Instant-runoff voting

Example instant-runoff voting ballot

Instant-runoff voting (IRV), also known as the: alternative vote (AV), transferable vote, ranked choice voting, or preferential voting is an electoral system used to elect a single winner from a field of more than two candidates. It is a preferential voting system in which voters rank the candidates in order of preference rather than voting for a single candidate.

Ballots are initially distributed based on each elector's first preference. If a candidate secures more than half of votes cast, that candidate wins. Otherwise, the candidate with the fewest votes is eliminated. Ballots assigned to the eliminated candidate are recounted and added to the totals of the remaining candidates based on who is ranked next on each ballot. This process continues until one candidate wins by obtaining more than half the votes.

IRV has the effect of avoiding split votes when multiple candidates earn support from like-minded voters. For example, suppose there are two similar candidates A & B, and a third opposing candidate C, with raw popularity of 35%, 25% and 40% respectively. In a plurality voting system, candidate C may win with 40% of the votes, even though 60% of electors prefer either A or B. Alternatively, voters are pressured to choose the seemingly stronger candidate of either A or B, despite personal preference for the other, in order to help ensure the defeat of C. It is often the resulting situation that candidate A or B would never get to ballot, whereas voters would be presented a two candidate choice. With IRV, the electors backing B as their first choice can allocate their preferences as #1 for B and #2 for A, which means A will win despite the split vote in first choices.

Instant-runoff voting is used to elect members of the Australian House of Representatives and most Australian State Governments, the President of India, members of legislative councils in India, the President of Ireland, and the parliament in Papua New Guinea. It is also used in Northern Ireland by-elections and for electing hereditary peers for the British House of Lords.

The system is also used in local elections around the world: to elect the mayor in cities such as London in the United Kingdom (in the variant known as supplementary vote) and Dunedin and Wellington in New Zealand. Variations of instant-runoff voting are employed by several jurisdictions in the United States, including San Francisco, San Leandro, and Oakland in California; Portland, Maine; Minneapolis and Saint Paul in Minnesota. The single transferable vote, a multi-seat form of IRV, is used in Cambridge, Massachusetts.

It is used to elect the leaders of the Labour Party and the Liberal Democrats in the United Kingdom and was used in elections in 2013 for the leader of the Liberal Party of Canada and in Canada's New Democratic Party leadership election, 2012.

Many private associations also use IRV, including the Hugo Awards for science fiction and the Academy of Motion Picture Arts and Sciences in selection of the Oscar for best picture. IRV is described in Robert's Rules of Order Newly Revised, 10th edition.

Terminology

Instant-runoff voting derives its name from how ballot-count simulates a series of two-round system runoffs except that voter preference do not change between rounds.

Britons and Canadians generally call IRV the "Alternative Vote" (AV). Australians, who use IRV for most single winner elections call IRV "preferential voting," as does Robert's Rules of Order. Americans in San Francisco, California, Portland, Maine and Minneapolis, Minnesota call IRV "ranked choice voting". IRV occasionally is referred to as Ware's method after its inventor, American William Robert Ware.

North Carolina law uses "instant runoff" to describe the contingent vote or "batch elimination" form of IRV in one-seat elections. A single second round of counting produces the top two candidates for a runoff election. Election officials in Hendersonville, North Carolina use "instant runoff" to describe a multi-seat election system that simulates in a single round of voting their previous system of multi-seat runoffs. State law in South Carolina and Arkansas use "instant runoff" to describe the practice of having certain categories of absentee voters cast ranked choice ballots before the first round of a runoff and counting those ballots in any subsequent runoff elections.

When the single transferable vote (STV) system is applied to a single-winner election it becomes IRV. Some Irish observers mistakenly call IRV "proportional representation" based on the fact that the same ballot form is used to elect its president by IRV and parliamentary seats by STV, but IRV is a winner-take-all election method.

History

Main article: History and use of instant-runoff voting

Instant-runoff voting was devised in 1871 by American architect William Robert Ware, although it is, in effect, a special case of the single transferable vote system, which emerged independently in the 1850s. Unlike the single transferable vote in multi-seat elections, however, the only ballot transfers are from backers of candidates who have been eliminated.

The first known use of an IRV-like system in a governmental election was in 1893 in an election for the colonial government of Queensland, in Australia. The variant used for this election was a "contingent vote". IRV in its true form was first used in 1908 in a State election in Western Australia.

IRV was introduced nationally in Australia in 1918 after the Swan by-election, in response to the rise of the conservative Country Party, representing small farmers. The Country Party split the anti-Labor vote in conservative country areas, allowing Labor candidates to win on a minority vote. The conservative government of Billy Hughes introduced preferential voting as a means of allowing competition between the two conservative parties without putting seats at risk. It was first used at the Corangamite by-election on 14 December 1918. Thomas Hare and Andrew Inglis Clark had previously introduced it in the Tasmanian House of Assembly.

Election procedure

Process

Flowchart for counting IRV votes

In instant-runoff voting, as with other ranked election methods, each voter ranks the list of candidates in order of preference. Under a common ballot layout, the voter marks a '1' beside the most preferred candidate, a '2' beside the second-most preferred, and so forth, in ascending order. This is shown in the example Australian ballot above.

