Quadratic Voting

What Is Quadratic Voting?

Quadratic voting (QV) is a governance mechanism designed to fix a fundamental problem with standard token-weighted voting: the richest participants win every argument. Understanding what is quadratic voting in DAOs starts with a simple math rule — your nth vote on any proposal costs n² credits rather than n credits. One vote costs 1. Four votes cost 16. Ten votes cost 100.

That cost structure changes everything.

Economist Glen Weyl and programmer Eric Posner formalized the concept in their 2018 book Radical Markets, arguing that QV lets participants signal how much they care about an issue, not just which side they're on. In traditional one-token-one-vote systems, a whale with 10 million tokens can steamroll a proposal that 10,000 small holders collectively oppose. QV doesn't eliminate that asymmetry entirely, but it taxes it heavily.

How the Math Works

Think of it like buying influence at a price that compounds against you. In a session where each participant receives 100 voice credits:

A participant who burns all 100 credits on a single proposal to cast 10 votes has zero influence everywhere else. Compare that to someone who spreads 1 vote across 100 different proposals. The system rewards breadth of participation over concentration of power — which is precisely the point.

This is closer to how municipal bond referendums work in some US cities than how most crypto governance operates today.

Quadratic Voting vs Token-Weighted Voting

Standard token-weighted voting is blunt. If you hold 51% of governance tokens, you control 51% of outcomes. Full stop. I've watched protocols get captured by a single entity holding a supermajority, passing treasury withdrawals that clearly benefited insiders over the broader community.

Quadratic voting isn't a silver bullet, but it fundamentally changes the incentive structure:

• Token-weighted voting: 1 token = 1 vote. Linear scaling. Whale-friendly.
• Quadratic voting: Votes cost credits quadratically. Favors participants with moderate, distributed preferences over single-issue maximalists.

The DAO voting systems comparison analysis covers the full spectrum — including conviction voting, holographic consensus, and optimistic approval — if you want to see how QV fits within the broader governance design space.

Real-World Implementations

Gitcoin Grants is the most cited example. Gitcoin's quadratic funding rounds — a related but distinct mechanism — distributed millions of dollars based on the number of contributors rather than the size of contributions. Projects with many small donors received proportionally more matching funds than those with one large donor. In some rounds, a project with 500 contributors giving $1 each outperformed a project with 2 contributors giving $1,000 each.

Optimism's RetroPGF rounds have experimented with similar mechanics, with hundreds of badgeholders allocating OP tokens across public goods projects.

Snapshot supports QV natively as a voting strategy, meaning any DAO using Snapshot can enable it without writing custom governance contracts.

The Sybil Attack Problem

Here's the uncomfortable truth most QV advocates undersell: quadratic voting is only as good as your identity system.

If one whale can create 1,000 wallets and distribute 100 voice credits to each, they've just bypassed the quadratic cost entirely. The attack is trivially cheap on-chain. This is why QV and sybil resistance are inseparable concepts — you can't have one without the other.

Gitcoin has used Passport, a composable identity layer, to assign trust scores that gate quadratic funding eligibility. Projects like Worldcoin and Proof of Humanity attempt on-chain proof-of-personhood. None are perfect. All introduce their own centralization or privacy trade-offs.

Critical warning: Any DAO deploying quadratic voting without a robust sybil-resistance layer is creating an illusion of fairness, not the real thing. Budget for identity infrastructure before you budget for QV mechanics.

Myth vs Reality

Myth: Quadratic voting eliminates whale dominance.
Reality: It reduces it. A participant with 100x more credits than average still has 10x more effective votes (√100 = 10). The distribution is more equitable, not equal.

Myth: QV always leads to better governance outcomes.
Reality: It depends on participation quality. If most voice credits go unspent — which happens frequently in low-engagement DAOs — the system reverts to being dominated by whoever bothers to show up.

For context on why participation rates matter so much, the analysis on on-chain voting participation rates and their effect on DAO outcomes is worth reading before any governance redesign.

When Quadratic Voting Makes Sense

QV works best when:

1. The governance community is large and diverse (hundreds of participants, not dozens)
2. You have a workable sybil-resistance mechanism already deployed
3. Proposals vary significantly in their importance to different sub-groups
4. You want to surface minority preferences that might otherwise be drowned out

It's overkill for a 15-person multisig and potentially dangerous for a protocol with billions in TVL but no identity layer. The governance attack vectors in token-based DAOs are worth reviewing — many of them apply directly to poorly designed QV implementations.

Resources and Further Reading

• Radical Markets by Glen Weyl & Eric Posner — the foundational text on QV theory
• Gitcoin Grants documentation — real-world QV-adjacent implementation
• Snapshot voting strategies — where you can see QV deployed in live DAO governance
• Ethereum.org on DAOs — solid primer on DAO governance basics

Quadratic voting is genuinely one of the more intellectually honest attempts to solve plutocratic governance. But it's a mechanism, not magic. Deploy it without solving for identity, and you've just built a more expensive version of the same problem you started with.