First-past-the-post (FPTP) — also called choose-one, first-preference plurality (FPP), or simply plurality — is a single-winner voting rule. Each voter marks one candidate as their favorite, or first-preference, and the candidate with more first-preference votes than any other candidate (a plurality) is elected, even if they do not have more than half of votes (a majority).
FPP has been used to elect part of the British House of Commons since the Middle Ages before spreading throughout the British Empire, usually in conjunction with plurality block voting. Throughout the 20th century, the former British colonies of Australia and New Zealand and many other countries that were using FPP abandoned it in favor of other electoral systems. FPP is still used in the majority of US states for most elections. However, the combination of partisan primaries and a two-party system in these jurisdictions means that most American elections behave effectively like two-round systems, in which the first round chooses two main contenders (of which one of them goes on to usually receive a majority of votes).
Example
In FPTP, only the first preferences matter. As such, the votes would be counted as 42% for Memphis, 26% for Nashville, 17% for Knoxville, and 15% for Chattanooga. Since Memphis has the most votes, it would win a FPTP election, even though it is far from the center of the state and most voters would prefer Nashville to any other option. Conversely, instant-runoff voting would elect Knoxville, the easternmost city, because Chattanooga would be eliminated in the first round and its 15% given to Knoxville in the second, giving Knoxville 32% to Nashville's 26% (the elimination and absorption of which would put Knoxville over Memphis' 42%). Such an election result is an example of center squeeze. By contrast, Condorcet methods would elect Nashville (the actual capital) because it beats all others in a head-to-head which takes into account all preferences in order (Nashville beats Memphis in 3 of the blocks, and beats Chattanooga and Knoxville in the biggest 2).
Properties and effects +Table of pathological behaviorsFelsenthal, Dan S. (2010) Review of paradoxes afflicting various voting procedures where one out of m candidates (m ≥ 2) must be elected . In: Assessing Alternative Voting Procedures, London School of Economics and Political Science, London, UK.PathologyExplanation/detailsFrustrated majorityThe frustrated majority paradox occurs when a majority of voters prefer some candidate Brighton to every other candidate, but Brighton still loses the election. First-past-the-post is vulnerable to this paradox because of vote-splitting.Condorcet loser paradoxThe Condorcet loser paradox happens when a majority of voters prefer every other candidate to Brighton, but Brighton still wins. First-past-the-post is vulnerable to this paradox because of vote-splitting.Center squeezeThe center squeeze describes a type of violation of :Independence of irrelevant alternatives primarily affecting voting rules in the Plurality-rule family where the Condorcet winner is eliminated in an early round or otherwise due to a lack of first-preference support.Spoiler effectA spoiler effect happens when the results of an election between A and B is affected by voters' opinions on an unrelated candidate C. First-past-the-post does not meet this criterion, which makes it vulnerable to spoilers.Cloning paradoxThe cloning paradox is a particular kind of spoiler effect that involves several perfect copies, "clones", of a candidate. Candidate-cloning causes vote-splitting in FPP.Best-is-worst paradoxThe best-is-worst paradox occurs when an electoral system declares the same candidate to be in first and last place, depending on whether voters rank candidates from best-to-worst or worst-to-best. FPP demonstrates this pathology, because a candidate can be both the FPP winner and also the anti-plurality loser.Lesser-evil votingLesser-evil voting occurs when voters are forced to support a "lesser of two evils" by rating them higher than their actual favorite candidate. FPP is vulnerable to this pathology.Later-no-harmSince plurality does not consider later preferences on the ballot at all, it is impossible to either harm or help a favorite candidate by marking later preferences. Thus it passes both Later-No-Harm and Later-No-Help.Later-no-helpMultiple-district paradoxThe multiple-district paradox refers to a particularly egregious kind of gerrymander, when it is possible to draw a map where a party that wins when single-member districts are used is defeated when multiple-member districts are used.Franceschini, Fiorenzo; Maisano, Domenico A. (2022-06-01). "Analysing paradoxes in design decisions: the case of "multiple-district" paradox". International Journal on Interactive Design and Manufacturing (IJIDeM). 16 (2): 677–689. doi:10.1007/s12008-022-00860-x. ISSN 1955-2505. This is not possible under FPP, or other positional voting methods.Perverse responsePerverse response occurs when a candidate loses as a result of receiving too much support from some voters, i.e. it is possible for a candidate to lose by receiving too many votes. FPP is not affected by this pathology.No-show paradoxThe no-show paradox is a situation where a candidate loses as a result of having too many supporters. In other words, adding a voter who supports A over B can cause A to lose to B. FPP is not affected by this pathology.
