Why New Political Parties and Independent Candidates Struggle to Become Electorally Viable in the United States

Full report

On this page

Executive finding: the viability problem is real, but it is not one thing

The best-supported explanation for the weakness of new U.S. parties is not a single act of deliberate suppression. It is a layered equilibrium produced by electoral rules, state ballot laws, strategic behavior by voters and elites, accumulated organizational advantages, and several institutions that use prior demonstrations of support as a proxy for seriousness. Some components were consciously designed in ways that favor established parties; others arose historically or follow rationally from winner-take-all competition. Crucially, the system can reproduce two-party dominance even when nearly every individual actor is behaving rationally and no one is coordinating to exclude outsiders. The Supreme Court itself has recognized both sides of this problem: it has struck down ballot rules that gave established parties an impermissible advantage while also upholding reasonable restrictions designed to require a “modicum of support” and maintain administrable ballots.

The strongest causal core is the interaction of single-winner elections plus plurality voting plus expectations about who can win. Political science's “M+1” result predicts roughly one more serious candidate than the number of seats being filled; in a one-seat district, that means pressure toward two viable candidates. The mechanism is not just mechanical. Voters, donors, candidates, endorsers and campaign professionals anticipate the mechanical rule and coordinate around the competitors they expect other people to coordinate around. Gary Cox's Making Votes Count and Steven Reed's work on the M+1 rule treat this as a strategic-coordination problem rather than a literal legal prohibition on third parties.

The complete mechanism is approximately:

underlying voter preference → ballot qualification and organization → polling / early observable support → donor, staff, endorsement and media judgments → perceived competitiveness → debate access and additional coverage → voter expectations about who can win → strategic voting → final vote share → future ballot status, fundraising credibility and party infrastructure → back to the next election.

That loop contains genuine feedback. A candidate can be weak because voters do not support the candidate, but a candidate can also remain weak partly because everyone expects everyone else not to support the candidate.

The empirical evidence is not equally strong for every arrow. My synthesis of the literature is:

There is therefore no defensible scholarly decomposition such as “40% ballot access, 30% strategic voting, 20% media.” The mechanisms are endogenous. A poll affects donations; donations affect organization; organization affects polling; electoral rules affect every actor's willingness to participate in the first place. Assigning additive percentages would create false precision.

A better overall judgment is:

The United States has genuinely competitive elections between two large coalitions, but unusually strong structural pressures that channel political competition through those two coalitions rather than allowing multiple independent party organizations to coexist at comparable strength.

That is a different claim from saying that U.S. elections are generally “fake” or predetermined.

Ballot access and electoral rules create the underlying entry problem

Ballot access is a decentralized fixed-cost problem

There is no single national ballot-access procedure for a presidential candidate. Election administration is substantially state-based, and official state guidance reflects different nomination, petition and write-in routes. A presidential organization seeking nationwide access must therefore solve a separate legal and administrative problem in every state plus the District of Columbia; congressional, state and local candidates confront the corresponding rules in their own jurisdictions. Colorado, for example, explicitly describes nomination, petition and write-in as alternative routes into candidacy, while the Election Assistance Commission emphasizes that election procedures vary substantially by state and locality.

Across the states, the relevant requirements fall into a recurring set of categories. A new party may have to establish party status by petition or registration; an existing party may retain automatic status by satisfying a previous-election vote or registration threshold; an unaffiliated candidate may need a separate nominating petition; deadlines may precede the general election by months; signatures must satisfy voter-eligibility and geographic rules and can be challenged; presidential candidates also need valid slates of presidential electors and associated filings; and write-in candidates generally must file declarations to have write-in votes counted. The precise thresholds and combinations change by state and office, so “ballot access” is better understood as fifty-plus different regulatory regimes rather than a single American requirement.

Supreme Court doctrine demonstrates just how consequential the details can be.

In Williams v. Rhodes in 1968, Ohio required a new party to obtain signatures equal to 15% of the number voting in the preceding gubernatorial election, with an early filing schedule, while established Democrats and Republicans could retain ballot status under much less burdensome conditions. The Supreme Court concluded that the cumulative structure gave established parties a “decided advantage” and violated constitutional rights of voting and association. The case is unusually strong evidence that formal electoral rules can, at least in particular instances, produce constitutionally impermissible incumbent-party advantages.

