---
title: "Curve DAO"
url: "https://caper.network/wiki/daos/dexs/curve-dao"
updated: 2026-09-10
license: CC-BY-4.0
license_url: "https://creativecommons.org/licenses/by/4.0/"
---

# Curve DAO

|  |  |
| --- | --- |
| **Name** | Curve DAO |
| **Type** | Protocol DAO (DeFi — stablecoin exchange + lending) |
| **Governance token** | CRV, vote-escrowed as **veCRV** (non-transferable) |
| **Governance model** | [Vote-escrow](/wiki/dao-governance/concepts/treasury/dao-tokenomics): lock CRV up to 4 years for time-decaying voting power; Aragon-based DAO; weekly gauge-weight votes set CRV emissions |
| **Products** | StableSwap AMM · crvUSD stablecoin (2023) · LlamaLend lending |
| **Notable for** | Pioneered vote-escrow (ve) tokenomics; the **"Curve Wars"**; one of the longest-lived Aragon DAOs |
| **Primary sources** | [curve.finance](https://curve.finance/), [docs.curve.finance](https://docs.curve.finance/user/curve-tokens/vecrv), [gov.curve.fi](https://gov.curve.finance/), [curve-dao-contracts](https://github.com/curvefi/curve-dao-contracts) |
| **Related** | [Uniswap DAO](/wiki/daos/dexs/uniswap-dao), [Aave DAO](/wiki/daos/lending/aave-dao), [Token-weighted voting](/wiki/dao-governance/concepts/voting/token-weighted-voting), [DAO tokenomics](/wiki/dao-governance/concepts/treasury/dao-tokenomics) |

The **Curve DAO** governs [Curve Finance](https://curve.finance/), a decentralized exchange specialised in low-slippage swaps between like-priced assets (stablecoins, staked-ETH pairs) and, since 2023, an issuer of the **crvUSD** stablecoin. Curve matters to DAO governance out of all proportion to its size because of _how_ it is governed: it introduced **vote-escrowed tokenomics (veCRV)**, the design that most later DeFi protocols copied, and it became the arena for the **"Curve Wars"** — the clearest real-world demonstration of how token-based voting power becomes a market to be bought. ([Curve DAO docs](https://docs.curve.finance/user/curve-tokens/vecrv))

## veCRV — the vote-escrow model

Curve does not let you vote with liquid **CRV**. To gain governance power you must **lock** CRV in the [VotingEscrow](https://docs.curve.finance/developer/curve-dao/voting-escrow/) contract for a fixed term of up to **four years**, receiving non-transferable **veCRV** in return. Voting power scales with lock length and **decays linearly** as the unlock date approaches, so influence must be continually renewed. Locking buys three things at once: governance votes, a share of protocol trading fees, and a **boost of up to 2.5×** on the CRV rewards your own liquidity positions earn.

The point of the lock is to _couple control to commitment_: a veCRV holder cannot vote and immediately sell, so voting weight is tied to a multi-year stake rather than to whoever happens to hold tokens at a snapshot. This is the ve model's answer to the [plutocracy and vote-renting problems](/wiki/dao-governance/concepts/voting/token-weighted-voting) of plain token voting, and it was influential enough that "ve-tokenomics" became a [category of its own](/wiki/dao-governance/concepts/voting/vote-escrow) across DeFi (see [DAO tokenomics](/wiki/dao-governance/concepts/treasury/dao-tokenomics)) — later protocols copied it directly — [Yearn's veYFI](/wiki/daos/lending/yearn-finance) and [Frax's veFXS](/wiki/daos/stablecoins/frax-finance) among them — and its most prominent fork, [Balancer's veBAL](/wiki/daos/dexs/balancer-dao), ran the same 80/20-lock playbook before [deprecating it entirely in 2026](/wiki/daos/dexs/balancer-dao). ([curve-dao-contracts](https://github.com/curvefi/curve-dao-contracts))

## Gauge weights and the Curve Wars

veCRV holders do more than pass proposals — each week they vote on **gauge weights**, which decide how Curve's ongoing **CRV emissions** are split across liquidity pools. Directing emissions to a pool deepens its liquidity, so any protocol that wants cheap, deep liquidity for its own token has a direct incentive to control veCRV. That incentive produced the **"Curve Wars."**

