Why Not Fusion Voting?¶
A Position Paper on the Limits of Cross-Endorsement Reform and the Case for Structural Electoral Architecture¶
Published March 2026¶
Fusion voting -- also called cross-endorsement or multiple party nomination -- has reemerged as a prominent reform proposal in American electoral politics. The Center for Ballot Freedom, Protect Democracy, and allied organizations are investing significant litigation and legislative resources into restoring the practice across the states. Their argument is that fusion voting is the "missing link" needed to counter polarization, empower minor parties, and break the two-party doom loop.
This paper argues that fusion voting, while not without modest historical merit, is a workaround that accepts the disease while treating one symptom. It is structurally dependent on the very electoral system it claims to improve and incapable of producing the multiparty governance its advocates promise. More importantly, substantial reform energy directed at fusion represents an opportunity cost -- resources that could instead support integrated electoral architecture changes that address the root causes of American democratic dysfunction.
The Federal Elections Modernization Act (FEMA), hereafter referred to as "the Act," provides that integrated architecture: STAR voting, multi-member districts with proportional representation, algorithmically neutral districting, House expansion, and Joint Endorsement Lists -- reforms that render fusion voting not merely unnecessary, but structurally irrelevant.
What Fusion Voting Is¶
Under traditional (disaggregated) fusion voting, a candidate may receive the nomination of more than one political party and appear on the ballot multiple times -- once per nominating party. Votes cast on each party line are tallied separately, then aggregated to produce the candidate's total. If Claire Farmer receives 46% on the Democratic line and 4.8% on the Common Sense Party line, she wins with 50.8%, and her margin of victory is visibly attributable to the minor party.
This practice was once universal in the United States. Prior to the adoption of the Australian ballot in the 1880s and 1890s, parties printed their own ballots, and cross-nomination required no state permission. Beginning in the late 1890s, major parties -- particularly Republicans in the North and Democrats in the South -- used their newly acquired control over state-printed ballots to ban fusion, explicitly to eliminate the coalition threats that Populist, labor, and multiracial fusion tickets posed to their dominance.
Today, traditional fusion remains legal and regularly practiced only in Connecticut and New York. A weaker variant called "dual labeling" -- where a candidate lists multiple party endorsements on a single ballot line without separate vote tallies -- exists in Oregon and Vermont. Efforts to restore fusion through litigation and legislation are underway in New Jersey, Michigan, Kansas, and other states, driven primarily by the Center for Ballot Freedom and allied reform organizations.
The Strongest Case for Fusion¶
Intellectual honesty requires engaging with the best version of the fusion argument, not a strawman. The case for fusion rests on several genuine observations.
First, fusion has a legitimate historical pedigree. The practice was not a fringe experiment but a mainstream feature of American elections for over a century. Its suppression was not motivated by good governance concerns but by naked partisan self-interest. As one Republican legislator stated during the fusion ban era, the intent was to prevent opposition coalitions from forming, regardless of voter preferences. The racial dimension is especially significant: Republican-Populist fusion in the post-Civil War South enabled multiracial coalitions that elected governors and members of Congress, and their destruction was intertwined with the rise of Jim Crow.
Second, fusion addresses the spoiler problem within the existing plurality framework. Under winner-take-all elections, minor party candidates either don't run (rendering themselves invisible) or do run and risk splitting the vote against the major-party candidate closest to their position. Fusion creates a third option: minor parties endorse a major-party candidate on their own ballot line, giving voters the ability to express minor-party support without "wasting" their vote. The voter signals a preference; the minor party demonstrates measurable leverage; the candidate wins with a visible coalition.
Third, the Brennan Center for Justice and others have argued that fusion promotes voter turnout, encourages major-party differentiation, and gives organized blocs of voters tools to influence specific legislative priorities. New York's Conservative Party has historically pressured Republican candidates rightward, and the Working Families Party has claimed credit for policy outcomes like the 2004 state minimum wage increase.
These are not trivial arguments. But they are arguments about making a broken system slightly more tolerable -- not about fixing the system itself.