Yet three years later, Jenness v. Fortson upheld Georgia's requirement that an independent or minor-party candidate secure signatures from 5% of eligible voters. The Court stressed countervailing features: candidates had 180 days, petition signers were not tightly restricted, and write-ins remained possible. Thus, a substantial numerical threshold is not automatically unconstitutional.

In Anderson v. Celebrezze, the Court invalidated Ohio's March deadline for independent presidential candidate John Anderson. The Court established the balancing framework still central to ballot-access litigation: weigh the character and magnitude of the burden against the state's asserted interests and consider whether the burden is necessary. It also specifically warned that restrictions falling unequally on independents and new parties implicate associational rights, and that “political stability” cannot simply mean protecting existing parties from competition.

Conversely, Burdick v. Takushi upheld Hawaii's prohibition on write-in voting because the Court regarded Hawaii's ordinary routes to the printed ballot as sufficiently accessible. Severe restrictions receive demanding review; reasonable and nondiscriminatory regulation can be justified by legitimate administrative interests.

Those cases rule out two simplistic interpretations. “Any access requirement is suppression” is wrong. “Ballot laws are irrelevant because anyone can technically run” is also wrong. State governments have legitimate reasons to demand evidence of support, prevent ballot overcrowding, administer deadlines and avoid voter confusion. But a collection of individually defensible requirements can still impose significant fixed costs on organizations that do not already possess lawyers, volunteers, voter lists and standing ballot status.

The burden is especially asymmetric because an established party often begins an election cycle with something a new organization must purchase with effort: a recognized legal identity and an existing ballot line. That is path dependence embedded in law even when the law's current purpose is administrative rather than suppressive.

Plurality voting changes what a rational vote means

Ballot access only determines whether Candidate C may compete. The electoral system determines what happens once C is present.

Suppose the expected vote is:

Under a one-round plurality rule, C's supporter faces a real tradeoff. A sincere vote for C expresses a first preference but has little chance of determining who holds office. A vote for A sacrifices expressive accuracy but may decrease the probability that B wins. Voting for A is therefore not necessarily evidence that A is the voter's true favorite. It can be instrumentally rational strategic voting. Formal models of strategic voting show that voters' actions depend on expected closeness and beliefs about other voters, not merely on ordinal preference.

This generates Duverger-style coordination. There are two mechanisms.

The mechanical effect is that coming third wins no fraction of a single seat. A party receiving 15% uniformly across hundreds of single-member districts can receive substantial national support and still win few or no seats.

The psychological/strategic effect occurs before ballots are cast: voters, politicians and donors anticipate that mechanical effect and consolidate around the alternatives they think can finish first or second. Reed's empirical work on the M+1 rule shows why the expected number of viable competitors tends toward seats plus one, although the result is probabilistic rather than an iron law.

That distinction matters. Plurality does not literally prohibit third parties, and it does not guarantee exactly two national parties. Geography can sustain additional parties: a party with 15% concentrated in one region may win many seats while a party with 15% spread uniformly across the country may win none. Britain and Canada therefore have more than two parliamentary parties despite extensive use of single-member plurality. The U.S. combination of geographically broad presidential politics, single-winner congressional elections and two deeply established nationwide organizations makes the coordinating pressure especially powerful.

Presidential elections reinforce the same incentives. The presidency is intrinsically one office, Electoral College votes are allocated on a winner-take-most basis in almost every state, and winning the presidency ordinarily requires an Electoral College majority. That gives voters additional reason to ask not simply “Whom do I prefer?” but “Which candidate has a plausible path to the relevant state victories and 270 electoral votes?” The constitutional architecture makes the presidency a particularly hostile environment for a geographically diffuse third party.

The result is an important paradox: a third candidate may need visible support before strategic voters are willing to support them, but cannot demonstrate that support because strategic voters are waiting for others to move first. That is a coordination problem, not necessarily suppression.

Money, media, debates, polling, personnel and endorsements amplify the initial disadvantage

The electoral rule does not operate in isolation. It supplies the expectations on which an entire campaign ecosystem acts.