Rather than each protocol locking CRV itself, [Convex Finance](https://www.convexfinance.com/) emerged as an aggregation layer: deposit CRV with Convex, it locks the maximum term on your behalf and hands back a liquid receipt, and [Convex](/wiki/daos/lending/convex-finance) votes the accumulated veCRV as a bloc. Within months Convex controlled **more than half of all veCRV**, becoming the effective power broker of Curve governance. **It still is, and the share is drifting up.** At Ethereum block 25,943,617 (10 September 2026) the Convex voter proxy held 418,549,575 veCRV of a 776,158,764 total — **53.93%**, up from 53.68% a month earlier — while holding only 49.88% of the 849,681,279 CRV actually locked in the escrow. The gap is decay: every other lock runs down toward its unlock date, and Convex's is re-locked to the four-year maximum. That maximum is a _rolling_ date rather than a fixed one, which is why it should not be quoted as a standing fact — this page previously gave it as 25 July 2030, and the same on-chain read now returns **22 August 2030**. All four figures are `eth_call` reads of the [VotingEscrow](https://etherscan.io/address/0x5f3b5DfEb7B28CDbD7FAba78963EE202a494e2A2) and [CRV](https://etherscan.io/address/0xD533a949740bb3306d119CC777fa900bA034cd52) contracts and can be reproduced at any block. The full on-chain read, including the CVX mint reaching its final cliff, is on the [Convex Finance](/wiki/daos/lending/convex-finance) page. On top of that grew an open **bribe market** (Votium and others) where projects pay veCRV/vlCVX voters directly to steer emissions — turning voting power into a rentable yield. Major protocols ran their own strategies here too — [Frax](/wiki/daos/stablecoins/frax-finance) amassed one of the largest vlCVX positions to steer Curve emissions toward its own pools. The template also travelled: the Solidly ve(3,3) fork of this gauge-and-bribe design became the dominant DEX on the OP-stack chains, most prominently [Aerodrome](/wiki/daos/dexs/aerodrome-finance) on Base. The Curve Wars are the canonical case study for the lesson on the [token-weighted voting](/wiki/dao-governance/concepts/voting/token-weighted-voting) page: even a lock-based system can have its votes bought, the leak just moves to a new place. ([Curve governance forum](https://gov.curve.finance/))

The amplification coefficient those votes sit alongside is not only an economic parameter. Curve's `ramp_A` moves it over a rate-limited window because the invariant a trade is priced against is solved by iteration rather than by formula – see [Numerical pricing in AMMs and bonding curves](/wiki/economics/numerical-pricing).

## crvUSD, and the 2023–2024 stress tests

In 2023 Curve launched **crvUSD**, an over-collateralized stablecoin with a novel "soft liquidation" mechanism (LLAMMA) that partially converts collateral instead of liquidating it all at once, plus the **LlamaLend** lending markets built on the same engine. The same period exposed Curve's risks in public:

- **July 2023 — the Vyper exploit.** A reentrancy-lock bug in specific versions of the [Vyper](https://vyper.readthedocs.io/) compiler (0.2.15, 0.2.16, 0.3.0) — not in Curve's own logic — let attackers drain several Curve pools for a cumulative **$60M+**. A large share was ultimately returned by white-hats and the attackers, but it was a systemic scare for the whole Vyper-based ecosystem. ([Halborn post-mortem](https://www.halborn.com/blog/post/explained-the-vyper-bug-hack-july-2023))
- **June 2024 — the founder-liquidation cascade.** Founder **Michael Egorov** had borrowed roughly **$95M** in stablecoins against about **$141M** of CRV spread across five lending venues. A ~24% CRV drop on 13 June triggered cascading liquidations that wiped out most of that collateral and briefly left ~$10M of bad debt in Curve's own LlamaLend market. Egorov repaid the bad debt within days, so no permanent loss was passed to depositors — but the episode is a standing lesson about founder over-leverage against a governance token. ([CoinDesk](https://www.coindesk.com/markets/2024/06/13/crv-slides-30-as-loans-tied-to-curves-founder-face-liquidation-risk))

## How Caper approaches this

Curve's veCRV set out to solve a real problem — transient token holders voting with no lasting stake — by forcing a **lock**. It worked, and then the lock itself became a tradable asset: Convex tokenised it and a bribe market rented it out, so control drifted back toward whoever could pay. A [caper](/wiki/foundations/what-is-a-caper) pursues the same goal — tying influence to genuine commitment — without a lock to tokenise. Its [voting weight](/wiki/governance/voting) combines the _stake_ a member holds with the _participation_ they have actually shown, so weight is _earned_ rather than rented, and a one-off borrowed position buys little say. There is also no gauge system and **no emissions to direct**: a caper's tokens come from a [bonding curve](/wiki/markets/bonding-curve), not a weekly inflation vote, so there is simply no emission stream for a "war" to capture. And because that same earned weight is also each member's [claim on the treasury at exit](/wiki/foundations/leaving-a-caper) — that same weight, not the [pro-rata slice by balance](/wiki/dao-governance/concepts/membership/rage-quit-and-exit-rights) a rage-quit pays, control stays coupled to real economic exposure — the property the lock was reaching for — with nothing to unbundle and sell. This is a design contrast, not a claim of superiority; the mechanics are on the linked pages and verified against the contract.

## References

- [Curve DAO: Voting Escrow (veCRV)](https://docs.curve.finance/user/curve-tokens/vecrv) — the canonical vote-escrow reference (primary).
- [curvefi/curve-dao-contracts](https://github.com/curvefi/curve-dao-contracts) — the DAO and gauge/escrow contracts (primary).
- [Curve governance forum](https://gov.curve.finance/) — proposals and gauge-weight discussion (primary).
- Halborn, [Explained: The Vyper Bug Hack (July 2023)](https://www.halborn.com/blog/post/explained-the-vyper-bug-hack-july-2023) — Curve reentrancy post-mortem.
- CoinDesk, [CRV slides as founder loans face liquidation](https://www.coindesk.com/markets/2024/06/13/crv-slides-30-as-loans-tied-to-curves-founder-face-liquidation-risk) (2024).