The Road to Curve Wars

Curve has become one of the most important protocols in decentralized finance (DeFi) today. Liquidity is king in DeFi, and Curve has become a kingmaker.

Consequently, an industry in DeFi has emerged, rife with bribery, to help protocols increase their control over the liquidity Curve provides. The situation has been labeled the Curve Wars, and it's not hyperbole to state that gaining an edge in this war means the difference between a project's success or its demise.

Now, how exactly did we get here? It's impossible to say where this war will end up, but we can look back and see how it originated. In this article, we'll focus on the origins of the Curve Wars and cover some notable events that made it possible.

There Would Be No Curve War Without CRV

The whitepaper for Curve was published in late 2019, and in January of 2020, the protocol was launched without its own token. Introducing a governance token that would make Curve a decentralized autonomous organization (DAO) was discussed in the Curve DAO whitepaper, which appeared at the end of June.

The DAO whitepaper introduced a system where users could lock CRV for different lengths of time to receive veCRV (vote escrowed CRV). Users would have an incentive to lock in their CRV for veCRV, because locking could boost their future CRV rewards up to 2.5x.

As a governance token, veCRV could also be used to vote in the DAO's decision-making process. One especially important decision that could be made with veCRV was how CRV rewards were distributed among liquidity providers (LPs).

So, on top of earning fees for providing liquidity to Curve's automated market maker (AMM) pools, users could earn CRV rewards that they could either sell or lock for additional benefits, including control of the protocol.

Today, protocols are fighting for a share of around 640,000 CRV distributed every day. CRV's emissions schedule has been reduced once, in 2021, and it is set to reduce by 15% every August for the next 300 years.

CRV Got an Impromptu Launch and Some Got a Headstart

Liquidity mining, or farming, had become all the rage in the early summer of 2020. CoinMarket Cap was questioning if liquidity mining was a useful tool or just a passing fad, but it pointed out that rewarding users with governance tokens incentivized users' participation in a protocol dramatically.

Curve teased it would launch a farmable governance token at the end of May 2020, and it mentioned that the protocol would retroactively reward LPs once the token went live. After this announcement, Curve's daily volume increased 50x over the next two months.

Curve became a "black hole" of liquidity as token hunters deposited liquidity en masse. The launch of CRV tokens was forecast to occur sometime by mid to late August, but the Curve team was vague about dates.  

Then, one user's participation in the protocol went above and beyond the norm.

On August 13th, 2020, the user known as 0xc4ad (Chad) pillaged Curve's GitHub, pulled the code for CRV contracts, and paid the ETH necessary for surprise-launching Curve's DAO governance—from five different addresses starting 0xc4ad (funded by the same address, which is active today).

The impromptu launch of CRV by an outsider was hailed by some as an important moment in the decentralization of Curve. On the other hand, a lot of people cried foul, saying 0xc4ad was an inside job.

Despite the surprise launch, the smart contracts were ready and had been launched correctly. Curve shipped the UI, and users began vote-locking CRV right away on August 14th.

Curve's docs stated (and still do at the time of writing) that boosts wouldn't begin until the late hours of August 26th (ended up being the 28th), but users were locking CRV ASAP.

The First Gauge Weight Vote Demonstrated CRV's Power

The first gauge weight vote to put veCRV to work was held on August 19th, and the Compound pool received votes that skyrocketed its APY much higher than any other pool.

Stake Capital, the parent of StakeDAO, which played/is playing a significant role in the Curve Wars, published an article explaining the effects of CRV's tokenomics and the role veCRV gauge weight voting played on attracting liquidity.

Graphics taken from Stake Capital's article, "Curve Tokenomics and First-Mover Advantage," highlight the correlation between voting, APY, and attracting liquidity:

After voting concluded on Wednesday, the 19th, the APY shot up as soon as Thursday, the 20th, arrived at midnight. As more people deposited in the Compound pool to capture the extremely high APY, the returns began settling down.

The above chart shows just how much liquidity was attracted to the Compound pool in a short amount of time. Even after the APY settled back down at 4AM, users were still making large deposits.

This led Stake Capital to note that, "CRV is not just money, it’s power."

Here's another explanation of what happened to the Compound pool taken from the Curve Discord. The names of public figures have not been redacted.

The Community Predicted the Curve Wars Right Away

As Curve's Discord community began doing the maths, it started to dawn on several members that a war was brewing. As early as August 18th, 2020, it was predicted, "This will turn into a very interesting boost-war once people realise that others' boost is your own uh.. un-boost."

What we now know as the Curve Wars was being called, throughout the rest of that August two years ago, "whale wars," "governance wars," "vote wars," "gaugeweight wars," and "pool-turf wars" on Curve's Discord.  

Then, the Curve Wars' First Shots Were Fired

On August 23rd, Andre Cronje, one of the masterminds behind Yearn, stated publicly that Curve had been taken over by its founders, and "voting is pointless now."

Curve's founder, Michael Egorov, explained what happened on Discord.

The Compound pool's APY broke through the ceiling just days before, riding on the back of veCRV votes. Yearn wanted a piece of that action for the yPool, but it didn't have a chance to compete. Cronje was not a happy camper, and he may have had a point.

A few days later, the veCRV distribution had become less dominated by founders, but they still held the majority of locked votes.

Cronje found a solution that would bring DeFi a step closer to what we now call the Curve Wars. He proposed, on August 25th, creating a whitelist that would allow a project's smart contracts to interact with Curve.

This proposal put the key mechanism in place that would allow a protocol to consolidate voting power over Curve's governance.

Currently, there are only four smart contracts that have been whitelisted: Yearn, StakeDAO, Convex, and, most recently, Abracadabra (May 2022). Several other protocols have attempted whitelisting and failed.