Funding: legal equality can coexist with network inequality

At the federal level, private contribution rules do not simply say that Democrats and Republicans may raise money while independents may not. For the 2025–26 cycle, individuals are subject to the same federal candidate contribution limit regardless of the candidate's party, and independent-expenditure-only committees can receive unlimited contributions for independent spending subject to applicable disclosure and noncoordination rules. An independent candidate can therefore, in principle, acquire enormous outside financial support.

The more important disadvantage is usually expected return on political investment.

A donor deciding where to place $1,000 or $100,000 of independent expenditure has several possible motives: ideological expression, helping to elect someone, gaining policy influence, building a movement or establishing relationships with future officials. If Candidate C is believed to have a 2% chance of winning and Candidate A has a 48% chance, election-oriented donors have a rational reason to favor A even when they prefer C's platform. The same calculation occurs among bundlers, political-action committees and large organizational donors. This is not necessarily corrupt; it is partly an expected-value response to perceived viability.

There is, however, also a formally path-dependent component in presidential public financing. The FEC's statutory system provides full general-election grants to qualifying major-party nominees, while a minor party's entitlement depends partly on its previous presidential vote: a party receiving between 5% and 25% in the preceding presidential election can qualify for proportional general-election funding. A new party may qualify for post-election funding only after demonstrating at least 5% support in the current election. Primary matching funds likewise require candidates to establish geographically dispersed fundraising and accept expenditure limits.

That is a genuine demonstrated-viability rule: prior success can unlock future resources. But its practical importance should not be overstated. Candidates accepting public financing also accept spending restrictions, making the program less attractive in modern high-spending campaigns. Thus public-finance design is evidence of institutional path dependence, but it is not sufficient to explain today's financing gap.

Media: competitiveness itself is newsworthy

Commercial and journalistic media face scarce airtime, reporter attention and audience attention. A candidate believed likely to become president or senator is inherently more consequential than a candidate polling at 1%, so allocating unequal coverage can arise from ordinary news judgment without any agreement to exclude outsiders. “Horse-race” research has long documented that election journalism devotes substantial attention to who is ahead, behind, gaining and losing rather than treating every legally qualified candidacy as equally newsworthy; Thomas Patterson's Out of Order and subsequent campaign-media research document this orientation.

This creates an endogenous problem. Coverage is rationally allocated partly according to viability, yet coverage itself supplies name recognition, legitimacy, donors and volunteers that help create viability. Observational research can readily show the association; establishing how much low coverage causes low support rather than simply reflects it is much harder.

Federal broadcasting law does not amount to a requirement that all candidates receive identical news treatment. The equal-opportunities provision in 47 U.S.C. §315 contains exceptions for bona fide news programs, interviews and news events. Consequently, the legal fact of candidacy does not entitle a low-polling candidate to equal editorial attention.

The appropriate empirical conclusion is therefore moderate rather than maximal: perceived competitiveness strongly influences coverage; coverage plausibly reinforces competitiveness; but “the media causes third-party failure” overstates a reciprocal relationship.

Debates: an unusually visible viability threshold

Presidential debates provide one of the clearest examples of a private institutional rule translating apparent viability into exposure.

There is no constitutional entitlement to participate in a televised presidential debate. Federal campaign-finance regulations instead allow qualifying nonprofit organizations to stage candidate debates subject to requirements including pre-established, objective candidate-selection criteria. See 11 C.F.R. §110.13. Thus the legally relevant principle is generally objective criteria, not “every ballot-qualified candidate must appear.”

The 2024 election illustrates the consequences. CNN required candidates for its June debate to reach at least 15% in four approved national polls and to appear on enough state ballots to have a mathematical route to 270 electoral votes. Robert F. Kennedy Jr. met the polling threshold in only three qualifying polls and had not obtained certified ballot access covering enough electoral votes by the deadline, so he was excluded.

That episode is important for three reasons.

First, ballot access, polling and debate access became multiplicative rather than separate barriers. A candidate had to clear both ballot and poll hurdles.

Second, the 2024 debates were arranged outside the longstanding Commission on Presidential Debates structure. That weakens the argument that there is a single permanent two-party institution controlling presidential debate access; the gatekeeper can change while viability-based selection persists.