The mechanics of the process are the same regardless of how many candidates the voter ranks, and how many are left unranked. In some implementations, the voter ranks as many or as few choices as they wish, while in other implementations the voter is required to rank either all candidates, or a prescribed number of them.

Optical scan IRV ballot

In the initial count, the first preference of each voter is counted and used to order the candidates. Each first preference counts as one vote for the appropriate candidate. Once all the first preferences are counted, if one candidate holds a majority, that candidate wins. Otherwise the candidate who holds the fewest first preferences is eliminated. If there is an exact tie for last place in numbers of votes, tie-breaking rules determine which candidate to eliminate. Some jurisdictions eliminate all low-ranking candidates simultaneously whose combined number of votes is fewer than the number of votes received by the lowest remaining candidates.

Ballots assigned to eliminated candidates are recounted and assigned to one of the remaining candidates based on the next preference on each ballot. The process repeats until one candidate achieves a majority of votes cast for continuing candidates. Ballots that 'exhaust' all their preferences (all its ranked candidates are eliminated) are set aside.

In Australian elections the allocation of preferences is performed efficiently at the polling booth by having the returning officer pre-declare the two likely winners. (In the event that the returning officer is wrong the votes need to be recounted.)

Ballots

As seen above, voters in an IRV election rank candidates on a preferential ballot. IRV systems in use in different countries vary both as to ballot design and as to whether or not voters are obliged to provide a full list of preferences. In elections such as those for the President of Ireland and the New South Wales Legislative Assembly, voters are permitted to rank as many (or as few) candidates as they wish. This is known in Australia as optional preferential voting.

Under optional preferential voting, voters may make only a first choice, known as "bullet voting". Allowing voters to rank only as many candidates as they wish may better reflect their preferences, but may result in ballot exhaustion (where all the voters' preferences are eliminated before a candidate is elected).

One IRV variant requires voters to express an order of preference for every candidate and thus they consider ballots that do not contain a complete ordering of all candidates to be spoilt. In Australia, this variant is known as 'full preferential voting'. This can become burdensome in elections with many candidates and can lead to 'donkey voting' in which the voter simply chooses candidates at random or in top-to-bottom order. [This variant is used in some Australian federal elections and some state elections].

Candidate order on the ballot paper

The common way to list candidates on a ballot paper is alphabetically or by random lot. In some cases, candidates may also be grouped by political party. Alternatively, Robson Rotation involves randomly changing candidate order for each print run.

Party strategies

Where preferential voting is used for the election of an assembly or council, parties and candidates often advise their supporters on their lower preferences, especially in Australia where a voter must rank all candidates to cast a valid ballot. This can lead to "preference deals", a form of pre-election bargaining, in which smaller parties agree to direct their voters in return for support from the winning party on issues critical to the small party. However, this relies on the assumption that supporters of a minor party will mark preferences for another party based on the advice that they have been given.

Counting logistics

Most IRV elections historically have been tallied by hand, including in elections to Australia's House of Representatives and most state governments. In the modern era, voting equipment can be used to administer the count either partially or fully.

In Australia, the returning officer now usually declares the two candidates that are most likely to win each seat. The votes are always counted by hand at the polling booth monitored by scrutineers from each candidate. The first part of the count is to record the first choice for all candidates. Votes for candidates other than the two likely winners are then allocated to them in a second pass. The whole process of counting the votes by hand and allocating preferences is typically completed within two hours on election night at a cost of $7.68 per elector in 2010 to run the entire election.

(The declaration by the returning officer is simply to optimize the counting process. In the unlikely event that the returning officer is wrong and a third candidate wins, then the votes would simply have to be counted a third time.)

Ireland in its presidential elections has several dozen counting centers around the nation. Each center reports its totals and receives instructions from the central office about which candidate or candidates to eliminate in the next round of counting based on which candidate is in last place. The count typically is completed the day after the election, as in 1997.

In the United States, some Californian cities, e.g. Oakland and San Francisco, administer IRV elections on voting machines, with optical scanning machines recording preferences and software tallying the IRV algorithm. Cary, North Carolina's pilot program in 2007 tallied first choices on optical scan equipment at the polls and then used a central hand-count for the IRV tally. Portland, Maine in 2011 was due to use its usual voting machines to tally first choice at the polls, then a central scan with different equipment if an IRV tally was necessary.

Examples

Some examples of IRV elections are given below. The first two (fictional elections) demonstrate the principle of IRV. The others offer examples of the results of real elections.

Five voters, three candidates

A simple example is provided in the accompanying table. Three candidates are running for election, Bob, Bill and Sue. There are five voters, "a" through "e". The voters each have one vote. They rank the candidates first, second and third in the order they prefer them. To win, a candidate must have a majority of vote; that is, three or more.

In Round 1, the first-choice rankings are tallied, with the results that Bob and Sue both have two votes and Bill has one. No candidate has a majority, so a second "instant runoff" round is required. Since Bill is in bottom place, he is eliminated. The ballot ranking him first is added to the totals of the candidate listed second. This results in the Round 2 votes as seen below. This gives Sue 3 votes, which is a majority.

