{
  "id": "proportional-representation",
  "title": "Proportional representation",
  "lang": "en",
  "source_lang": "en",
  "summary": "Proportional representation comprises electoral systems designed to allocate seats in close correspondence with voters’ support for parties or groups of candidates.",
  "category": "society",
  "subcategory": "politics",
  "type": "concept",
  "aliases": [
    "PR",
    "proportional electoral representation"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/proportional-representation/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/proportional-representation/",
      "json": "https://aiwiki.page/zh/proportional-representation.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Proportional representation (PR) is a family of electoral systems intended to make the distribution of seats in an elected body correspond closely to the distribution of electoral support. If a party receives 30 percent of the relevant votes, the proportional ideal is that it should receive approximately 30 percent of the seats. Unlike plurality voting in single-member constituencies, proportional systems normally allow several political preferences within the same electorate to obtain representation.\n\nPR is widely used to elect national and regional legislatures, municipal councils, and the European Parliament. Its principal forms are party-list representation, the single transferable vote, and compensatory mixed systems. These arrangements differ in whether voters choose parties, individual candidates, or both, and in how votes are translated into seats.\n\nProportionality is a matter of degree rather than an all-or-nothing property. Constituency size, electoral thresholds, allocation formulas, and compensatory rules can produce substantially different results within the PR family. Its institutional significance extends beyond arithmetic: proportional representation affects party competition, opportunities for minority representation, coalition formation, and the relationship between voters and elected representatives."
    },
    {
      "heading": "The proportional principle and its limits",
      "text": "The central distinction is between allocating representation to the range of preferences within an electorate and identifying a single winner in each constituency. In a single-member district, only one candidate can be elected; votes for other candidates do not directly produce a seat there. Under PR, several representatives are usually elected together, or a compensatory allocation corrects distortions created by constituency contests. This enables groups smaller than a local majority to obtain representation without being geographically concentrated.\n\nFor a party receiving a vote share \\(v_i\\) in an assembly with \\(S\\) seats, its exact proportional entitlement would be\n\n\\[s_i^{*}=S v_i.\\]\n\nActual seats must normally be whole numbers, so exact proportionality is frequently impossible. A party receiving 12.5 percent of the votes for a ten-seat assembly has an ideal entitlement of 1.25 seats, but must receive an integer allocation. The rules therefore have to decide how to distribute the unavoidable rounding differences. The same electorate can produce different results under different mathematically defensible formulas.\n\nThe relevant unit of proportionality must also be specified. A system may be proportional within each electoral district but less proportional nationally. If every district elects only three members, a party with evenly distributed support of 10 percent may repeatedly fail to win a seat. Conversely, a national compensatory tier can produce a close correspondence between nationwide votes and seats even when most representatives are initially elected in geographically defined constituencies. Proportionality within districts and proportionality across the whole legislature are distinct institutional objectives.\n\nPR primarily concerns the relationship between votes and representation, not every possible meaning of political equality. It does not by itself guarantee equal turnout, equal campaign resources, a socially representative candidate pool, or impartial constituency boundaries. Those matters also involve suffrage, party organization, electoral administration, and the broader rules of democracy. Nor does proportional representation require representatives to resemble the electorate demographically: proportional party representation and descriptive representation are separate concepts.\n\nThe proportional principle also differs from majority rule. A party can legitimately receive a majority of seats under PR if it wins enough votes. What PR generally seeks to avoid is awarding a large legislative majority to a party supported by a substantially smaller share of voters. The allocation concerns membership of the representative body; subsequent decisions within that body may still be taken by ordinary majority voting."