Why Fusion Falls Short¶
Structural Dependency on the System It Claims to Fix¶
The most fundamental problem with fusion voting is that it is conceptually dependent on the plurality, single-member-district electoral system. Every benefit fusion claims to provide -- solving the spoiler problem, giving minor parties measurable leverage, enabling protest votes that "count" -- is a response to pathologies generated by winner-take-all elections. Remove the underlying system, and fusion has nothing to do.
Under STAR voting, the spoiler problem disappears entirely. Voters score all candidates on a 0-5 scale; the two highest-scoring candidates advance to an automatic runoff determined by voter preference. There is no strategic reason to vote for a "safe" major-party candidate instead of your actual preference -- the scoring mechanism handles preference aggregation directly. Fusion's primary selling point -- vote for the minor party without wasting your vote -- becomes irrelevant because no vote is wasted in the first place.
Under proportional representation with multi-member districts, minor parties do not need to piggyback on major-party candidates to achieve relevance. A party with 20% support in a five-seat district wins approximately one seat. The endorsement-leverage model that fusion provides -- "we gave you 4.8% of your winning margin, now listen to us" -- is replaced by something categorically more powerful: actual legislative votes. A minor party with seats in Congress does not need to negotiate from the position of a junior endorser hoping for policy concessions; it negotiates as a coalition partner whose votes are required to govern.
Fusion, in other words, does not merely exist within the plurality/SMD framework. It requires that framework to justify its existence. It is a patch, not a solution.
The New York Problem¶
Fusion's advocates point to New York as proof of concept. But New York's experience over more than a century raises as many concerns as it resolves.
The Liberal Party -- for decades one of New York's fusion parties -- devolved into what observers widely regarded as a patronage operation, bargaining ballot-line endorsements for payoffs rather than advancing a genuine policy agenda. The Brennan Center for Justice acknowledged this directly in its 2006 primer on fusion, noting that the Liberal Party "bargained for patronage for family members and associates of party leaders." The Center's response -- that the political marketplace eventually dispatched the Liberal Party and replaced it with the more issue-oriented Working Families Party -- is fair, but it does not eliminate the structural incentive that produced the corruption in the first place. Fusion creates an asset (the ballot line) that has value independent of the party's policy mission, which generates ongoing temptation toward rent-seeking behavior.
More recently, commentators have noted that New York's ballot access thresholds create a dependency relationship that undermines the theoretical leverage fusion is supposed to provide. The Working Families Party must cross a vote-received threshold every gubernatorial cycle to maintain its ballot line, which in practice means it has little choice but to endorse the Democratic nominee regardless of that nominee's alignment with the party's priorities. The leverage runs in one direction: the major party can take the minor party's endorsement for granted because the minor party cannot afford to withhold it.
If fusion has not broken the two-party doom loop in New York after more than a century of continuous operation, the burden of proof falls on advocates to explain why restoring it in other states would produce a different result.
Media Invisibility¶
Fusion's leverage model depends on the public seeing disaggregated vote totals -- knowing not just that a candidate won, but that a specific minor party's votes provided the margin of victory. In practice, media rarely report results at this level of granularity. News coverage reports winners and losers, vote percentages, and occasionally margins. The party-line breakdown that makes fusion meaningful is available to political operatives and close observers, but largely invisible to the general electorate. This means fusion's signaling function -- the mechanism through which minor parties are supposed to gain public leverage -- is substantially degraded in the real-world information environment.
A Uniquely American Accident¶
Fusion voting's intellectual weakness is revealed by a simple comparative question: does any significant movement for fusion voting exist outside the United States?
It does not. And the reason is structural, not accidental. Other democracies that operate plurality, single-member-district systems -- the United Kingdom, Canada, India -- experience the same Duverger pressures that make minor parties irrelevant under American plurality rules. But none of them have responded by developing fusion-style ballot mechanisms. Instead, they use informal electoral alliances (India's pre-electoral coalition system), strategic withdrawals in two-round systems (France's New Popular Front in 2024), or list-linking mechanisms within proportional systems (the Dutch lijstverbinding, Swiss apparentement). In most cases, minor parties in these systems advocate for proportional representation -- they push to change the system rather than create workarounds within it.
