral Studies, Volume 37, 2015, Pages 41-49, ISSN 0261-3794,
- Link1: https://doi.org/10.1016/j.electstud.2014.11.006
- Link2: https://cpb-us-w2.wpmucdn.com/u.osu.edu/dist/e/1083/files/2014/12/ElectoralStudies-2fupfhd.pdf
- "We analyze data taken from images of more than 600,000 ballots cast by voters in four recent local elections. We document a problem known as ballot “exhaustion,” which results in a substantial number of votes being discarded in each election. As a result of ballot exhaustion, the winner in all four of our cases receives less than a majority of the total votes cast, a finding that raises serious concerns about IRV and challenges a key argument made by the system's proponents."
- "Second, IRV does not ensure that the winning candidate will have received a majority of all votes cast, only a majority of all valid votes in the final round of tallying. Thus, it is possible that the winning candidate will fall short of an actual majority when a substantial number of ballots are eliminated, or “exhausted,” during the vote redistribution process. Third, and related to the previous point, there is some probability that a voter's ballot will become exhausted, eliminating their influence over the final outcome."
- "If at any point the voter did not rank a next choice (assuming her most favored choice or choices are eliminated), or all of the choices on the voter's ballot have been eliminated, the ballot is “exhausted” d meaning that it is excluded from future vote redistributions, and it does not affect the final outcome of the election. The ballot, in essence, is discarded. The process ends once a candidate receives a majority of the remaining valid votes."
References[edit | edit source]
- ↑ "Ballot exhaustion". Ballotpedia. Retrieved 2023-02-27.
- ↑ User:RobLa quoted oldid 1141090457 of [[w:Instant-runoff voting|]]: https://en.wikipedia.org/w/index.php?title=Instant-runoff_voting&oldid=1141090457
- ↑ "Second Report: Election of a Speaker". House of Commons Select Committee on Procedure. 15 February 2001. Retrieved 18 February 2008.
- ↑ https://fairvote.org/our-reforms/ranked-choice-voting-information/#_13-what-are-inactive-or-exhausted-ballots
- ↑ Quoting w:Issues affecting the single transferable vote https://en.wikipedia.org/w/index.php?title=Issues_affecting_the_single_transferable_vote&oldid=1137726277
- ↑ 4. How did this change to IRV come about? Over 64% of Burlington voters voted in favor of the IRV Charter amendment in March, 2005, and it went into effect on May 12, 2005, when the governor signed the ratification bill, H.505, which had been passed by both the House and Senate.
- ↑ "Mayor Bob Kiss". City of Burlington. Archived from the original on 2007-11-29. Retrieved 2007-11-16.
- ↑ "ChoicePlus Pro 2009 Burlington Mayor Round Detail Report". 2011-07-25. Archived from the original on 2011-07-25. Retrieved 2018-01-03.
- ↑ "ChoicePlus Pro 2009 Burlington Mayor Round 4 Report". March 3, 2009. Archived from the original on 2011-07-25. Retrieved 2011-02-28.
- ↑ "Voting in the House of Representatives". Australian Electoral Commission. 28 June 2016. Retrieved 9 December 2018.
- ↑ "Busting the Myths of AV". No2av.org. 25 October 2010. Archived from the original on 8 February 2011. Retrieved 17 April 2011.
- ↑ "Informal Voting – Two Ways of Allowing More Votes to Count". ABC Elections. 28 February 2011. Retrieved 15 August 2011.
- ↑ "Instant Runoff Voting and Its Impact on Racial Minorities" (PDF). New America Foundation. 1 August 2008. Archived from the original (PDF) on 15 June 2011. Retrieved 15 August 2011.
- ↑ Burnett, Craig M.; Kogan, Vladimir (March 2015). "Ballot (and voter) 'exhaustion' under Instant Runoff Voting: An examination of four ranked-choice elections". Electoral Studies. 37: 41–49. doi:10.1016/j.electstud.2014.11.006. Unknown parameter
- ↑ "Box".
Other external websites[edit | edit source]
- Ballot (and voter) “exhaustion” under Instant Runoff Voting: An examination of four ranked-choice elections