    },
    {
      "heading": "Historical development and adoption",
      "text": "Modern proportional representation developed during nineteenth-century debates about the expansion of representative institutions. As electorates grew, reformers questioned whether territorial winner-take-all elections adequately represented religious, political, and social minorities. Earlier arrangements such as cumulative voting and the limited vote offered partial minority representation, but did not necessarily allocate seats in proportion to support. The emergence of organized mass parties made the relationship between party votes and legislative seats increasingly important.\n\nTwo foundational strands concerned transferable voting and party-list allocation. In Denmark, Carl Christoffer Georg Andræ devised a transferable-vote arrangement used in elections beginning in 1855. In Britain, Thomas Hare published proposals for proportional personal representation in the 1850s, including The Election of Representatives, Parliamentary and Municipal in 1859. John Stuart Mill supported Hare’s approach in Considerations on Representative Government in 1861. These proposals sought to allow voters with shared preferences to elect representatives even when they did not constitute a territorial majority.\n\nParty-list systems developed alongside these candidate-centred proposals. The Belgian mathematician and lawyer Victor D’Hondt published his seat-allocation method in 1878. Belgium adopted list proportional representation for parliamentary elections in 1899, becoming a major early example of its nationwide application to a lower chamber. Finland used proportional representation in its landmark parliamentary election of 1907, the first under its new system of universal and equal parliamentary suffrage for women and men.\n\nAdoption spread particularly rapidly in Europe around the end of the First World War. The Netherlands introduced proportional representation for its 1918 parliamentary election, and Switzerland first elected its National Council by PR in 1919. Political coalitions supporting reform varied. Smaller parties sought access to representation, while established parties sometimes preferred proportional rules to the possibility of severe defeat under an expanding electorate. PR consequently emerged through different combinations of democratic principle, social accommodation, and strategic institutional bargaining.\n\nLater adoption often accompanied changes in political order. West Germany’s first federal election in 1949 used a mixed system combining constituency representation with proportional allocation. South Africa’s first election under universal adult suffrage in 1994 used party-list PR. New Zealand adopted mixed-member proportional representation following its 1993 referendum and first used it in 1996. These cases illustrate that PR is not a single institutional package: it has been adopted in established parliamentary democracies, newly constituted states, and transitions from exclusionary electoral arrangements."
    },
    {
      "heading": "Party-list systems and voter choice",
      "text": "In party-list proportional representation, parties or other eligible electoral groups nominate lists of candidates. Votes determine how many seats each list receives, while separate rules determine which listed candidates occupy those seats. The distinction between allocating seats among parties and allocating them among candidates is fundamental. Two countries can use the same party-allocation formula while giving voters very different levels of influence over individual representatives.\n\nUnder a closed-list system, the party establishes the order of candidates before the election. If the list wins four seats, its first four eligible candidates are elected. Voters generally choose among lists rather than altering their internal order. Spain’s Congress of Deputies uses closed lists in its multi-member constituencies. Closed lists make the party’s proposed team visible before voting and permit deliberate arrangements such as alternating women and men, but give party nomination procedures considerable influence over individual political careers.\n\nOpen-list systems allow voters to support particular candidates, although the importance of those preferences varies. In Finland, voters select an individual candidate; votes are aggregated for the relevant list, and candidates’ personal vote totals determine their order within it. In other systems, a preference vote changes the party’s nominated order only if a candidate passes a specified threshold. Such arrangements are often called flexible lists. Consequently, “open list” can describe systems ranging from substantial voter control to more limited opportunities to override party rankings.\n\nSome list systems permit more extensive combinations. In Switzerland’s National Council elections in multi-member cantons, voters can modify lists, give a candidate more than one vote within legal limits, and select candidates from different lists. The practice of combining candidates across lists is known as panachage. Such ballots separate support for a party organization from support for particular candidates more sharply than a simple closed-list vote does.\n\nThe list structure influences campaigning and political accountability. Where personal votes determine election, candidates from the same political party may compete against one another as well as against rival parties. They may cultivate local organizations, particular policy constituencies, or personal reputations. Closed lists generally concentrate more electoral competition between parties. Neither arrangement eliminates candidate selection by parties: even highly candidate-centred systems usually require access to an approved list before voters can express preferences.\n\nList PR also requires rules for vacancies and independent candidates. A departing member is often replaced by the next eligible candidate on the same list, preserving the election’s party balance without a by-election. Independent participation may take the form of an individual candidacy, a non-party list, or a group meeting nomination requirements; its practical accessibility depends on the electoral law."
