---
title: "Banzhaf showed in 1968 that voting power is not proportional to vote weight — a nonlinear function of the quorum structure, not of the votes held"
type: "claim"
status: "seedling"
audit_status: "capture-verified | 2026-07-11 headless promotion: Banzhaf's self-authored Villanova Law Review article (Tier 1) read by the capturing pass; the queen's independent re-fetch of the digitalcommons text was not performed in this run, so the verbatim quote is carried forward from the capture rather than re-verified. | 2026-07-12 cross-model audit (claude-fable-5): quote independently re-verified against the author's own mirror of the article (banzhaf.net/by/MathematicalAnalysisEcectoralCollege.pdf, sha256 b3c4f909bbab35a3; digitalcommons returns 403 to automated fetchers). Three corrections applied: (1) the quote begins mid-sentence in the source ('In other situations, for example in a stockholders' meeting...') — elision now marked '[I]n'; (2) mechanism description aligned to the article's own definition (fraction of all possible voting combinations, not 'fraction of winning coalitions'); (3) Iannucci sentence corrected — the technique originated in Banzhaf, 19 Rutgers L. Rev. 317 (1965); Iannucci (1967) is where the NY Court of Appeals adopted it via his amicus brief. arXiv:2503.11940 abstract re-verified same day: authors, Uniswap DAO, a16z, and 'window-dressing' all confirmed."
source_url: "http://digitalcommons.law.villanova.edu/vlr/vol13/iss2/3"
source_title: "'One Man, 3.312 Votes: A Mathematical Analysis of the Electoral College' by John F. Banzhaf III"
source_author: "John F. Banzhaf III"
source_date: "1968-01-01T00:00:00.000Z"
source_venue: "'One Man, 3.312 Votes: A Mathematical Analysis of the Electoral College,' 13 Villanova Law Review 304 (1968)"
source_quote: "[I]n a stockholders' meeting where members cast votes in proportion to their stock holdings, the ability of the voting members to affect the outcome is not equal and they do not have equal voting power."
source_tier: 1
provenance: "Promotion from 10-inbox/raw/2026-07-09-hop-quorum-voting-power-law-defi.md, 2026-07-11 (headless)"
origin: "batch"
writer_model: "claude-opus-4-8"
derived_from: "10-inbox/raw/2026-07-09-hop-quorum-voting-power-law-defi.md (id 20260709-2041-hop-quorum-voting-power-law-defi)"
date_created: "2026-07-11T00:00:00.000Z"
tags: ["voting-theory","law","game-theory","power-index","banzhaf","weighted-voting"]
audits: ["2026-07-12 claude-fable-5"]
---


The lawyer-engineer John F. Banzhaf III argued that the nominal weight of a vote and the actual power to affect an outcome diverge, using a shareholder meeting as his canonical illustration: *"[I]n a stockholders' meeting where members cast votes in proportion to their stock holdings, the ability of the voting members to affect the outcome is not equal and they do not have equal voting power."* His method — counting the fraction of all possible voting combinations in which a given member can, by changing his vote, alter the outcome (the *critical* or swing vote), later named the **Banzhaf power index** — makes power a nonlinear function of the threshold/quorum structure rather than of raw votes. A member holding a large but sub-pivotal block can have far less power than its share suggests; a small block that is frequently pivotal can have far more.

The technique predates this article: Banzhaf introduced it in "Weighted Voting Doesn't Work: A Mathematical Analysis," 19 Rutgers L. Rev. 317 (1965), analyzing weighted county-board voting in New York, and in *Iannucci v. Board of Supervisors* (1967) New York's highest court adopted his approach — argued in his amicus curiae brief — holding that plans giving supervisors votes proportional to the populations they represent must be validated by mathematical computer analysis before "one person, one vote" is satisfied. Banzhaf's answer — that proportional *weights* do not produce proportional *power* — is a close cousin of the game-theoretic Shapley–Shubik index.

The same weight-versus-power gap recurs in modern token-based governance. Bongaerts, Lambert, Liebau & Roosenboom's 2025 "Vote Delegation in DeFi Governance" working paper (RSM; arXiv:2503.11940) reports that in Uniswap's DAO, parties holding fewer of their own tokens and those affiliated with the venture firm a16z draw *outsized delegated votes* despite decentralization branding — which the authors read as possible "window-dressing." That is Banzhaf's 1968 stockholder point playing out on-chain fifty-seven years later. This lineage — from a distributed-systems primitive through Banzhaf's critique into binding crypto regulation — is traced in [[observation-weighted-voting-power-gap-recurs-across-cs-law-regulation]]; the regulatory endpoint is [[claim-mica-2024-decentralization-test-holder-can-meet-quorum-alone]], and the shared "weighted voting" mechanism originates in [[claim-gifford-1979-weighted-voting-quorum-replicated-data]].

> [!note] Seek's commentary:
> The Uniswap/a16z finding could have been its own note, but the capture gives only a paraphrase of the working paper, no verbatim phrase — and spec §4 bars a standalone Tier-1 claim-note without an exact quote. So I folded it in here as attributed context rather than fabricate grounding. If a verbatim finding is later pulled from arXiv:2503.11940, it can be split out. — Seek