	Round 1						Round 2
Candidate	a	b	c	d	e	Votes	a	b	c	d	e	Votes
Bob	1	2	3	1	2	2	1	2	2	1	2	2
Sue	3	1	2	3	1	2	2	1	1	2	1	3
Bill	2	3	1	2	3	1

Tennessee capital election

Most instant-runoff voting elections are won by the candidate who leads in first-choice rankings. In such cases, IRV chooses the same winner as first-past-the-post voting. Some IRV elections are won by a candidate who finishes second after the first-round count. In this case, IRV chooses the same winner as a two-round system if all voters were to vote again and maintain their same preferences. A candidate may also win who is in third place or lower after the first count, but gains majority support in the final round. In such cases, IRV would choose the same winner as a multi-round system that eliminated the last-place candidate before each new vote, assuming all voters kept voting and maintained their same preferences. Here is an example of this last case.

Suppose that Tennessee is holding an election on the location of its capital. The population is concentrated around four major cities. All voters want the capital to be as close to them as possible. The options are:

• Memphis, the largest city, but far from the others (42% of voters)
• Nashville, near the center of the state (26% of voters)
• Chattanooga, somewhat east (15% of voters)
• Knoxville, far to the northeast (17% of voters)

The preferences of each region's voters are:

42% of voters Far-West	26% of voters Center	15% of voters Center-East	17% of voters Far-East
Memphis Nashville Chattanooga Knoxville	Nashville Chattanooga Knoxville Memphis	Chattanooga Knoxville Nashville Memphis	Knoxville Chattanooga Nashville Memphis

In the first round no city gets a majority:

Votes in round/ City Choice	1st
Memphis	42%
Nashville	26%
Knoxville	17%
Chattanooga	15%

If one of the cities had achieved a majority vote (more than half), the election would end there. If this were a first-past-the-post election, Memphis would win because it received the most votes. But IRV does not allow a candidate to win on the first round without having an absolute majority of the vote. While 42% of the electorate voted for Memphis, 58% of the electorate voted against Memphis in this first round.

So we move to the second round of tabulation to determine which of the front-running cities had broader support. Chattanooga received the lowest number of votes in the first round, so it is eliminated. The ballots that listed Chattanooga as “first-choice” are added to the totals of the second-choice selection on each ballot. Everything else stays the same.

Chattanooga’s 15% of the total votes are added to the second choices selected by the voters for whom that city was first-choice (in this example Knoxville):

Votes in round/ City Choice	1st	2nd
Memphis	42%	42%
Nashville	26%	26%
Knoxville	17%	32%
Chattanooga	15%

In the first round, Memphis was first, Nashville was second and Knoxville was third. With Chattanooga eliminated and its votes redistributed, the second round finds Memphis still in first place, followed by Knoxville in second and Nashville has moved down to third place. No city yet has secured a majority of votes, so we move to the third round with the elimination of Nashville, and it becomes a contest between Memphis and Knoxville.

As in the second round with Chattanooga, all of the ballots currently counting for Nashville are added to the totals of Memphis or Knoxville based on which city is ranked next on that ballot. In this example the second-choice of the Nashville voters is Chattanooga, which is already eliminated. Therefore the votes are added to their third-choice: Knoxville.

The third round of tabulation yields the following result:

Votes in round/ City Choice	1st	2nd	3rd
Memphis	42%	42%	42%
Nashville	26%	26%
Knoxville	17%	32%	58%
Chattanooga	15%

Result: Knoxville, which was running third in the first tabulation, has moved up from behind to take first place in the third and final round. The winner of the election is Knoxville. However, if 6% of voters in Memphis were to put Nashville first, the winner would be Nashville, a preferable outcome for voters in Memphis. This is an example of potential tactical voting, though one that would be difficult for voters to carry out in practice. Also, if 17% of voters in Memphis were to stay away from voting, the winner would be Nashville. This is an example of IRV failing the participation criterion.

For comparison, note that traditional first-past-the-post voting would elect Memphis, even though most citizens consider it the worst choice, because 42% is larger than any other single city. As Nashville is a Condorcet winner, Condorcet methods would elect Nashville. A two-round system would have a runoff between Memphis and Nashville where Nashville would win, too.

2006 Burlington mayoral election

See also: Instant-runoff voting in the United States § 2006 Burlington results

Candidate	Round 1		Round 2
Bob Kiss	3,809	(38.9%)	4,761	(48.6%)
Hinda Miller	3,106	(31.7%)	3,986	(40.7%)
Kevin Curley	2,609	(26.7%)	—
Other	254	(2.6%)	—
Exhausted ballots	10	(0.1%)	1,041	(10.5%)
Total	9,778	(100%)	9,778	(100%)

In 2006, the U.S. city of Burlington, Vermont, held a mayoral election using instant-runoff voting. Progressive Bob Kiss won in two rounds with 48.6% of the second round ballots, defeating Democrat Hinda Miller who achieved 40.7%. 10.6% (1,031) of the ballots were exhausted before the final round, because those voters (largely backers of Republican candidate Kevin Curley) offered no preference between the final two candidates, Miller and Kiss.

After the first round, all but two candidates were eliminated, as their combined vote total (2,863) was less than Miller's, so that none could pull ahead of Miller, even by receiving every vote from the other minor candidates. The votes for these candidates were recounted and redistributed between Kiss and Miller. After the second round count, Kiss was declared the winner as he had obtained a majority (54.4%) of the remaining unexhausted ballots.