Two-party rule
Perhaps the most striking effect of FPP is the fact that the number of a party's seats in a legislature is determined not by its overall vote count in an election but by how those votes were geographically distributed. Parties with few votes sometimes take more than few seats; often the most-popular party takes 20 percent more seats than its portion of the popular vote. Some criticize FPP for this, arguing that a fundamental requirement of an election system is to accurately represent the views of voters. FPP often creates "false majorities" by over-representing larger parties (giving a majority of the parliamentary/legislative seats to a party that did not receive a majority of the votes) while under-representing smaller ones.
In Canada, majority governments have been formed often but usually they are made up of a party that received less than a majority of votes in the election. A party forming a majority government and also winning a majority of the votes cast has happened only six times since 1900: 1900; 1904; 1917; 1940, 1958 and 1984.
In the United Kingdom, 19 of the 24 general elections since 1922 have produced a single-party majority government. In only two of them (1931 and 1935), the leading party took a majority of the votes across the UK.
In elections in the UK, the FPTP system has produced strange results, with huge gaps between the votes a party receives and the seats it receives. In the 2024 General Election, Reform UK won 14.3% of the popular vote (the third-highest share nationally) but only received 5 seats (0.7% of Parliament). Whilst, the Liberal Democrats won 12.2% of the vote—fewer than Reform UK—but won 72 seats.
In some cases, this can lead to a party receiving the plurality or even majority of the votes cast overall, yet still failing to gain a plurality of legislative seats. This results in a situation called a majority reversal or electoral inversion or wrong-winner result. Famous examples of the second-place party (in votes nationally) winning a majority of seats include the elections in Ghana in 2012, New Zealand in 1978 and 1981, and the United Kingdom in 1951.
Famous examples of the second most popular party (in votes nationally) winning a plurality of seats include the elections in Canada in 1957, 1979, 2019 and 2021. Even when a party wins more than half the votes in an almost purely two-party-competition, it is possible for the runner-up to win a majority of seats. This happened in Saint Vincent and the Grenadines in 1966, 1998, and 2020 and in Belize in 1993. Even with only two parties and equally-sized constituencies, winning a majority of seats just requires receiving more than half the vote in more than half the districts—even if the other party receives all the votes cast in the other districts—so just over a quarter of the vote is theoretically enough to win a majority in the legislature. With enough candidates splitting the vote in a district, the total number of votes needed to win can be made arbitrarily small.
Under first-past-the-post, a small party may draw votes and seats away from a larger party that it is more similar to, and therefore give an advantage to one it is less similar to. For example, in the 2000 United States presidential election, the left-leaning Ralph Nader drew more votes from the left-leaning Al Gore, resulting in Nader spoiling the election for the Democrats. According to the political pressure group Make Votes Matter, FPTP creates a powerful electoral incentive for large parties to target similar segments of voters with similar policies. The effect of this reduces political diversity in a country because the larger parties are incentivized to coalesce around similar policies. The ACE Electoral Knowledge Network describes India's use of FPTP as a "legacy of British colonialism".
Duverger's law is an idea in political science which says that constituencies that use first-past-the-post methods in single-winner elections will lead to two-party systems, given enough time. Economist Jeffrey Sachs explains:
However, most countries with first-past-the-post elections have multiparty legislatures (albeit with two parties larger than the others), the United States being the major exception. There is a counter-argument to Duverger's Law, that while on the national level a plurality system may encourage two parties, in the individual constituencies supermajorities may cause the largest party to suffer fracturing.
Landslide victories In multi-district elections, first-past-the-post voting can result in landslide victories and electoral wipeouts, particularly in case of political fragmentation or low electoral competition. Parties not voted for by a majority of voters may win a majority of seats. However, in Western European legislative elections, the Gallagher indexmeasuring the disproportionality between votes received and seats receiveddoes not appear to have a direct effect on political legitimacy, as people instead mostly care about government composition.
Strongholds, key constituencies and kingmakers The distortions in geographical representation (artificial regionalism) provide incentives for parties to "write off regions" where they are weak and not have much chance of being elected. So they ignore the interests of areas in which they are too weak to stand much chance of gaining representation, leading to governments that do not govern in the national interest.
Further, during election campaigns the campaigning activity of parties tends to focus on marginal seats held by opponents where there is a prospect of a change in representation. These decisions leave safer areas (safe to one party or the other) excluded from participation in an active campaign. Political parties operate by targeting districts, directing their activists and policy proposals toward those areas considered to be marginal, either possible to be lost or won, where each additional vote is potentially more critical and has more value.