    },
    {
      "heading": "Seat-allocation formulas",
      "text": "Party-list PR requires a method for converting votes into whole seats. The main families are largest-remainder methods and divisor methods. Both begin with vote totals, but they handle rounding differently. Their effects are most visible in smaller districts or close contests for the final seat. No formula can remove the indivisibility of seats, and no single method satisfies every possible mathematical criterion for fair allocation.\n\nA largest-remainder method first establishes a quota. Under the Hare quota,\n\n\\[Q_{\\mathrm{Hare}}=\\frac{V}{S},\\]\n\nwhere \\(V\\) is the number of votes included in the allocation and \\(S\\) the number of seats. Each list receives the whole-number part of its votes divided by the quota. Any remaining seats go to the lists with the largest unused remainders. With 100,000 votes, ten seats, and party totals of 47,000, 32,000, and 21,000, the initial allocations are four, three, and two seats. The remaining seat goes to the first party, whose remainder is 7,000, producing a final allocation of five, three, and two.\n\nA divisor method instead ranks quotients obtained by dividing each party’s votes by a sequence of numbers. Under the D’Hondt method, the divisors are \\(1,2,3,\\ldots\\), and seats go to the highest quotients until all are filled. Under the Sainte-Laguë method, the usual sequence is \\(1,3,5,\\ldots\\). D’Hondt generally gives larger parties a greater advantage in rounding than standard Sainte-Laguë. Modified versions can change the first divisor, making an initial seat harder to obtain.\n\nFor example, suppose three parties receive 50,000, 30,000, and 20,000 votes in a five-seat district. D’Hondt produces an allocation of three, one, and one. Standard Sainte-Laguë produces two, two, and one. The difference does not arise from different ballots or vote totals, but from how the rules resolve competing claims to indivisible seats. As the number of available seats increases, rounding differences usually become smaller relative to the assembly’s size.\n\nThe formulas also differ in their mathematical properties. Certain largest-remainder methods can exhibit the Alabama paradox: increasing the total number of seats can cause a party to lose a seat even when vote totals remain unchanged. Divisor methods avoid that particular problem but can allocate outside a party’s lower or upper quota, the integers immediately surrounding its exact proportional entitlement. These properties connect electoral allocation to the wider mathematics of apportionment, including the distribution of legislative seats among territorial units."
    },
    {
      "heading": "The single transferable vote",
      "text": "The single transferable vote (STV) seeks proportional representation through votes for candidates rather than an indispensable party-list allocation. Voters rank candidates in order of preference in a multi-member constituency. Each ballot supports one continuing candidate at a time. Votes can transfer when a candidate is elected with more support than necessary or when a candidate is eliminated, allowing lower preferences to influence the outcome.\n\nMost public STV systems use a form of the Droop quota:\n\n\\[Q=\\left\\lfloor\\frac{V}{S+1}\\right\\rfloor+1,\\]\n\nwhere \\(V\\) is the number of valid votes and \\(S\\) the seats to fill. In a five-seat election with 60,000 valid votes, the quota is 10,001. Candidates reaching the quota are elected. Votes surplus to their requirements are transferred according to subsequent preferences, using the applicable transfer procedure. If no further candidate reaches the quota, the lowest continuing candidate is normally eliminated and that candidate’s votes are transferred.\n\nThe details of surplus transfers matter. Some procedures transfer selected ballots, while others transfer ballots at fractional values so that the total value moved equals the surplus. Variants differ in which ballots are eligible for transfer, how transfer values are calculated, and how rounding is handled. Ballots become exhausted when no further valid preference remains for a continuing candidate. Thus, the effective body of votes still participating can decline during the count, particularly when voters rank only a few candidates.\n\nSTV is used for elections to Ireland’s Dáil Éireann, Malta’s House of Representatives, and the Australian Senate, although their detailed rules differ. Ireland’s constituencies return three to five members. Malta uses five-member districts, with additional constitutional mechanisms affecting the overall parliamentary result. The Australian Senate normally elects six senators per state at a periodic election and twelve per state at a double-dissolution election. These differences affect the support required to secure representation.\n\nSTV belongs to the broader family of ranked-choice voting, but ranked ballots alone do not make an election proportional. A single-winner ranked election can select only one representative and therefore cannot distribute that one seat proportionally among several groups. STV’s proportional potential depends on electing multiple members and allowing votes to transfer.\n\nIts results are not identical to party-list proportionality. Voters may rank candidates across party boundaries, parties may nominate different numbers of candidates, and personal popularity can affect transfers. STV can therefore represent coalitions of voter preferences not fully captured by party labels. Its proportionality is generally constrained by constituency magnitude: a three-seat district offers fewer opportunities for smaller groups than a much larger district."