Third, any threshold presents a real normative tradeoff. A debate containing dozens of candidates may become unusable, so some filtering is defensible. But a high poll threshold can convert a prior lack of visibility into continued lack of visibility. The fairness question is therefore not simply whether a threshold exists, but whether it is proportionate, transparent, neutral and realistically attainable by an entrant before exposure is denied.

Polling: both measurement and coordination device

Polls play two conceptually separate roles.

They measure support, imperfectly. Candidate-list design can matter because a respondent cannot select a named candidate who is not offered as a response option except through whatever “other” or volunteered-response procedure the poll uses. Polling organizations also differ over when to add minor candidates to ballot tests.

But polls also communicate beliefs about everyone else's behavior. This is extraordinarily important under plurality. A voter cannot vote strategically merely from knowing their own preferences; they need beliefs about who is viable. Polls supply those beliefs. Strategic-voting models therefore imply that even an unbiased poll can affect coordination by making information about likely rankings public.

Where debate eligibility is conditioned on poll numbers, the interaction becomes direct: polling no longer merely reports the campaign; it helps determine access to another campaign institution, as the 2024 CNN criteria demonstrated.

Yet this is one of the areas where strong claims outrun evidence. There is good reason to think candidate inclusion changes measured support at the margin and that polls inform strategic voters. There is much weaker evidence that pollsters' omission of minor candidates, considered independently of their underlying weakness, is a major cause of their eventual electoral failure. Polling should therefore be treated as an amplifier and coordination technology, not as a demonstrated primary cause.

Personnel and candidate recruitment: the major parties are labor markets as well as labels

A serious campaign needs far more than a candidate: ballot lawyers, fundraisers, field directors, data staff, communication specialists, opposition researchers, digital specialists, volunteers, voter files and local organizational knowledge.

Established parties supply networks connecting these complementary assets. Research that treats parties as networks rather than merely formal committees—such as Cohen, Karol, Noel and Zaller's The Party Decides, and work by Seth Masket and by Koger, Masket and Noel—finds that party actors coordinate endorsements, information, candidates and campaign resources well beyond the official national committee.

For an experienced professional, joining an established party's campaign has rational career advantages: a higher probability of victory, repeat employment in subsequent campaigns, connections to officeholders, access to established vendors and a recognizable résumé signal. Talented prospective candidates face a similar choice. Rather than spend years constructing Party C, they can often challenge the leadership or ideological direction of Party A through its primary.

This is one of the most important U.S. features often missing from “third parties are excluded” accounts: the two parties are unusually permeable coalitions. Movements that might form new parties in parliamentary systems can instead contest Democratic or Republican primaries, take over local committees, defeat incumbents and gradually move the national coalition. That process limits organizational pluralism while preserving substantial ideological competition.

The relevant structural effect is therefore not simply “outsiders are kept out.” It is more accurately:

outsiders are strongly incentivized to enter politics through one of two existing franchises.

That helps explain why dissatisfaction can produce insurgencies inside the Republican or Democratic Party rather than durable new parties outside them.

Endorsements: an expected-impact calculation

Labor unions, advocacy groups, trade associations, elected officials and ideological organizations typically care about both policy proximity and the probability that support will matter. An organization may sincerely prefer C but endorse A if A is sufficiently closer than B and has a much greater probability of winning. An endorsement can bring members, volunteers, donations and credibility, so those decisions add another feedback channel.

The mechanism is analogous to strategic voting, except that organizations are allocating bundles of resources rather than one ballot. It need not reflect hostility toward C. An environmental group that thinks C is perfect, A is acceptable and B disastrous may maximize expected policy value by helping A.

Directly isolating “endorsement viability bias” from ideology, candidate quality and party relationships is difficult, so the evidence is less decisive than the evidence for electoral-system effects. Nevertheless, party-nomination and endorsement research strongly supports the broader proposition that elite coordination helps narrow candidate fields before ordinary voters make final choices.

The two-party equilibrium is self-reinforcing, but some links are much stronger than others

The proposed feedback loop is theoretically coherent and substantially supported:

little support → little money/coverage/personnel → low apparent viability → strategic voter desertion → little support

But it is necessary to separate true feedback from a theory that becomes unfalsifiable by explaining every loss as exclusion.

Consider each link.

The feedback hypothesis is therefore strongest when treated as a coordination equilibrium, not as a conspiracy model.