1990 Irish presidential election

See also: Irish presidential election, 1990

Irish presidential election, 1990
Candidate	Round 1		Round 2
Mary Robinson	612,265	(38.9%)	817,830	(51.6%)
Brian Lenihan	694,484	(43.8%)	731,273	(46.2%)
Austin Currie	267,902	(16.9%)	—
Exhausted ballots	9,444	(0.6%)	34,992	(2.2%)
Total	1,584,095	(100%)	1,584,095	(100%)

The result of the 1990 Irish presidential election provides an example of how instant-runoff voting can produce a different result than first-past-the-post voting. The three candidates were Brian Lenihan of the traditionally dominant Fianna Fáil party, Austin Currie of Fine Gael, and Mary Robinson, nominated by the Labour Party and the Worker's Party. After the first round, Lenihan had the largest share of the first-choice rankings (and hence would have won a first-past-the-post vote), but no candidate attained the necessary majority. Currie was eliminated and his votes reassigned to the next choice ranked on each ballot; in this process, Robinson received 82% of Currie's votes, thereby overtaking Lenihan.

Voting system criteria

Scholars rate voting systems using mathematically derived voting system criteria, which describe desirable features of a system. No ranked preference method can meet all of the criteria, because some of them are mutually exclusive, as shown by statements such as Arrow's impossibility theorem and the Gibbard-Satterthwaite theorem.

Many of the mathematical criteria by which voting systems are compared were formulated for voters with ordinal preferences. If voters vote according to the same ordinal preferences in both rounds, criteria can be applied to two-round systems of runoffs, and in that case, each of the criteria failed by IRV is also failed by the two-round system as they relate to automatic elimination of trailing candidates. Partial results exist for other models of voter behavior in the two-round system: see the two-round system article's criterion compliance section for more information.

The criteria that IRV meets, and those that it does not, are listed below.

Majority criterion

Y The majority criterion states that "if one candidate is preferred by an absolute majority of voters, then that candidate must win". IRV meets this criterion.

Mutual majority criterion

Y The mutual majority criterion states that "if an absolute majority of voters prefer every member of a group of candidates to every candidate not in that group, then one of the preferred group must win". IRV meets this criterion.

Later-no-harm criterion

Y The later-no-harm criterion states that "if a voter alters the order of candidates lower in his/her preference (e.g. swapping the second and third preferences), then that does not affect the chances of the most preferred candidate being elected". IRV meets this criterion.

Resolvability criterion

Y The resolvability criterion states that "the probability of an exact tie must diminish as more votes are cast". IRV meets this criterion.

Condorcet winner criterion

N The Condorcet winner criterion states that "if a candidate would win a head-to-head competition against every other candidate, then that candidate must win the overall election". It is incompatible with the later-no-harm criterion, so IRV does not meet this criterion.

IRV is more likely to elect the Condorcet winner than plurality voting and traditional runoff elections. The California cities of Oakland, San Francisco and San Leandro in 2010 provide an example; there were a total of four elections in which the plurality voting leader in first choice rankings was defeated, and in each case the IRV winner was the Condorcet winner, including a San Francisco election in which the IRV winner was in third place in first choice rankings.

Condorcet loser criterion

Y The Condorcet loser criterion states that "if a candidate would lose a head-to-head competition against every other candidate, then that candidate must not win the overall election". IRV meets this criterion.

Consistency criterion

N The consistency criterion states that if dividing the electorate into two groups and running the same election separately with each group returns the same result for both groups, then the election over the whole electorate should return this result. IRV, like all preferential voting systems which are not positional, does not meet this criterion.

Monotonicity criterion

Further information: Monotonicity criterion § Instant-runoff voting and the Two-round system are not monotonic

N The monotonicity criterion states that "a voter can't harm a candidate's chances of winning by voting that candidate higher, or help a candidate by voting that candidate lower, while keeping the relative order of all the other candidates equal." IRV does not meet this criterion. Allard claims failure is unlikely, at a less than 0.03% chance per election. Some critics argue in turn that Allard's calculations are wrong and the probability of monotonicity failure is much greater, at 14.5% under the impartial culture election model in the three-candidate case, or 7-10% in the case of a left-right spectrum. Lepelly et al. find a 2%-5% probability of monotonicity failure under the same election model as Allard.

Participation criterion

N The participation criterion states that "the best way to help a candidate win must not be to abstain". IRV does not meet this criterion: in some cases, the voter's preferred candidate can be best helped if the voter does not vote at all. Depankar Ray finds a 50% probability that, when IRV elects a different candidate than Plurality, some voters would have been better off not showing up.

Reversal symmetry criterion

N The reversal symmetry criterion states that "if candidate A is the unique winner, and each voter's individual preferences are inverted, then A must not be elected". IRV does not meet this criterion: it is possible to construct an election where reversing the order of every ballot paper does not alter the final winner.