The international democratic reform conversation simply does not include fusion voting. The concept is a second-order effect of a very specific set of American historical conditions: the shift from party-printed to government-printed ballots in the 1880s-1890s, the deliberate suppression of cross-nomination by major parties during that transition, the survival of fusion in a handful of states (principally New York), and the constitutional framework created by Timmons v. Twin Cities Area New Party (1997). No other democracy went through this sequence. Every other country that faced the problems fusion claims to solve either changed its electoral system or developed coordination mechanisms that don't require ballot-level intervention.
This should give pause to any reformer evaluating where to invest scarce political capital. Fusion is not a proven international model waiting to be adopted. It is a restoration project for a 19th-century American workaround to a 19th-century American problem that the rest of the democratic world solved by moving to different electoral architectures entirely.
The Affirmative Case: Integrated Electoral Architecture¶
The Act provides an alternative approach that addresses the root causes of the dysfunction fusion attempts to patch. Its reforms are designed as an integrated system where each component reinforces the others.
STAR Voting Eliminates the Spoiler Problem Entirely¶
STAR voting replaces plurality voting for all federal elections. Voters score candidates 0-5 stars; the two highest-scoring candidates advance to an automatic runoff determined by voter preference. This eliminates vote-splitting, removes the strategic incentive to vote for the "lesser evil," and enables honest preference expression across the full candidate field.
Where fusion gives minor-party voters a way to support a major-party candidate while signaling minor-party affiliation, STAR voting makes the signal unnecessary. Voters can score their preferred minor-party candidate a 5 and a tolerable major-party candidate a 3 without any risk that their honest preference will produce a perverse outcome. The scoring mechanism does what fusion attempts through ballot mechanics -- it lets voters express nuanced preferences -- but it does so mathematically rather than through institutional workarounds.
Proportional Representation Gives Minor Parties Actual Seats¶
Beginning at the First FEMA Election, all qualifying states elect Representatives from multi-member districts (3-7 seats) using STAR Proportional Representation (STAR-PR). Under this system, 20% voter support yields roughly 20% of seats. Minor parties do not need endorsement leverage or bargaining chips -- they have representatives who vote on legislation.
This is the difference between influence and power. Fusion gives minor parties influence: the ability to signal, to negotiate, to hope that their measurable vote share will be rewarded with policy concessions. Proportional representation gives them power: actual seats, actual votes, actual participation in the legislative process. No amount of ballot-line leverage can substitute for having members of Congress who answer to your voters.
Algorithmically Neutral Districting Eliminates the Boundaries Problem Fusion Ignores¶
Fusion voting operates inside gerrymandered districts and accepts the boundary distortions that make those districts uncompetitive. A minor party's 4.8% endorsement leverage means nothing if the district was drawn to produce a 20-point partisan margin -- the major-party candidate wins regardless, and the minor party's bargaining position evaporates.
The Act mandates algorithmically neutral, deterministic districting for all states electing two or more Representatives. The Electoral Science Office certifies the algorithm, which operates without reference to racial, ethnic, linguistic, partisan, or incumbent-residence data and produces publicly verifiable outputs. Shortest splitline serves as the statutory default pending ESO certification. No human draws a line. No partisan actor touches a boundary.
This addresses a dimension of electoral dysfunction that fusion cannot even conceptually reach. Fusion is a ballot-level intervention -- it changes how votes are counted within a district but has no mechanism for addressing whether the district itself was drawn to predetermine the outcome. Algorithmically neutral districting eliminates the manipulation at its source, and proportional representation ensures that even imperfect boundaries cannot distort representation, because communities do not need to be packed into a single district to win seats. The combination removes two layers of structural dysfunction that fusion leaves entirely intact.
