Proportional representation

The proposal

Proportional representation means electing several representatives from each district so that a party’s share of seats tracks its share of votes, replacing winner-take-all single-member districts. The variants are multi-member districts with the single transferable vote, mixed-member systems, and party-list systems. All differ from ranked-choice voting in single-member districts, which changes how one winner is chosen, not the seat math; nearly all American evidence concerns the second. Multi-member congressional districts were lawful for most of American history; a 1967 statute, 2 U.S.C. § 2c, ended them.¹ The standing bill, the Fair Representation Act, would repeal that rule; its sponsor and all seven cosponsors are Democrats, as were all eight on the prior version, and no Republican has cosponsored it since its 2017 introduction.² Our redistricting amendment covers the mid-decade re-draw ban only.

The evidence

No independent national poll with a party breakdown asks about multi-member districts or proportional seat allocation; searches of Pew, Gallup, AP-NORC, and seven other independent pollsters found none.

The only direct measurement is advocacy-commissioned. Verasight, for Fix Our House and New America (2026), found 47% to 21% for “proportional representation for U.S. Congressional elections,” and 50% to 22% for seats “in proportion to their share of the overall vote”; three in ten unsure. The party crosstabs are not merely un-located: the release publishes toplines and wordings only and directs readers to the authors for breakdowns, so no floor can be computed from the published record.³ The one partisan figure released is a share of a filtered subgroup, not a floor.⁴

The closest real-world evidence is about ranked-choice voting in single-member districts, not proportional representation. On November 5, 2024, statewide adoption measures lost by wide margins in Idaho (30.4% yes), Oregon (42.3%), Colorado (46.5%), and Nevada (45.7%, reversing 53% approval in 2022).⁵⁶ Nineteen states have banned ranked-choice voting, seventeen under single-party Republican control; in Ohio’s 2026 roll calls, 63 of 65 House votes in favor were Republican.⁷ The one proportional result is local: Portland, Oregon, where a 2022 charter measure carried 58.08% yes, though it bundled multi-member districts with a change in the form of city government, so it is not a clean read on the mechanism.⁸

Independent items with party splits are proxies, and the main one has moved. On Gallup’s question whether a third major party is needed, support ran 63% in 2023 (58% of Republicans, 46% of Democrats, 75% of independents), 58% in 2024 (48% of Republicans, 53% of Democrats, 69% of independents), and 62% in 2025 (43% of Republicans, 58% of Democrats, 74% of independents). The party lines crossed: the Republican reading fell 58 to 48 to 43 while the Democratic reading rose 46 to 53 to 58, and Gallup attributes the movement to which party holds the presidency. In the 2025 wave, 15% say they are very likely to vote for a third-party candidate (7% of Republicans, 9% of Democrats, 29% of independents), and 54% would switch to a major-party candidate if their preferred one looked unlikely to win.⁹ A deliberative University of Maryland survey (2021) found 61% approving ranked-choice voting for federal elections — 49% of Republicans, 55% of independents, 73% of Democrats.¹⁰

Where it stands

Verdict: Unpolled. No bipartisan floor can be computed, and the absence is now confirmed as a property of the published record rather than of our search.³ The nearest proxy floor is 43% of Republicans (Gallup, 2025), which was 46% of Democrats two years earlier; the nearest mechanism-adjacent floor is 49% of Republicans on ranked choice (Maryland, 2021).

• Staleness: cleared. Direct evidence is weeks old; ballots 2024; legislation and the proxy series through 2026.
• Framing dependency: failing. The same idea draws 24% cold, 47% by name, 50% by description, 61.5% after a metaphor, and 62% as a call for a third party.
• Abuse-vs-remedy gap: wide. 76% were concerned that 48% of a state’s votes could yield 65% of its seats (Verasight, 2026, advocacy-commissioned, no party breakdown published);³ the remedy has cleared nothing.
• Depth: untested on the mechanism, thin on the proxy. No survey has named the costs of proportional representation: larger districts, weaker constituent ties, coalition government. Where depth has been probed on the nearest proxy, support thins (Gallup, 2025).⁹
• Regime coding: demonstrated on the proxy. Proportionality cuts the seat premium of whichever party the current map favors, and the third-party proxy swapped floor parties across three waves (Gallup, 2023–2025), so any floor read off it is time-bound.⁹
• Symmetry: open. Neutral by construction, untested with voters; the legislative record is one-sided in both directions.
• Existential problem: arguably passes. Single-member plurality districts are what makes district-line gerrymandering possible, and 45% of adults identified as independents in 2025, a record (Gallup).¹¹
• Why-an-amendment: fails as things stand. Congress imposed single-member districts by statute and can lift them the same way.¹ The case for constitutional status is entrenchment — nineteen states have banned ranked choice by ordinary law; the counter is that district magnitude and seat formulas are machinery our drafts send to Congress by law.

Framings we tested

None clears the bar. By name, 47%; by description, 50%; several representatives for a larger area over one local representative, 61.1% (Verasight, 2026, advocacy-commissioned, no party breakdown published).³ After an explanation, 52% or 61.5%, from 24–27% cold; 59% had never heard of it (Citizen Data for Protect Democracy, advocacy-commissioned, no party breakdown published).¹² “A third major party is needed”: 62%, floor 43% of Republicans (Gallup, 2025) — a diagnosis, not a remedy, and one whose floor party has switched since 2023.⁹ “I often wish there were more parties to choose from”: 37%, floor 25% of Republicans (Pew Research Center, 2023).¹³ Ranked choice after briefing: 61%, floor 49% of Republicans (Maryland, 2021).¹⁰ Ranked choice at the ballot box: 30.4%–46.5% yes across four states, all defeated (2024).⁵⁶ The best-polling framings are not the rule that would be drafted.

What would move it

In: an independent national poll asking the mechanism in forward-looking, both-parties wording with crosstabs — e.g., “Would you favor or oppose electing House members from larger districts that each choose three to five members, so that a party winning 40% of the vote in an area wins about 40% of that area’s seats?” — with at least 60% in both parties, holding through a costs-named follow-up (larger districts, more parties, coalition bargaining). Statewide adoption of multi-member districts by ballot measure, or a cross-party roll call on the Fair Representation Act, would also count. None of that answers the why-an-amendment test: why a rule Congress can enact by repealing one 1967 sentence belongs in the Constitution.

Out: an independent poll putting either party below 45% on the mechanism, or a costs-named collapse, would make this a rejection on the floor and depth tests; further statewide defeats would do the same. A settled judgment that the why-an-amendment test has no answer would move the category to the statutory list.

← All candidates