    },
    {
      "heading": "Mixed-member proportional and compensatory systems",
      "text": "Mixed-member proportional representation (MMP) combines constituency representatives with an allocation designed to make overall party representation proportional. Voters often have two votes: one for a constituency candidate and one for a party list. The party vote normally determines each qualifying party’s total seat entitlement, and constituency seats already won are counted toward that entitlement. List seats then fill the difference.\n\nA simplified example shows the compensatory logic. In a 100-seat legislature, a party receiving 40 percent of the relevant votes might be entitled to 40 seats. If it wins 25 constituency seats, it receives 15 list seats. Another party entitled to 30 seats but winning only five constituencies would receive 25 list seats. The list tier does not merely add a second election to the first; it corrects the overall party balance.\n\nThis distinguishes MMP from parallel voting, sometimes called a mixed-member majoritarian system. Under parallel voting, constituency seats and list seats are allocated independently. A party winning many constituency seats normally keeps that advantage in addition to its proportional share of the separate list tier. Japan’s House of Representatives uses such a parallel arrangement. The presence of both constituency and list members is therefore insufficient to establish that a system is proportional overall.\n\nNew Zealand’s MMP system normally begins with a 120-seat Parliament. A party generally qualifies for allocation by receiving at least 5 percent of the party vote or winning an electorate seat. Electorate victories are deducted from the party’s overall entitlement, with list members filling the remainder. If a party wins more electorate seats than its party vote would ordinarily justify, those excess seats create an overhang and can increase the size of Parliament.\n\nGermany has repeatedly revised the relationship between constituency results and proportional entitlement. Earlier federal arrangements allowed overhang seats and, later, extensive compensatory additions. Electoral legislation enacted in 2023 introduced a fixed size of 630 seats for the Bundestag and made constituency victories subject to coverage by the party’s second-vote entitlement. The central principle remains that the party vote governs proportional representation; finishing first in a constituency does not necessarily guarantee a seat under these rules.\n\nCompensation need not involve a full MMP system. Some countries allocate ordinary seats in regional districts and then distribute leveling seats to correct national or regional imbalances. Norway, for example, uses district seats together with nationwide leveling seats for elections to the Storting. The degree of correction depends on the number of compensatory seats, eligibility rules, and whether the mechanism has enough capacity to offset distortions created in the initial allocation."
    },
    {
      "heading": "District magnitude, thresholds, and electoral geography",
      "text": "District magnitude means the number of representatives elected from a district, not its population or geographical area. It is one of the strongest influences on proportionality. A district electing twenty members can generally represent smaller shares of voters than one electing three. Increasing magnitude reduces the relative importance of rounding and broadens the range of parties capable of winning seats, although formulas and thresholds continue to matter.\n\nA legal electoral threshold establishes a minimum vote share or other condition for participation in seat allocation. Israel, which elects its 120-member Knesset in a nationwide constituency, has used a threshold of 3.25 percent since legislation passed in 2014. The Netherlands has no comparable higher percentage barrier for its 150-seat lower chamber; eligibility is linked to obtaining the electoral quota, approximately one seat’s worth of votes. These arrangements produce different opportunities for small parties despite both using national list PR.\n\nA natural or effective threshold arises from the scarcity of seats even without an explicit legal barrier. In a five-seat district, a party with 2 percent of the vote is unlikely to win representation under ordinary proportional formulas. There is no universal effective-threshold figure because outcomes depend on the distribution of votes among competitors. A commonly used comparative approximation is \\(75/(M+1)\\) percent, where \\(M\\) is district magnitude, but this is a heuristic rather than a legal cutoff or guaranteed minimum.\n\nGeographical distribution interacts with these rules. A regional party concentrated in a few districts can win seats under arrangements that exclude a party with the same national vote share spread thinly across the country. Spain illustrates how a system formally based on PR can create different electoral conditions in large and small provincial constituencies. Regional concentration is not inherently an advantage or disadvantage under PR; its effect depends on where allocation occurs and whether a higher-level tier compensates for local outcomes.\n\nSeat distribution among territories creates an additional issue. If sparsely populated districts receive more seats per resident than densely populated ones, the legislature may be malapportioned even when seats within each district are allocated proportionally. Population data, often obtained through a population census, may guide periodic redistribution. Rules guaranteeing minimum territorial representation can nevertheless create deliberate departures from equal population per seat.\n\nTerritorial design can serve objectives other than national proportionality, including representation of remote areas, linguistic communities, or constituent units of a federal system. The important analytical distinction is between proportional allocation among competitors and equal weighting among geographical electorates. A system can perform well on one dimension and less well on the other."