House Expansion Creates the Mathematical Space for Proportionality¶
The Act grows the House from 435 to approximately 720 members over 14 years using the cube root rule. This expansion is not incidental -- it creates the seats that proportional representation needs to function with meaningful granularity. Multi-member districts of 3-7 seats in a 720-member House produce representation that accurately tracks voter preferences at a level of precision that single-member districts -- with or without fusion -- cannot approach.
Joint Endorsement Lists Capture Fusion's Informational Benefit Without Its Baggage¶
Title II of the Act establishes Joint Endorsement Lists: each candidate appears once on the ballot with up to three party endorsements displayed alongside their name.
Example ballot format: Sarah Chen | Registered: Democratic | Endorsed by: Democratic, Working Families, Green
This provides every informational benefit that fusion's advocates claim -- voters see which parties support which candidates, coalition structures are transparent, minor parties gain visibility through endorsement power -- without the structural problems fusion creates. There is no vote aggregation across party lines, no multiple ballot lines for the same candidate, no asset (the ballot line) that can be converted into patronage, and no dependency relationship created by vote-threshold requirements for maintaining party status.
Joint Endorsement Lists are explicitly distinguishable from fusion under existing constitutional doctrine. Timmons v. Twin Cities Area New Party addressed state authority to prohibit fusion -- multiple party nominations with vote aggregation. The Act's system displays endorsements, not nominations; candidates appear once, not multiple times; and the provision operates under federal Elections Clause authority rather than as a state-law challenge.
The Integrated System Makes Fusion Irrelevant¶
The Act's reforms are not a menu of independent options. They are an integrated system that must launch together. STAR voting eliminates spoiler effects. Multi-member districts translate preferences into proportional outcomes. Algorithmically neutral districting ensures those districts are drawn without human discretion or partisan manipulation. House expansion creates seats for proportional representation to fill. Joint Endorsement Lists make coalition structures transparent. If any major component launches without the others, the result is dysfunction at a different scale rather than genuine modernization.
This integrated approach is why fusion becomes not merely unnecessary but structurally irrelevant under the Act. Every problem fusion was designed to solve -- spoiler effects, minor party invisibility, coalition opacity -- is addressed directly by components of the Act that operate at the correct level of the electoral architecture. And the problems fusion cannot even reach -- gerrymandered boundaries, human discretion in map-drawing, the geometric distortions of winner-take-all -- are addressed by components fusion has no mechanism to replicate. Fusion patches the ballot. The Act rewrites the operating system.
The Opportunity Cost Problem¶
The fusion voting movement is not cost-free. The Center for Ballot Freedom and its allies are investing substantial resources -- legal fees, organizing capacity, political capital, media attention -- into state-by-state litigation and legislation to restore a practice whose maximum achievable outcome is modest improvement within a fundamentally broken system. Every dollar and every hour spent on fusion is a dollar and an hour not spent on structural reform.
This is not to say that fusion advocates have bad intentions. Many are sophisticated analysts who understand the limitations of fusion and view it as a stepping stone toward larger reforms. Lee Drutman has argued that fusion builds party infrastructure that can later advocate for proportional representation. This bootstrapping theory is plausible in the abstract, but the historical evidence does not support it. New York has had fusion for over a century. Its minor parties have not used their fusion-based leverage to achieve proportional representation, ranked-choice voting, or any other structural electoral reform at the state level. The stepping stone has not led to the next step.
Reform movements face a finite attention span in the public and in legislatures. The question is not whether fusion is better than the status quo -- it probably is, marginally. The question is whether fusion is the best use of the reform community's limited resources. The answer, measured against the scale of the problem and the availability of structural alternatives, is no.