Independence of irrelevant alternatives criterion

N The independence of irrelevant alternatives criterion states that "the election outcome remains the same even if a candidate who cannot win decides to run." IRV does not meet this criterion; in the general case, instant-runoff voting can be susceptible to strategic nomination: whether or not a candidate decides to run at all can affect the result even if the new candidate cannot themselves win. This is much less likely to happen than under plurality.

Independence of clones criterion

Y The independence of clones criterion states that "the election outcome remains the same even if an identical candidate who is equally preferred decides to run." IRV meets this criterion.

Comparison to other voting systems

Comparison of mechanics

Instant-runoff voting is one of many ranked ballot systems. For example, the elimination of the candidate with the most last-place rankings, rather than the one with the fewest first-place rankings, is called Coombs' method, and universal assignment of numerical values to each rank is used in the Borda count method. A chart in the article on the Schulze method compares various ranked ballot systems.

Comparison to first-past-the-post

At the most recent Australian federal election, in September 2013, 135 out of the 150 House of Representatives seats (or 90 percent) were won by the candidate who led on first preferences. The other 15 seats (10 percent) were won by the candidate who placed second on first preferences.

Resistance to tactical voting

The Gibbard–Satterthwaite theorem demonstrates that no voting system using only the preference rankings of the voters can be entirely immune from tactical voting unless it is dictatorial (there is only one person who is able to choose the winner) or incorporates an element of chance. This implies that IRV is susceptible to tactical voting in some circumstances.

Nonetheless, IRV is considered one of the less-manipulable voting systems, with theorist Nicolaus Tideman noting that, "alternative vote is quite resistant to strategy" and Australian political analyst Antony Green dismissing suggestions of tactical voting. James Green-Armytage finds the alternative vote to be second most resistant to tactical voting among the methods tested, only beaten by a class of AV-Condorcet hybrids, although the alternative vote resists strategic withdrawal by candidates less well.

By not meeting the monotonicity, Condorcet winner, and participation criteria, IRV permits forms of tactical voting when voters have sufficient information about other voters' preferences, such as from accurate pre-election polling. FairVote mentions that monotonicity failure can lead to situations where "Having more voters rank [a] candidate first, can cause [them] to switch from being a winner to being a loser." That assessment is accurate, although it only happens in particular situations. The change in lower candidates is important: whether votes are shifted to the leading candidate, shifted to a fringe candidate, or discarded altogether is of no importance.

Tactical voting in IRV seeks to alter the order of eliminations in early rounds, to ensure that the original winner is challenged by a stronger opponent in the final round. For example, in a three-party election where voters for both the left and right prefer the centrist candidate to stop the "enemy" candidate winning, those voters who care more about defeating the "enemy" than electing their own candidate may cast a tactical first preference vote for the centrist candidate.

The 2009 mayoral election in Burlington, Vermont provides an example in which strategy theoretically could have worked but would have been unlikely in practice. In that election, most supporters of the candidate who came in second (a Republican who led in first choices) preferred the Condorcet winner, a Democrat, to the IRV winner, the Progressive Party nominee. If about 20% of the backers of the Republican candidate had insincerely raised the Democrat from their second choice to their first, the Republican would have dropped from first to third in first choices, and the Democrat would then have won the instant runoff. But given that the Republican was a strong candidate who nearly won in the instant runoff, few of his backers would have risked giving up on his candidacy based on a chance, unknown before the fact, to elect the compromise Condorcet winner.

Spoiler effect

Main article: Spoiler effect

The spoiler effect is where two or more politically similar candidates divide the vote for the more popular end of the political spectrum. That is, each receives fewer votes than a single opponent on the unpopular end of the spectrum who is disliked by the majority of voters but who wins from the advantage that, on that unpopular side, he or she is unopposed.

Proponents of IRV note that by reducing the spoiler effect, IRV makes it safe to vote honestly for marginal parties, and so discourages tactical voting: under a plurality system, voters who sympathize most strongly with a marginal candidate are strongly encouraged to instead vote for a more popular candidate who shares some of the same principles, since that candidate has a much greater chance of being elected and a vote for the fringe candidate is largely wasted.

An IRV system reduces this problem, since the voter can rank the marginal candidate first and the mainstream candidate second; in the likely event that the fringe candidate is eliminated, the vote is not wasted but is transferred to the second preference. However, in the event that the fringe candidate is not eliminated immediately, there is an increased chance for the mainstream candidate to lose, especially when multiple fringe candidates are running. If that happens, the opposing mainstream candidate has a much easier path to victory. As a result, voting honestly for a fringe candidate is unlikely to benefit the voter, and may backfire.

In Australia's national elections in 2007, the average number of candidates in a district was seven, and at least four candidates ran in every district; notwithstanding the fact that Australia only has two major political parties. Every seat was won with a majority of the vote, including several where results would have been different under plurality voting. While IRV reduces the severity of the spoiler effect compared to plurality voting, the problem still persists.

Proportionality

IRV is not a proportional voting system. Like all winner-take-all voting systems, IRV tends to exaggerate the number of seats won by the largest parties; small parties without majority support in any given constituency are unlikely to earn seats in a legislature, although their supporters will be more likely to be part of the final choice between the two strongest candidates. A simulation of IRV in the 2010 UK general election by the Electoral Reform Society concluded that the election would have altered the balance of seats between the three main parties, but the number of seats won by minor parties would have remained unchanged.