A useful game-theoretic way to see it is to imagine millions of voters who would all prefer C to A but believe everyone else will stay with A because B is unacceptable. Each individual's optimal response may be to remain with A. The resulting low C vote then validates the expectation. A second equilibrium might exist in which enough voters coordinate on C and C replaces A. But moving from the first equilibrium to the second requires a coordination shock—an incumbent-party collapse, extraordinary candidate, geographic concentration, institutional reform or major realignment.

That pattern helps explain why plurality systems often produce party replacement rather than stable three-party competition. The Republican Party's rise in the 1850s is not a counterexample to two-party pressure so much as a demonstration of its dynamic form: the Republicans supplanted the Whigs rather than becoming a permanent third party alongside two durable old parties.

History also challenges the idea that today's party duopoly was simply created by modern ballot-access laws. Two-party competition developed in the United States long before the present regulatory structure. The dominant organizations repeatedly changed—Federalists, Democratic-Republicans, Whigs, Democrats, Republicans—but a two-camp structure repeatedly reconstituted itself. Later adoption of government-printed ballots and formal qualification procedures increased the state's role in deciding which party labels appeared automatically, adding path-dependent legal advantages to an already existing strategic structure.

Several other observations challenge a maximal structural-exclusion theory.

Major-party takeover is possible. Because nominations are contested in primaries, an entrepreneurial movement can use an existing party's ballot infrastructure instead of building a new party. This is a genuine route to political change, although it channels that change into two organizations.

Money alone does not solve the problem. Ross Perot in 1992 demonstrated that a wealthy, nationally known independent could acquire extensive coverage and win nearly one-fifth of the popular vote yet still win no electoral votes. The example is difficult to reconcile with theories claiming lack of money or media attention is the entire problem; it is highly consistent with theories emphasizing geographic winner-take-all rules and strategic coordination. The FEC's official 1992 results report Perot at approximately 18.9% of the national popular vote.

Plurality does not always yield two national parties. Canada and the United Kingdom sustain additional parties because concentrated regional support can convert votes into seats. That indicates that district geography, parliamentary institutions and historical cleavages mediate Duverger's pressure.

Ballot rules sometimes survive stringent legal review. Jenness and Burdick make clear that constitutional doctrine recognizes ballot administration as a legitimate state function and does not treat every burden on an entrant as an illicit attempt to suppress competition.

The structural hypothesis survives these objections, but in a narrower and stronger form:

The U.S. system does not make third-party success impossible. It substantially raises the coordination and organizational threshold at which third-party support becomes self-sustaining, and it gives existing parties advantages that accumulate precisely because they have already crossed that threshold.

That is a classic path-dependent network effect.

Comparative evidence shows which institutions make entry easier

Cross-national comparison is especially useful because it tests whether the American outcome follows from universal voter psychology or from institutions.

Proportional representation lowers the penalty for being third

Germany combines constituency voting with proportional allocation. The Bundestag's official explanation states that party representation is principally linked to the second, party-list vote and that parties generally must receive more than 5% nationwide to enter the Bundestag, subject to specified exceptions. A party therefore does not need to finish first in hundreds of separate districts to turn, for example, 10% national support into meaningful parliamentary representation.

Germany still has an entry barrier: 5% is consequential. The crucial distinction is that above the threshold, a party's votes translate proportionally into legislative power. That dramatically improves the expected value of supporting a party polling at 8%. A German voter need not conclude that an 8% party is “throwing away” the vote merely because it will not become the largest party.

New Zealand provides an unusually informative quasi-historical contrast. Its older first-past-the-post system strongly favored the two largest parties; after adopting Mixed Member Proportional representation in 1996, parliamentary representation became substantially more multiparty. Under MMP, voters have separate electorate and party votes, and qualifying parties receive representation roughly in proportion to their party vote, subject to the statutory threshold. The institutional change did not create previously nonexistent ideological diversity; it changed the ability of that diversity to survive as separate organizations. This is consistent with comparative theories linking district magnitude and proportionality to party-system fragmentation.

The Netherlands goes farther: highly proportional allocation and a very low effective threshold permit relatively small national vote shares to yield seats. The recurring appearance of new parliamentary parties there demonstrates that high turnover in party organizations is feasible when voters can support an entrant without first believing that it will finish first in a geographic district.