    },
    {
      "heading": "Party competition, coalition government, and accountability",
      "text": "PR generally makes legislative entry more feasible for smaller parties than single-member plurality elections do. This operates through both mechanical and behavioural effects. Mechanically, votes for a smaller party can yield seats without the party finishing first in a locality. Behaviourally, voters and political organizers may be more willing to support or establish such parties when representation appears attainable. These mechanisms form part of the comparative literature associated with Duverger’s law and related propositions about electoral rules and party systems.\n\nThe relationship is not deterministic. Electoral thresholds, district magnitude, social divisions, political history, and party organization all influence the number of competitive parties. A proportional system with a high threshold and small districts can sustain fewer parliamentary parties than one with a large nationwide district and a low barrier to entry. Conversely, a plurality system can accommodate several significant parties when their support is geographically concentrated.\n\nBecause PR often produces legislatures in which no party holds a majority, coalition government is common in parliamentary settings. Parties may negotiate a joint programme, distribute ministerial posts, and agree on procedures for handling disagreement. A minority cabinet can also govern without a formal majority coalition if other parties tolerate it or provide support on confidence and budget votes. The connection between PR and cabinet formation is especially important in a parliamentary system, where the executive depends on legislative confidence.\n\nCoalition bargaining creates a distinction between electoral strength and governing influence. A small party may become pivotal if its seats are necessary for a viable majority. Nevertheless, small size alone does not confer bargaining power: ideological compatibility, alternative coalitions, internal cohesion, and public commitments constrain negotiations. The party winning the most votes is not automatically entitled to govern if other parties can form a majority coalition.\n\nArguments about government stability require similar distinctions. Cabinet duration, continuity of governing parties, legislative productivity, and stability of the constitutional order are different outcomes. A cabinet may change while much of its coalition remains in office, and lengthy government-formation talks do not necessarily imply a breakdown of ordinary administration. Comparisons therefore need to account for confidence procedures, investiture requirements, party discipline, and other institutional arrangements.\n\nAccountability can also take different forms. Coalition government may make responsibility for a particular compromise harder to assign to one party, but PR can give voters a wider range of meaningful alternatives. In candidate-centred systems, voters may replace individual representatives while continuing to support their party. Electoral rules shape the channels of accountability without determining whether parties provide clear programmes, explain compromises, or respond to voters."
    },
    {
      "heading": "Minority representation and the composition of legislatures",
      "text": "PR can improve the electoral prospects of political minorities by allowing representation without a local plurality. This is especially relevant to groups whose supporters are geographically dispersed. A party receiving a modest but consistent share of votes across the country may obtain seats under national PR while winning none under single-member plurality. The effect concerns political support, however: membership in a religious, linguistic, ethnic, or social group does not necessarily imply support for one party.\n\nTerritorially concentrated minorities face a different institutional calculation. They may already be able to elect constituency representatives under a non-proportional system. Under PR, their opportunities depend on district boundaries, thresholds, and access to national or regional allocation. Some electoral laws provide threshold exemptions or other special arrangements for recognized minority parties. Such provisions are supplementary institutional choices, not inherent features of proportional representation.\n\nThe relationship between PR and gender equality has received substantial comparative attention. Multi-member party lists allow parties to nominate a balanced group of candidates rather than select only one candidate for a constituency. Closed lists can make placement rules particularly consequential: a requirement to include women has limited effect if they appear only in unwinnable positions. Alternating women and men on a list, often called a zipper arrangement, directly addresses candidate order.\n\nGender quotas can be established by law or adopted voluntarily by parties. Their effects depend on the proportion required, ranking provisions, sanctions for non-compliance, and the number of seats a list wins. Open-list systems introduce another stage because voters’ preferences may alter the order of election. Representation therefore reflects the interaction of nomination rules, electoral incentives, voter behaviour, and party support rather than PR alone.\n\nProportionality among parties should also be distinguished from representation of age groups, occupations, income levels, disabled people, or other social categories. Electoral systems operate on votes and candidacies; they do not automatically reproduce the population’s composition. Candidate recruitment, campaign finance, accessibility, and party leadership structures remain important. A legislature can be highly proportional in party terms while remaining socially unrepresentative.\n\nBeyond national elections, proportional methods are used in some local councils, organizational elections, and committee allocations. Within legislatures, committee seats may be divided proportionally among parliamentary groups even when the legislature itself was elected under a different electoral system. This extends the principle from choosing representatives to distributing opportunities for participation in legislative oversight and parliamentary work. The relevant proportional unit may then be parliamentary seats rather than popular votes."