Australia, a nation with a long record of using IRV for the election of legislative bodies, has had representation in its parliament broadly similar to that expected by plurality systems. Medium-sized parties, such as the National Party of Australia, can co-exist with coalition partners such as the Liberal Party of Australia, and can compete against it without fear of losing seats to other parties due to vote splitting. IRV is more likely to result in legislatures where no single party has an absolute majority of seats (a hung parliament), but does not generally produce as fragmented a legislature as a fully proportional system, such as is used for the House of Representatives of the Netherlands or the New Zealand House of Representatives, where coalitions of numerous small parties are needed for a majority.

Costs

The costs of printing and counting ballot papers for an IRV election are no different from those of any other system using the same technology. However, the more-complicated counting system may encourage officials to introduce more advanced technology, such as software counters or electronic voting machines. Pierce County, Washington election officials outlined one-time costs of $857,000 to implement IRV for its elections in 2008, covering software and equipment, voter education and testing. In 2009, the auditor of Washington counties reported that the ongoing costs of the system were not necessarily balanced by the costs of eliminating runoffs for most county offices, because those elections may be needed for other offices not elected by IRV. Other jurisdictions have reported immediate cost savings.

Australian elections are counted by hand. The 2010 federal election cost a total of $7.68 per elector of which only a small proportion is the actual counting of votes. Counting is now normally performed in a single pass at the polling center as described above.

The perceived costs or cost savings of adopting an IRV system are commonly used by both supporters and critics. In the 2011 referendum on the Alternative Vote in the UK, the NOtoAV campaign was launched with a claim that adopting the system would cost £250 million; commentators argued that this headline figure had been inflated by including £82 million for the cost of the referendum itself, and a further £130 million on the assumption that the UK would need to introduce electronic voting systems, when ministers had confirmed that there was no intention of implementing such technology, whatever the outcome of the election. Automated vote counting is seen by some to have a greater potential for election fraud; IRV supporters counter these claims with recommended audit procedures, or note that automated counting is not required for the system at all.

Because it does not require two separate votes, IRV is accepted to cost less than two-round primary/general or general/runoff election systems.

Negative campaigning

John Russo, Oakland City Attorney, argued in the Oakland Tribune on 24 July 2006 that "Instant runoff voting is an antidote to the disease of negative campaigning. IRV led to San Francisco candidates campaigning more cooperatively. Under the system, their candidates were less likely to engage in negative campaigning because such tactics would risk alienating the voters who support 'attacked' candidates", reducing the chance that they would support the attacker as a second or third choice.

No formal studies have been conducted in the United States. Internationally, Benjamin Reilly suggests instant-runoff voting eases ethnic conflict in divided societies. This feature was a leading argument for why Papua New Guinea adopted instant-runoff voting.

Critics allege there is a lack of evidence that such an effect occurs as often as suggested. Indeed, Lord Alexander's objections to the conclusions of the British Independent Commission on the Voting System's report cites the example of Australia saying "their politicians tend to be, if anything, more blunt and outspoken than our own".

Plural voting

In Ann Arbor, Michigan arguments over IRV in letters to newspapers included the belief that IRV "gives minority candidate voters two votes", because some voters' ballots may count for their first choice in the first round and a lesser choice in a later round. The argument that IRV represents plural voting is sometimes used in arguments over the 'fairness' of the system, and has led to several legal challenges in the United States. The argument was addressed and rejected by a Michigan court in 1975; in Stephenson v. the Ann Arbor Board of City Canvassers, the court held "majority preferential voting" (as IRV was then known) to be in compliance with the Michigan and United States constitutions, writing:

Under the 'M.P.V. System', however, no one person or voter has more than one effective vote for one office. No voter's vote can be counted more than once for the same candidate. In the final analysis, no voter is given greater weight in his or her vote over the vote of another voter, although to understand this does require a conceptual understanding of how the effect of a 'M.P.V. System' is like that of a run-off election. The form of majority preferential voting employed in the City of Ann Arbor's election of its Mayor does not violate the one-man, one-vote mandate nor does it deprive anyone of equal protection rights under the Michigan or United States Constitutions.

Invalid ballots and exhausted ballots

Because the ballot marking is more complex, there can be an increase in spoiled ballots. In Australia, voters are required to write a number beside every candidate, and error rates 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. 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.

Robert's Rules of Order

The sequential elimination method used by IRV is described in Robert's Rules of Order Newly Revised, 10th edition. as an example of "preferential voting", a term covering "any of a number of voting methods by which, on a single ballot when there are more than two possible choices, the second or less-preferred choices of voters can be taken into account if no candidate or proposition attains a majority. While it is more complicated than other methods of voting in common use, and is not a substitute for the normal procedure of repeated balloting until a majority is obtained, preferential voting is especially useful and fair in an election by mail if it is impractical to take more than one ballot. In such cases, it makes possible a more representative result than under a rule that a plurality shall elect...."Preferential voting has many variations. One method is described ... by way of illustration." And then the instant runoff voting method is detailed.

