DAO Governance 2026: 1% Holds 90% Power, Turnout Stalls at 12%
On-chain data reveals that DAO governance remains deeply plutocratic despite billions in treasuries, with median participation at 5-12% and the top 10% of holders controlling 76.2% of voting power.
The numbers land like a hammer on glass. Across the major DAOs operating in 2026, median voting participation hovers somewhere between 5% and 12% of eligible governance tokens [^claim_697]. Do the subtraction yourself: that means 88% to 95% of token holders never cast a vote. Meanwhile, the top 10% of token holders control 76.2% of all voting power, and over 90% of voting power is concentrated in less than 10% of voters [^claim_698]. This is not a bug in the software. It is not a failure of community outreach. It is the logical, predictable outcome of one-token-one-vote rules—a mechanism that an ACM study formalizes as inherently plutocratic [^claim_699]. Even concave voting rules like quadratic voting cannot fully remove plutocracy without introducing other attack surfaces; the study recommends bicameral governance combined with Sybil-resistant identity [^claim_699]. The math is not neutral. It was never designed to be.
These governance realities matter because the stakes are enormous, and they are denominated in real dollars. DAOs collectively oversee over $26 billion in on-chain treasuries, with Lido DAO alone holding $1.4 billion [^claim_700][^claim_711]. Uniswap DAO’s on-chain governance has executed fee-switch proposals with turnouts between 62.84M and 77.83M UNI voting ‘For,’ and a ‘UNIfication’ proposal passed with 125.34M UNI ‘For’ [^claim_701]. It also voted to return 12.5 million loaned UNI tokens worth roughly $42 million [^claim_702]. Aave DAO approved the ‘Aave Will Win’ plan, requesting $42.5 million in funding and routing all branded product revenue back to the DAO treasury [^claim_703]. It also renewed BGD Labs as a core service provider, explicitly treating the on-chain approval as a binding agreement [^claim_704]. Lido DAO voted on a treasury stablecoin strategy and a NEST proposal to use stETH for LDO buybacks [^claim_706]. These are not abstract signaling exercises. They are binding financial decisions with real-world consequences.
Protocols are experimenting with governance frameworks to manage these decisions, building layers of process in an attempt to distribute power more evenly. Arbitrum DAO uses a two-stage process: an off-chain Snapshot temperature check, then an on-chain Tally vote requiring a simple majority and a quorum of 4.5% (constitutional) or 3% (non-constitutional) of votable tokens [^claim_707]. BeamX DAO requires staking before a two-week voting epoch, locks tokens until the epoch ends, and uses fixed quorums of 5–20% for parameter changes and 15–40% for governance-mechanism changes [^claim_708]. Yet even these designs do not address the underlying concentration: about 1% of token holders wield roughly 90% of voting power, and many DAO treasuries are still controlled by 3–5 multisig wallets [^claim_710]. The architecture of participation has been refined, but the architecture of power remains stubbornly intact.
The gap between theory and practice is wide, and it is getting wider. DAOs are using governance to decide deep structural changes—legal wrappers like Uniswap’s Wyoming DUNA [^claim_709], protocol upgrades like Aave V4 [^claim_705], and multi-million-dollar treasury allocations—but the electorate is tiny and concentrated. For DeFi protocols, this means that control over fee switches, oracle configurations, and cross-chain liquidity routing can pivot on the preferences of a narrow cohort of whales and delegates. Any mechanism design that assumes a broad, engaged electorate is ignoring the data. The data is telling us something uncomfortable: the promise of decentralized governance is being hollowed out by the very mechanisms meant to enable it. The question is not whether DAOs can work. The question is whether they can work for anyone other than the people who already hold the most tokens.
Provenance ledger
15 claims web-citedEvery claim below cites a source URL, and each URL was checked for validity before publish. The excerpt shown is the researcher's own summary of the page — it is not re-derived from the source, so it is not a verified verbatim quote. Follow the link to confirm any claim against the original. Citation markers in the text jump here.