These cases strongly support the structural hypothesis—but specifically its electoral-system component.

Canada shows that plurality is a pressure, not a law of nature

Canada is a valuable challenge case because it also uses single-member plurality for the federal House of Commons yet has repeatedly sustained more than two relevant parties.

The reason does not falsify strategic theory. It illustrates it. Parties such as the Bloc Québécois or historically the Reform Party could concentrate support geographically. A party that is first in many western Canadian districts can win parliamentary representation even without being competitive nationally. Parliamentary government also means a smaller party can wield legislative or coalition influence without having a plausible chance of independently winning a separately elected presidency.

The lesson is:

single-member plurality is especially hostile to a third party whose support is broad but shallow; it is less hostile to one whose support is regionally concentrated.

That makes the U.S. presidency a particularly severe case because a durable new nationwide party needs both national infrastructure and geographically efficient support.

France demonstrates the importance of multiple rounds

Two-round presidential and legislative elections modify the coordination problem. Voters can use a first round to reveal preference among several alternatives and coordinate more aggressively in the decisive round. Emmanuel Macron's newly created political movement was able to break through France's established party structure in 2017 during a period of profound party-system realignment.

A runoff does not eliminate strategic voting—indeed it can create strategic calculations about who will qualify—but it postpones some “lesser evil” coordination until after voters have first demonstrated support.

Ranked-choice voting helps one mechanism but is not proportional representation

Single-winner ranked-choice voting can reduce one particular problem: a voter can rank C first and A second without necessarily helping B by withholding a first-choice vote from A. That reduces the classic spoiler incentive.

But it does not automatically produce multiparty legislatures. If every district still elects one representative, only one candidate ultimately wins each seat. The proportional-representation advantage arises mainly from multi-member districts plus proportional or semi-proportional allocation, not merely from ranking candidates.

This distinction is important when evaluating reform proposals. Ranked-choice voting can reduce strategic desertion; proportional representation can alter the deeper seat-conversion structure.

The comparative institutional prediction

The probability of successful new-party entry should rise when several of the following conditions are present:

The comparative record therefore supports neither “institutions do not matter” nor “plurality makes new parties literally impossible.” The better proposition is probabilistic: institutions change the support threshold at which a new organization becomes rationally investable by voters and elites.

Historical path dependence explains why dominance reproduces itself

Once two parties have become dominant, several advantages arise simply from the fact of dominance.

The label becomes an information shortcut. Voters do not need to investigate every Democratic or Republican candidate from zero; party labels carry information about likely issue positions, coalition partners and governing behavior. A new party must first create that brand.

An installed organizational base exists. Established parties possess state and local committees, recurring donor relationships, candidate pipelines, activist networks, consultants and legal expertise. Party-network research shows that much of what a party does occurs through distributed relationships rather than one centralized command.

Past votes can produce future legal advantages. Ballot-retention rules and presidential public-financing formulas may explicitly use prior vote share as evidence of party status or entitlement. The FEC's 5–25% prior-election public-funding rule is a direct federal example.

Existing officeholders create a recruitment network. A legislator, governor, mayor or activist can advance inside an existing party and inherit allies, endorsements and constituent networks. A new party must build the first generation without already having the second.

Donors and campaign professionals coordinate on the same focal points as voters. A major-party nomination is not merely a philosophical affiliation; it is a credibility signal that millions of other voters will recognize the candidate as potentially viable.

The parties can absorb challengers. This is perhaps the strongest path-dependent defense of the American duopoly. Instead of being ideologically fixed entities confronting new parties, the Democratic and Republican coalitions can change membership and policy. Insurgent factions therefore frequently find it more efficient to capture an existing ballot line than to create a third one.

The last mechanism deserves particular emphasis because it cuts both ways normatively. It is an entry barrier at the organizational level but an entry channel at the ideological level. A political entrepreneur may be prevented by incentives from creating Party C while nevertheless having a real route to transform Party A.

That is one reason the United States cannot straightforwardly be described as a system where voters have only two immutable ideological options. The organizations are old; their internal coalitions are not fixed.

Intent also needs careful treatment. A rule can be:

1. deliberately enacted;
2. advantageous to incumbents;
3. sincerely justified as administrative regulation;
4. and structurally exclusionary in effect,

all at the same time.