Robert's Rules continues: "The system of preferential voting just described should not be used in cases where it is possible to follow the normal procedure of repeated balloting until one candidate or proposition attains a majority. Although this type of preferential ballot is preferable to an election by plurality, it affords less freedom of choice than repeated balloting, because it denies voters the opportunity of basing their second or lesser choices on the results of earlier ballots, and because the candidate or proposition in last place is automatically eliminated and may thus be prevented from becoming a compromise choice." Two other books on parliamentary procedure take a similar stance, disapproving of plurality voting and describing preferential voting as an option, if authorized in the bylaws, when repeated balloting is impractical: The Standard Code of Parliamentary Procedure and Riddick's Rules of Procedure.

Global use

Main article: History and use of instant-runoff voting

See also: Table of voting systems by country

Australia

See also: Australian electoral system and Electoral systems of the Australian states and territories

Instant-runoff voting is used for national elections in Australia to elect members of the Australian House of Representatives. The Australian Senate uses a modified form, combining it with a proportional representation system (the Single transferable vote); candidates are eliminated until the remaining parties can be said to have a sufficient proportion of the vote to earn a seat. Most state and council elections also use the system.

Canada

More often called the Alternative Vote in Canada, IRV has never been used for federal elections but was used for provincial elections in British Columbia (1952 and 1953), Alberta (1926), and Manitoba (1927–1953).

IRV is used to elect the leaders of two of the largest federal political parties in Canada, the Liberal Party of Canada and the Conservative Party of Canada. The New Democratic Party of Canada uses a mixture of IRV and exhaustive voting, depending on the member. Canadian Prime Minister Stephen Harper won an IRV election to become party leader in the 2004 leadership election. In 2013, the Liberal Party picked Justin Trudeau with IRV in a national primary.

The province of Ontario has announced that they will allow municipalities the option to use instant-runoff voting for local elections starting in 2018.

Czech Republic

IRV is used to elect leaders of the Green party.

Hong Kong

IRV is used to elect a small number of functional constituencies of the Legislative Council, all of which have very small electorates.

India

IRV is used in numerous electoral college environments, including the election of the President of India by the members of the Parliament of India and of the Vidhan Sabhas – the state legislatures.

Ireland

While most elections in the Republic of Ireland use the single transferable vote (STV), in single-winner contests this reduces to IRV. This is the case in all Presidential elections and Seanad panel by-elections, and most Dáil by-elections In the rare event of multiple simultaneous vacancies in a single Dáil constituency, a single STV by-election may be held; for Seanad panels, multiple IRV by-elections are held.

New Zealand

IRV is used in the elections of mayors and councillors in single-member wards in some New Zealand cities, such as Dunedin and Wellington. Multi-member wards in these cities use STV.

IRV, under the name Alternative Vote, was one of the four alternative systems available (alongside MMP, STV and SM) in the 1992 referendum on the voting method to elect MP's to the New Zealand House of Representatives. It came third of the alternative systems (ahead of SM) with 6.6% of the vote. IRV, under the name Preferential Vote, was one of the four alternative systems choices presented in the 2011 voting method referendum, but the referendum resulted in New Zealanders choosing to keep their proportional system of representation instead, while IRV came last with 8.34%.

Papua New Guinea

Since 2003, the national parliament of Papua New Guinea has been elected using an IRV variant called Limited Preferential Voting, where voters are limited to ranking three candidates.

United Kingdom

In the United Kingdom the system is commonly known as the alternative vote. It is used to elect the leaders of the Labour Party and the Liberal Democrats. (The leader of the Conservative Party is elected under a similar system, a variant of the exhaustive ballot.) It is also used for by-elections to the British House of Lords, in which hereditary peers are selected for that body. AV is also used by members of parliament to elect the chairmen of select committees and the Speaker of the House of Lords. The Speaker of the House of Commons is elected by the exhaustive ballot.

In 2010, the Conservative—Liberal Democrat coalition government agreed to hold a national referendum on the alternative vote, held on 5 May 2011. The proposal would have affected the way in which Members of Parliament are elected to the British House of Commons at Westminster. The result of the referendum was a vote against adoption of the alternative vote by a margin of 67.9 percent to 32.1 percent.

United States

Main article: Instant-runoff voting in the United States

IRV is used by several jurisdictions in the United States, including San Francisco and Oakland, California, and Minneapolis and Saint Paul, Minnesota. United States private associations that use IRV include the Hugo Awards for science fiction, the Academy of Motion Picture Arts and Sciences in selection of the Oscar for Best Picture, and more than fifty colleges and universities for student elections.

Similar systems

Runoff voting

The term instant runoff voting is derived from the name of a class of voting systems called runoff voting. In runoff voting voters do not rank candidates in order of preference on a single ballot. Instead a similar effect is achieved by using multiple rounds of voting. All multi-round runoff voting systems allow voters to change their preferences in each round, incorporating the results of the prior round to influence their decision. This is not possible in IRV, as participants vote only once, and this prohibits certain forms of tactical voting that can be prevalent in 'standard' runoff voting.

Exhaustive ballot

A system closer to IRV is the exhaustive ballot. In this system—one familiar to fans of the television show American Idol—one candidate is eliminated after each round, and many rounds of voting are used, rather than just two. Because holding many rounds of voting on separate days is generally expensive, the exhaustive ballot is not used for large-scale, public elections.