[1] Across major DAOs in 2026, median voting participation is reported at 5–12% of eligible governance tokens, indicating that the vast majority of token supply does not vote in on‑chain decisions. web-cited
"Median voting participation across major DAOs: 5-12% of eligible tokens (ACM DAO Governance Review)... Think about that: in a typical DAO, 88-95% of token holders never vote."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[2] In ten major DAOs studied, over 90% of voting power is controlled by less than 10% of total voters, and the top 10% of token holders control 76.2% of all voting power, showing strong plutocratic concentration despite nominally open governance. web-cited
"The top 10% of token holders control 76.2% of all voting power (ScienceDirect: Who Controls DAOs?)... Over 90% of voting power is controlled by less than 10% of total voters in ten major DAOs."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[3] An ACM study on DAO governance formalizes that standard ‘one token, one vote’ rules make voting power proportional to wallet balance and are therefore plutocratic; the study shows that even concave (quadratic‑style) voting rules cannot fully remove plutocracy without introducing other attack surfaces, leading to recommendations for bicameral governance (one concave chamber, one linear chamber) combined with Sybil‑resistant identity. web-cited
"The dominant rule, voting power proportional to wallet balance, concentrates control among a small number of large holders... We define... a plutocratic voting rule... the most pragmatic defenses accept that concave voting alone is insufficient and combine it with a second mechanism whose failure modes are disjoint. Bicameral governance is one instance: a proposal must pass through two chambers with different rules, typically one concave and one linear... Proof of Personhood and other anti-Sybi
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[4] Industry analytics for 2026 report that DAOs collectively manage over $25–26 billion of on‑chain treasuries, with more than 5,000 DAOs tracked and Lido DAO’s treasury alone around $1.4 billion, making large treasury allocation and risk‑management votes a central on‑chain governance activity. web-cited
"Current industry data indicates that DAOs collectively oversee over $25 billion in treasury assets... More than 5,000 DAOs are now traceable through major analytics platforms" and "DAOs collectively control more than $26B in onchain treasuries, Lido ($1.4B) the largest individual treasuries per DeepDAO's ..."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[5] Uniswap DAO’s on‑chain governance executed a sequence of fee‑switch and structure votes in 2025–2026, including three ‘Protocol Fee Expansion’ proposals (Votes 1–3) with turnouts between roughly 62.84M and 77.83M UNI voting ‘For’ and 0–10.98M ‘Against, and a ‘UNIfication’ proposal that passed with 125.34M UNI ‘For’ and 742.5K ‘Against, illustrating that major protocol revenue‑share changes are now decided directly on‑chain with eight‑figure token participation. web-cited
"Protocol Fee Expansion: Vote 1... Executed... 62.84M For, 4.97K Against... Protocol Fee Expansion: Vote 2... 77.83M For, 10.98 Against... Protocol Fee Expansion: Vote 3... 72.98M For, 0 Against... UNIfication... Executed 2:04 pm Dec 28, 2025... 125.34M For, 742.5 Against."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[6] Uniswap DAO also executed an on‑chain proposal to return 12.5 million previously loaned UNI governance tokens (about $42 million at the time) from the Uniswap Foundation and top delegates back to the DAO’s governance timelock, with early tallies showing 53% ‘For’ and 46% ‘Abstain’ and almost no explicit ‘Against’ votes, demonstrating that temporary delegation programs can be unwound via on‑chain vote when governance liquidity is deemed sufficient. web-cited
"The digital cooperative governing Uniswap... is voting on a proposal to take back some $42 million worth of governance tokens... Between 2022 and 2023, Uniswap’s... DAO... loaned out 12.5 million UNI... 'UNI holders have been actively delegating voting power... passed proposals have averaged roughly 75 million votes in turnout'... So far, 53% of votes have been cast in favour, 46% voting to abstain, and a negligible amount against the proposal." and the Agora execution record: "Return 12.5M Del
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[7] Aave governance approved the ‘Aave Will Win’ plan in 2026, which requests $42.5 million in funding for Aave Labs and routes all branded product revenue back to the DAO treasury, effectively shifting protocol control and future cashflows under DAO oversight while funding core development via a time‑bounded budget. web-cited
"The Aave community approved the 'Aave Will Win' plan, strengthening DAO control over the protocol. Revenue from Aave applications now flows ..." and from the proposal discussion: "Aave's 'Will Win' proposal requesting $42.5 million for Aave Labs with all branded product revenue routed back to the DAO Treasury narrowly passed its first governance vote."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[8] In November 2025, the Aave DAO passed an on‑chain proposal renewing BGD Labs as a core security and development service provider until April 1, 2026 and explicitly treating the on‑chain approval as a binding agreement between the DAO and BGD Labs, illustrating that DAOs are using governance to enter quasi‑contractual service relationships. web-cited
"Present to the Aave DAO a proposal from BGD Labs to renew our involvement as a development and security coordinator services provider until 1st April 2026... The approval of this proposal by Aave governance acts as a binding agreement between the Aave DAO and BGD Labs."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[9] Aave V4, whose temp‑check passed Aave governance in May 2024, is being rolled out through 2025–2026 as an on‑chain protocol upgrade that replaces per‑chain liquidity pools with a two‑layer architecture of a shared ‘Liquidity Hub’ plus multiple ‘Borrow Spokes,’ introduces unified borrow rates and spoke‑level risk premiums, and plans cross‑chain liquidity fungibility via CCIP governed by future votes. web-cited