Conversely, observing incumbent advantage does not prove suppressive intent. Williams documents a system severe enough to violate constitutional rights, while Jenness and Burdick document the judicial conclusion that states may impose meaningful restrictions without those restrictions being unconstitutional suppression.

The historical record therefore suggests a mixed explanation:

Intentional rules matter, especially ballot qualification, party-retention rules, public-finance thresholds and debate criteria.

Historical development matters more broadly, because the Democratic and Republican organizations accumulated positions before today's contestants arrived.

Rational strategic behavior is essential, because voters, donors, candidates and organizations respond to probabilities of winning.

Network effects are substantial, because every actor's willingness to support an entrant depends partly on beliefs about everyone else's willingness.

Resource requirements matter, because national campaigns require enormous complementary infrastructure even if all laws were formally party-neutral.

Voter coordination is the final transmission mechanism, because electoral strength ultimately depends on people being willing to cast ballots for an entrant despite uncertainty about whether others will do so.

No deliberate suppression hypothesis is needed to generate the equilibrium.

Is the result a “false choice”? A more precise normative test

“False choice” is too imprecise to do the analytical work required here.

A political choice can be narrow, path-dependent, strategically constrained or structurally filtered while still being genuine. If A and B are independent organizations, advance meaningfully different coalitions and programs, can defeat one another, and election authorities honestly count votes, then a choice between them is not “false” merely because a voter would prefer C.

The term becomes defensible only in a stronger sense—when nominal alternatives are not truly autonomous, opposition cannot actually win, voters are coerced, results are fabricated, or the permitted candidates are controlled so that apparent competition cannot change political power.

A useful four-part classification can be made operational.

This framework draws on the democratic distinction between contestation and mere formal voting associated with Robert Dahl's work, and on contemporary election-integrity research distinguishing unequal competition from elections in which uncertainty and opposition rights have largely disappeared.

Under these criteria, the United States is best characterized at the national level as:

a competitive electoral democracy with high barriers to new-party entry and a substantially structurally restricted party market, not a generally noncompetitive electoral system.

The evidence for genuine competition between the two major parties is overwhelming: control of the presidency, Congress, governorships and state governments can and does change through elections. The evidence for restricted organizational entry is also strong: single-member plurality rules, state-by-state access, strategic voting and accumulated networks strongly favor operating through one of the two established coalitions.

That leads to a more accurate normative phrase than “false choice”:

The American system offers real choices through a structurally compressed party menu.

Those two propositions are compatible.

There are also elections where the distinction changes at a smaller scale. A district in which only one serious candidate appears may be practically noncompetitive in that election even though the national electoral regime remains competitive. Conversely, a close Democratic-Republican election may be intensely competitive despite offering little satisfaction to a voter outside both coalitions.

Awareness of the barriers and democratic legitimacy

The final part of the inquiry has a substantially thinner evidence base than the institutional analysis.

There is extensive research on perceived representation, political efficacy, election fairness, winning and losing, trust, satisfaction with democracy and institutional reform. There is much less research that randomly exposes Americans to information specifically about ballot-access laws, debate thresholds or Duverger's law and then tracks political attitudes. It would therefore be an error to claim that scholarship has demonstrated a simple causal chain:

learning about third-party barriers → losing trust → becoming anti-system.

What can be said more confidently is narrower.

Political efficacy. External efficacy is lower when citizens believe public institutions are unresponsive and that their choices cannot influence government. A voter who concludes that no feasible candidate represents them may therefore experience lower efficacy. But whether learning about third-party structural barriers itself causes that belief has not been established nearly as strongly as the broader efficacy relationship.

Trust and legitimacy. Comparative research consistently finds a “winner-loser gap”: people whose preferred side loses elections tend to report lower satisfaction with democratic performance. Christopher Anderson and Christine Guillory's cross-national research additionally linked majoritarian and consensus institutions to differing patterns of democratic satisfaction. Later electoral-integrity research, including Pippa Norris's work, finds that perceived procedural fairness is closely related to legitimacy. These results imply that a system perceived to make preferences ineffective can damage support, but they do not establish that multipartyism always increases trust.