    },
    {
      "heading": "Measuring results and evaluating institutional design",
      "text": "Researchers assess disproportionality by comparing parties’ vote shares with their seat shares. One widely used measure is the Gallagher index, introduced by political scientist Michael Gallagher in 1991. If \\(v_i\\) and \\(s_i\\) are vote and seat percentages for party \\(i\\), it is calculated as\n\n\\[G=\\sqrt{\\frac{1}{2}\\sum_i(s_i-v_i)^2}.\\]\n\nA value of zero indicates exact proportionality. Larger discrepancies contribute more heavily because the differences are squared. The index permits comparisons across elections, although decisions about how to classify independents and small parties can affect the result.\n\nThe Loosemore–Hanby index uses a different calculation:\n\n\\[D=\\frac{1}{2}\\sum_i |s_i-v_i|.\\]\n\nIt measures half the total absolute difference between vote and seat shares. With three parties receiving vote shares of 50, 30, and 20 percent and seat shares of 60, 25, and 15 percent, this index is 10 percentage points. The Gallagher index for the same result is approximately 8.66. Neither figure by itself identifies the institutional cause of the discrepancy: thresholds, district size, geography, or a mixed-system design may all contribute.\n\nThe denominator is important. Calculating proportionality against all valid party votes captures the exclusion of parties below the threshold. Recalculating only among parties admitted to allocation answers a narrower question and usually makes the result appear more proportional. In mixed systems, researchers must specify whether they compare seats with constituency votes, list votes, or another legally relevant vote. Under STV, aggregating first preferences by party gives useful information but does not capture every preference expressed through transfers.\n\nEvaluation also extends beyond a single numerical index. Electoral integrity requires accurate voter registration, secure ballot handling, transparent counting, and effective dispute resolution. A proportional formula cannot compensate for restrictions on genuine competition or unequal access to the ballot. Conversely, a carefully administered election can use a non-proportional formula. The proportionality of results and the democratic quality of the electoral process are related but distinct subjects within representative democracy.\n\nInstitutional design involves trade-offs among proportionality, geographical representation, candidate choice, simplicity, and governability. Large national districts can produce close party proportionality but provide a less localized constituency link. Open lists offer personal choice but may intensify competition among co-partisans. Compensation preserves local contests while introducing additional allocation rules. Thresholds can limit parliamentary fragmentation while leaving some voters without representation for their preferred party.\n\nProposals for reform consequently require specification of the entire system rather than endorsement of “PR” in the abstract. The ballot, nomination process, district structure, allocation formula, eligibility thresholds, compensatory mechanism, vacancy rules, and counting procedures jointly determine how an election works. Even apparently technical changes—such as replacing a first divisor, altering a threshold exemption, or reducing the number of leveling seats—can change which candidates enter the legislature and how closely its composition reflects electoral support."
    }
  ],
  "links": [
    "plurality-voting",
    "legislature",
    "european-parliament",
    "suffrage",
    "democracy",
    "majority-rule",
    "world-war-i",
    "referendum",
    "political-accountability",
    "political-party",
    "ranked-choice-voting",
    "storting",
    "population-census",
    "federalism",
    "parliamentary-system",
    "government",
    "gender-equality",
    "legislative-oversight",
    "representative-democracy"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}