What it is
Quadratic voting lets participants express not just which option they prefer, but how strongly, while making it mathematically expensive to dominate a single decision. Each additional vote on the same choice costs the square of the number of votes cast — so 1 vote costs 1 "credit," 2 votes cost 4 credits, 3 votes cost 9 credits, and so on — spent from a fixed budget of voting credits each participant is given.
How it works
- Every participant is allocated a fixed budget of voting credits (or tokens) for a given decision or funding round.
- For each option or proposal, a participant chooses how many votes to allocate, and the credits spent equal the number of votes squared (n votes costs n² credits).
- This pricing curve means casting a single extra vote on an option you already voted for gets steadily more expensive, discouraging piling all your budget onto one choice.
- Participants instead tend to spread credits across several proposals they care about, or concentrate heavily only on the few they feel strongest about, accepting the quadratic cost.
- Votes across all participants are tallied per option (summing raw vote counts, not credits spent), and the highest-tallied options win or receive proportional funding.
- In quadratic funding variants (grant matching), the mechanism instead uses the sum-of-square-roots-then-squared formula across many small contributors to compute a matching pool allocation, rewarding proposals with broad small-donor support over those with a single large donor.
- Sybil resistance (proving each participant is a unique, un-duplicated identity) is a prerequisite, since splitting one identity into many resets everyone's cheap first vote.
Why designers use it
- Lets intensity of preference matter, unlike one-person-one-vote which can't distinguish "slightly prefer" from "desperately need."
- Blunts whale dominance relative to simple token-weighted voting, since buying overwhelming influence over one decision costs quadratically more, not linearly more.
- In grant funding, rewards proposals with genuine broad community support over ones bankrolled by a single wealthy backer.
- Produces more efficient aggregate outcomes in public-goods-style decisions where preference intensity carries real information.
Failure modes
- Sybil attacks are the central threat: without robust unique-identity verification, one actor splitting into many wallets gets many "cheap" first votes instead of paying the quadratic cost, and the whole mechanism's guarantee collapses.
- Collusion rings: coordinated groups can pool resources and split votes across many colluding identities to simulate broad organic support in quadratic funding rounds.
- Identity verification is hard and often invasive (KYC, biometrics, social graphs), trading a plutocracy problem for a privacy or gatekeeping problem.
- Voter confusion: the quadratic cost curve is unintuitive to most participants, leading to suboptimal or accidental vote allocation compared to simpler mechanisms.
- Gameable in small populations: with few participants, a single sophisticated voter who understands the math can still exert outsized influence over naive voters who don't budget their credits well.
What to check before using it
- What sybil resistance mechanism is actually in place (proof-of-personhood, social attestation, stake-based identity), and how well does it hold up against a motivated attacker?
- Is the credit budget and cost curve clearly explained to participants so votes reflect genuine preference rather than confusion?
- For funding rounds, is there collusion detection or anti-fraud tooling to catch coordinated small-donor rings gaming the matching formula?
- Is the population large and diverse enough for the statistical benefits of quadratic aggregation to actually hold, or is it a small enough group that gaming dominates?
- What's the fallback if identity verification fails or excludes legitimate participants (accessibility, false negatives)?