Participation. Structural barriers can redirect participation rather than simply suppress it. A prospective Green, Libertarian, socialist, nationalist or other movement activist may participate intensely—but in a Democratic or Republican primary. Strategic voting similarly changes which candidate receives a vote without necessarily reducing turnout. Comparative studies often associate proportional systems and greater electoral competitiveness with participation, but institutional effects are intertwined with registration systems, compulsory voting, political culture and campaign mobilization.

Support for electoral reform. Evidence here is stronger. Citizens disadvantaged by existing rules, electoral losers and supporters of smaller parties tend to be more receptive to institutional change. That is partly principled and partly self-interested; supporters of incumbent arrangements likewise often become less enthusiastic about reform when reform might weaken their side. The implication is that awareness of entry barriers can plausibly increase support for ranked-choice voting, proportional representation, easier ballot access or debate reform, but support is mediated by partisan calculations.

Anti-system attitudes. Low efficacy, perceived unresponsiveness and distrust correlate with populist and anti-establishment sentiment. Causal direction is difficult: people with pre-existing anti-system attitudes are also more likely to perceive ordinary institutions as illegitimate. Treating third-party barriers as a demonstrated cause of extremism would therefore go beyond the evidence.

The distinction between actual unfairness and perceived unfairness is normatively important. Explaining the strategic logic of plurality elections can produce two opposite reactions. Some citizens may conclude, “The system prevents my real preferences from being represented.” Others may conclude, “Now I understand why broad coalitions form before elections rather than afterward.” Both interpretations are compatible with the same institutional facts.

Evidence supporting and challenging the structural-exclusion hypothesis

The strongest evidence supporting it is cumulative.

The Supreme Court has found some ballot-access structures unconstitutionally discriminatory toward new parties.

Single-member plurality generates a well-established strategic pressure toward two viable competitors.

Some public benefits are explicitly conditioned on prior support, producing path dependence.

Debate sponsors may condition exposure on polling and prior ballot qualification, creating a measurable viability feedback.

Established parties possess accumulated brands, networks and candidate pipelines that new parties must recreate.

Comparative proportional systems make it substantially easier for smaller parties to translate modest vote shares into legislative representation; Germany's official seat-allocation rules make the contrast particularly clear.

The strongest evidence challenging a maximal version is equally important.

Reasonable ballot regulation has repeatedly been upheld; states have legitimate administrative interests.

Private federal campaign-finance law does not categorically reserve fundraising or independent spending for major-party candidates.

The U.S. two-party tendency predates the modern ballot-access regime and has survived repeated changes in which parties occupy the two major positions.

The major parties are permeable enough that outsiders can compete for their nominations; political entrepreneurship is therefore often absorbed rather than prohibited.

Plurality systems such as Canada's can sustain additional parties when social and geographic conditions differ, showing that electoral rules create incentives rather than deterministic outcomes.

Exceptionally well-resourced outsiders can achieve enormous publicity and substantial popular support, yet still fail to convert that support into office—evidence that structural seat-conversion and coordination mechanisms matter more than a simple story of media blackout or lack of money.

And, most importantly, Democrats and Republicans are not sham competitors. They represent distinct coalitions, routinely attempt to defeat each other and can exchange control of government through elections.

The resulting judgment is therefore deliberately asymmetric:

The weak claim is strongly supported: U.S. institutions make durable new-party viability considerably harder than a surface reading of “anyone can run” would imply.

The stronger claim is also substantially supported: those institutions create self-reinforcing advantages for existing major parties even without suppressive intent.

The very strong claim is not supported: there is insufficient evidence that the system's two-party equilibrium is primarily the product of a coordinated deliberate project to prevent voters from having meaningful alternatives.

The “false choice” claim is generally too strong: the major-party alternatives are genuine, but the organizational menu through which political differences are expressed is structurally compressed.

The deepest causal conclusion is that American political competition is open at the level of ideas and factions more than it is open at the level of durable party organizations. An insurgent ideology can capture an existing party far more easily than a new organization can become a stable third major party. Plurality elections then reward that adaptation: movements learn to fight for control of one of two established coalitions, voters learn that supporting one of those coalitions is usually the safest way to affect who governs, donors and professionals follow the same logic, and every round of rational adaptation makes the next round of two-party coordination more rational still.