It is worth noting that the scholarly literature itself reflects this gap. A systematic review of fusion voting scholarship reveals approximately two dozen favorable sources -- primarily law review articles and policy organization reports -- against fewer than five peer-reviewed critical treatments (principally Tamas 2017, Pope 1998, and Chamberlain 2012). More significantly, no published peer-reviewed study systematically compares fusion voting against proportional representation, STAR voting, ranked-choice voting, or approval voting using common evaluative criteria. The comparative case that fusion advocates implicitly make -- that fusion deserves reform energy over structural alternatives -- has never been subjected to the rigorous empirical testing that would justify the resource investment the movement demands. The fusion literature is largely a conversation among proponents, and the absence of comparative analysis is itself an indictment of the field's analytical rigor on this question.
Conclusion¶
Fusion voting is a historically interesting, modestly beneficial workaround for problems created by America's winner-take-all, single-member-district electoral system. Its advocates are correct that it was unjustly suppressed, that its suppression served partisan interests rather than democratic values, and that its restoration would provide marginal improvements in minor party viability within the existing framework.
But marginal improvement within a broken framework is not reform. Fusion cannot break Duverger's law. It cannot produce proportional representation. It cannot give minor parties legislative seats. It cannot eliminate the spoiler problem -- it can only create a workaround that allows voters to navigate around it. It cannot address gerrymandering or the human discretion embedded in boundary-drawing -- it operates inside whatever districts it finds. And it consumes scarce reform resources that could be directed toward structural changes that address root causes rather than symptoms.
The Federal Elections Modernization Act offers those structural changes: STAR voting to eliminate vote-splitting, multi-member districts with proportional representation to give every voter's preference proportional weight, algorithmically neutral districting to eliminate human discretion from boundary placement, House expansion to create the mathematical space for proportionality, and Joint Endorsement Lists to provide coalition transparency without fusion's institutional baggage. Together, these reforms produce a system in which fusion voting has nothing to do -- because the problems it was designed to solve no longer exist.
The question for the reform community is not "why not fusion voting?" in isolation. It is: given finite resources and political capital, do we invest in making the broken system slightly more tolerable, or do we invest in replacing it?
Works Cited¶
Argensinger, Peter H. "'A Place on the Ballot': Fusion Politics and Antifusion Laws." American Historical Review 85, no. 2 (April 1980): 287-306.
Chamberlain, Adam. "Fusion Ballots as a Candidate-Centered Reform: Evidence from Oregon." Social Science Journal 49, no. 4 (2012): 458-464.
Drutman, Lee. "How Fusion Voting Builds the New Parties That Can Break the Two-Party Doom Loop." Lee Drutman's Newsletter, November 18, 2025.
Michelson, Melissa R., and Scott J. Susin. "What's in a Name: The Power of Fusion Politics in a Local Election." Polity 36, no. 2 (January 2004): 301-333.
Morse, Adam, and J.J. Gass. More Choices, More Voices: A Primer on Fusion. Brennan Center for Justice at NYU School of Law, October 2006.
Pope, James Gray. "Fusion, Timmons v. Twin Cities Area New Party, and the Future of Third Parties in the United States." Rutgers Law Review 50 (1998): 473-506.
Pocasangre, Oscar, and Maresa Strano. "What We Know About Fusion Voting." New America, July 2024.
Ramos, Albert E. Federal Elections Modernization Act, Rev 5.2. The American Policy Architecture Institute, February 2026.
Rogers, Joel, and Maresa Strano. "More Than Semantics: Distinguishing Dual Labeling from Traditional Fusion Voting." Ballot Access News, September 16, 2023.
Scarrow, Howard A. "Duverger's Law, Fusion, and the Decline of American 'Third' Parties." Western Political Quarterly 39, no. 4 (1986): 634-647.
Soyer, Daniel. Left in the Center: The Liberal Party of New York and the Rise and Fall of American Social Democracy. Ithaca, NY: Cornell University Press, 2021.
Tamas, Bernard. "Does Fusion Undermine American Third Parties? An Analysis of House Elections from 1870 to 2016." New Political Science 39, no. 4 (2017): 609-626.
Timmons v. Twin Cities Area New Party, 520 U.S. 351 (1997).
Revision history available in the raw file.
Download this document (opens on GitHub -- click the download button)
Prepared by Albert Ramos for The American Policy Architecture Institute