"Aave V4 is the largest architectural revision... The short version: V3 is one liquidity pool per chain... V4 is a 'Liquidity Hub' with 'Spokes'... Aave V4... shipping in phases through 2025–2026, restructures the protocol around two layers: a Liquidity Hub and Borrow Spokes... Unified borrow rate... Liquidity premiums... Cross-chain liquidity fungibility... via CCIP... The temp-check passed governance in May 2024; initial Hub deployments began on testnet in Q4 2024 with mainnet rollout phased t
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[10] Lido DAO conducted multiple treasury‑focused governance votes in 2025, including a ‘treasury stablecoin strategy’ vote scheduled from 24 November to 1 December 2025 that could change how treasury funds and risk are managed, and a ‘NEST’ proposal that builds a technical framework to use stETH from the Lido treasury to perform LDO buybacks, with activation gated behind a separate on‑chain vote. web-cited
"Lido has announced that voting is now open for the NEST proposal, which aims to establish a technical framework for potential LDO token buybacks. The proposal suggests using stETH from the Lido DAO treasury... The NEST proposal is not a direct buyback plan but rather a foundational step... its activation requiring a separate on-chain vote." and "The Lido DAO is voting on a plan involving stablecoin strategy for its treasury from 24 November to 1 December 2025. The proposal could change how the
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[11] Arbitrum DAO governance uses a two‑stage process where proposals first go to an off‑chain Snapshot ‘temperature check’ poll and, if they pass, proceed to an on‑chain Tally vote that must satisfy both a simple majority of votes in favor and a minimum quorum threshold of 4.5% of votable tokens for constitutional proposals or 3% for non‑constitutional proposals. web-cited
"If the proposal passes the temperature check, it will move on to an on-chain vote facilitated by Tally... To pass this stage, the proposal must meet two thresholds: 1. The proposal must receive more votes in favor than against; and 2. Constitutional AIPs must receive votes from at least 4.5% of votable tokens; non-Constitutional AIPs must receive votes from at least 3% of votable tokens."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[12] BeamX DAO’s governance framework requires BEAMX tokens to be staked before the start of a two‑week voting epoch to be eligible, locks voting tokens until the epoch ends, and uses fixed quorums for ‘special’ proposals between 5–20% of circulating supply for parameter and treasury changes (Type A) and 15–40% for governance‑mechanism changes (Type B), with one BEAMX equaling one vote and absolute‑majority passage. web-cited
"The standard voting cycle (voting epoch) lasts for two weeks or 20,160 blocks... To vote in the current epoch, users must have staked their BEAMX tokens before it begins and cannot withdraw their tokens until the voting epoch ends... One BEAMX token equals one vote, and each proposal must pass via an absolute majority... The initial quorum ranges between 5%-20% of the current BEAMX circulating supply for Type A proposals and 15%-40% of BEAMX circulating supply for Type B proposals."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[13] Uniswap DAO’s on‑chain ‘DAO Principles’ vote in January 2025 formally ratified a written set of principles for the DAO, while a separate on‑chain proposal in September 2025 established ‘Uniswap Governance as DUNI, a Wyoming DUNA’ with 52.97M UNI voting ‘For’ and 0.24M ‘Against, signaling that legal wrapper and meta‑governance structure changes are being decided directly by tokenholders. web-cited
"This onchain vote confirms a set of principles (the Principles) for the Uniswap DAO." and from the Agora governance log: "Establish Uniswap Governance as 'DUNI,' a Wyoming DUNA... executed... 52.97M For, 0.24 Against."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[14] An overview of DAO voting behavior in 2026 notes that typical voter turnout is only 5–15% and that around 1% of token holders often wield about 90% of voting power, with many DAO treasuries still controlled via 3–5‑signature multisig wallets rather than fully on‑chain governance, highlighting a persistent gap between the theoretical and actual decentralization of DAO control over treasuries and upgrades. web-cited
"A mere one percent of token holders wield approximately ninety percent of the voting power, while voter engagement typically hovers between five and fifteen percent. Treasury management in many DAOs still depends 3-5 signature, which is significantly below the separation-of-duties standards..."
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
[15] As of Q1 2026, DAOs are reported to control more than $26 billion in on‑chain treasuries, with governance decisions typically covering protocol parameter changes, grant programs, liquidity incentives, and treasury asset allocation, and with individual large DAOs such as Lido, Uniswap, and Aave each responsible for billions in cumulative governance‑directed flows. web-cited
"As of Q1 2026, DAOs collectively control more than $26B in onchain treasuries, Lido ($1.4B) the largest individual treasuries per DeepDAO's ..." and explanatory text describing DAO treasury spend via governance.
This excerpt was not re-derived from the source page, and may paraphrase or condense it. Check the source before relying on it.
Sources
- https://blockeden.xyz/forum/t/5-12-dao-voting-participation-in-2026-decentralized-governance-or-oligarchies-with-token-holder-th
- https://arxiv.org/html/2605.18990v1
- https://www.dlnews.com/articles/defi/daos-keep-centralizingdecades-of-governance-research-explain-why/
- https://vote.uniswapfoundation.org
- https://www.dlnews.com/articles/defi/uniswap-dao-votes-to-take-back-loaned-uni-tokens/
- https://finance.yahoo.com/markets/crypto/articles/aave-ends-month-long-wait-114215826.html
- https://vote.onaave.com/proposal/?proposalId=404
- https://eco.com/support/en/articles/14800886-aave-v3-vs-v4-what-changed-and-why-it-matters
- https://phemex.com/news/article/lido-opens-voting-on-nest-proposal-for-ldo-token-buyback-system-21305
- https://docs.arbitrum.foundation/how-tos/vote-dao-proposals
- https://www.beam.mw/docs/ecosystem/overview/beamx-dao-governance-framework
- https://vote.uniswapfoundation.org/proposals/78
- https://eco.com/support/en/articles/14799687-dao-treasury-management-onchain-governance-spend