Two-round systems

The simplest form of runoff voting is the two-round system, which typically excludes all but two candidates after the first round, rather than gradually eliminating candidates over a series of rounds. Eliminations can occur with or without allowing and applying preference votes to choose the final two candidates. A second round of voting or counting is only necessary if no candidate receives an overall majority of votes. This system is used in France and the Finnish presidential election.

Contingent vote

Top-two IRV

The contingent vote, also known as Top-two IRV, or batch-style, is the same as IRV except that if no candidate achieves a majority in the first round of counting, all but the two candidates with the most votes are eliminated, and the second preferences for those ballots are counted. As in IRV, there is only one round of voting.

Under a variant of contingent voting used in Sri Lanka, and the elections for Mayor of London in the United Kingdom, voters rank a specified maximum number of candidates. In London, the Supplementary Vote allows voters to express first and second preferences only. Sri Lankan voters rank up to three candidates for the President of Sri Lanka.

While similar to "sequential-elimination" IRV, top-two can produce different results. Excluding more than one candidate after the first count might eliminate a candidate who would have won under sequential elimination IRV. Restricting voters to a maximum number of preferences is more likely to exhaust ballots if voters do not anticipate which candidates will finish in the top two. This can encourage voters to vote more tactically, by ranking at least one candidate they think is likely to win.

Conversely, a practical benefit of 'contingent voting' is expediency and confidence in the result with only two rounds. Particularly in elections with few (e.g., fewer than 100) voters, numerous ties can destroy confidence. Heavy use of tie-breaking rules leaves uncomfortable doubts over whether the winner might have changed if a recount had been performed.

Larger runoff process

IRV may also be part of a larger runoff process:

• Some jurisdictions that hold runoff elections allow absentee (only) voters to submit IRV ballots, because the interval between votes is too short for a second round of absentee voting. IRV ballots enable absentee votes to count in the second (general) election round if their first choice does not make the runoff. Arkansas, South Carolina and Springfield, Illinois adopt this approach. Louisiana uses it only for members of the United States Service or who reside overseas.
• IRV can quickly eliminate weak candidates in early rounds of an exhaustive ballot runoff, using rules to leave the desired number of candidates for further balloting.
• IRV allows an arbitrary victory threshold in a single round of voting, e.g., 60%. In such cases a second vote may be held to confirm the winner.
• IRV elections that require a majority of cast ballots but not that voters rank all candidates may require more than a single IRV ballot due to exhausted ballots.
• Robert's Rules recommends preferential voting for elections by mail and requiring a majority of cast votes to elect a winner, giving IRV as their example. For in-person elections, they recommend repeated balloting until one candidate receives an absolute majority of all votes cast. Repeated voting allows voters to turn to a candidate as a compromise who polled poorly in the initial election.

The common feature of these IRV variations is that one vote is counted per ballot per round, with rules that eliminate the weakest candidate(s) in successive rounds. Most IRV implementations drop the "majority of cast ballots" requirement.

See also

• Alternative Vote Plus (AV+) or Alternative Vote Top-up proposed by the Jenkins Commission (UK)
• First-past-the-post voting
• None of the above (NOTA) or Re-Open Nominations (RON)
• Outline of democracy
• Single transferable vote, AV system for elections with multiple positions to be filled

External links

• 2010 articles from the Constitution Society and Electoral Reform Society summarizing the proposed change in the United Kingdom to IRV/Alternative Vote

Practice

• Advantages and disadvantages of AV from the ACE Project Electoral Design Reference Materials
• A Handbook of Electoral System Design from International IDEA
• Australian Electoral Commission Web Site
• Preferential Voting in Australia from Australian Politics.com
• San Francisco Department of Elections, California
• Alameda County Registrar of Voters, California
• City of Minneapolis, Minnesota
• State of North Carolina

Demos and simulations

• The Star Tribune: How ranked-choice voting works – an interactive graphic
• AmericanQuorum.com A ranked choice ballot tool from the Indaba Application Network, including the animated display of an instant runoff.
• BBC: Would the alternative vote have changed history?, illustration of how the results of the last six general elections might have looked had the 'alternative vote' system been in place.
• OpenSTV – Open source software for computing IRV and STV
• Favourite Futurama Character Poll
• Voting System Visualizations – 2-dimensional plots of results of various systems, with assumptions of sincere voting behavior.
• Simulation Of Various Voting Models for Close Elections Opposition article by Brian Olson.

Advocacy groups and positions

• Yes to Fairer Votes campaign site for the Yes side of the United Kingdom Alternative Vote referendum, 2011
• Washington Post
• Instant Runoff Voting at FairVote
• League of Women Voters of Vermont
• InstantRunoff.com
• Ranked Ballot Initiative of Toronto, Canada
• Roosevelt Institution
• Citizens for Voter Choice :: Massachusetts
• FairVote Minnesota
• Common Cause Massachusetts
• Brookings Institution's "Empowering Moderate Voters" paper
• Does the Alternative Vote Bring Tyranny to Australia? - Antony Green ABC

Opposition groups and positions

• Fair Vote Canada paper on the Alternative Vote
• IRV page at the Center for Range Voting
• Instant Runoff Voting Report Values and Risks Report by the N.C. Coalition for Verified Voting