Yearn's Backscratcher Vault Became Curve's First Governance Proxy

Notice how the infographic below is labeled "Yearn Strategy" under the category of "Miscellaneous." If this infographic were made now, there would be a lot more projects listed under this category, which could theoretically be renamed "Bribes."

It's possible that if Yearn had chosen, at one point, to lock CRV instead of flipping it to increase Yearn's vault yields, the protocol could have dominated the race to control Curve's governance during an early window of opportunity.

The Yearn community debated over what to do with all their CRV rewards in YIP 37: Participate in CRV governance and 2.5x CRV reward boost.

The vote passed, and early CRV rewards (spewing compared to current emissions) would be vote-locked for boosts instead of being returned directly to Yearn's users. Cue another step towards making the Curve Wars possible.

By fall, DeFi hit a speedbump, and Yearn's TVL decreased by more than half from a high of around $1 billion. In October, Yearn began discussing a CRV vault strategy to benefit all vaults using Curve, and this became the Backscratcher Vault.

This article claims that Yearn "aim[ed] to replenish lost liquidity" with the launch of the Backscratcher Vault, a new product that accepted users' CRV, forever, in return for "perpetual claim on Curve DAO admin fees across all Yearn products."

The Backscratcher Vault (yveCRV vault) was announced in the Yearn Finance Newsletter #10 dated November 9, 2020. On the same day, a Yearn governance proposal began a discussion titled Establish who votes in Curve DAO on behalf of Yearn.

The Curve team, which was on the sidelines watching another protocol discuss its own governance, joined the conversation.

At this point, the situation seems more like a cooperative effort than a war, but the "concert of Curve" would fall apart within a few months.  

Developments in January Pushed Curve Further Towards Battlefield Conditions

In January 2021, Yearn collaborated with Curve to launch factory pools. The factory pools made it possible for projects to permissionlessly add pools to Curve.

Still, two DAO votes would be necessary to reward the new pools (one vote to add a rewards gauge and one vote to set reward parameters). This is really important, since a pool without rewards draws less liquidity than one earning CRV.

Now, any project could pair its tokens with Curve's 3Pool (USDC, USDT, and DAI) to benefit from the pool's deep liquidity.

Pairing with 3Pool also stabilizes a token's price; the more tokens a stablecoin pairs with 3Pool, the less its price moves during larger transactions. The 3Pool itself helped stabilize DAI's price after it launched, and this effect translates over to stablecoins 3Pool is paired with as well.

The Conditions Were Primed for Curve Wars

Factory pools also made it possible for protocols to allow users to "have their cake and eat it too" with regards to locking tokens. One drawback to locking tokens is that no one knows what DeFi will look like in a month, much less four years.

The major draw for the protocols that are now built on Curve is that when users lock their CRV, they receive another token, like cvxCRV, which is liquid since it can be swapped for CRV again on Curve. So, it's likely there wouldn't be a Convex if it weren't for the introduction of factory pools.

StakeDAO (founded by the same Julien Bouteloup who announced Curve's debut of CRV) also launched in January, snagging the second-ever smart contract whitelist. In April, a third protocol managed to have its smart contracts whitelisted, and this brings us to Convex.

Convex Arrived and Hoovered CRV Like No Tomorrow

A proposal was introduced to add Convex to the Curve whitelist on April 23rd, 2021. Yearn was between a rock and a hard place on this vote, because if they voted no, then they wouldn't receive the "bribe" of airdropped CVX for voting Yes to whitelist Convex.

From the Yearn governance forum:

Yearn eventually voted Yes, since an overwhelming Curve voter turnout (hunting for a bribe) meant Convex would get whitelisted no matter what it voted.

Also, it was rumored that the Curve team was behind Convex...

...but Curve later denied they were working together.

Convex began accumulating CRV at a rapid pace by bribing users with airdrops. Its veCRV holdings surpassed StakeDAO in just two days. It took two weeks for Convex to surpass Yearn in veCRV holdings.

As of now, Convex has made it mathematically impossible for any individual or protocol to regain majority control of Curve. Since Convex controls Curve, users who held CVX controlled Curve.

Bribery Ramped Up the Curve Wars Another Notch

Cronje introduced a bribing mechanism, most likely the last piece of the Curve War puzzle up to now, in August 2021. It further complicated the Curve Wars by allowing protocols to bribe users for their votes.

The bribe tool Cronje launched has fallen out of favor these days, but bribery still plays a major role in Curve voting. In fact, it looks like bribes became pretty necessary.

Protocols can spend much less bribing voters than if they tried to buy the votes themselves, and the rewards remain the same.

Nowadays, Convex controls the largest stake in Curve's governance, but Votium at the moment controls Convex, since it adds the mechanism for bribes, and users want bribes.  

The Elements of a Liquidity War

The Curve Wars have essentially fractured into a dizzying array of belligerents fighting for control of the liquidity on Curve. Projects continue to enter the fray by offering optimized schemes for locking tokens and accepting bribes.

At this juncture, it's hard to keep track of every party involved in the Curve Wars. Instead, let's look back on some things that happened in order to reach this point:

• Curve launched an AMM with utility for stablecoins.
• Curve experienced a surprise launch of CRV.
• CRV gave users control over Curve governance and CRV rewards distribution.
• A vote for a smart contract whitelist made it possible for projects to consolidate CRV holdings, but few managed to whitelist.
• The introduction of factory pools opened Curve to any project and invited projects to incentivize their pools for deeper liquidity.
• Bribes were found to be the cheapest way to incentivize users to vote for pools.
• Multiple protocols now fight to attract users and their votes or proxy votes.

The question now is: How can another protocol replicate this kind of successful participation in its own governance model?

On Solana, the stableswap AMM Saber is setting the stage for its own liquidity wars with voting for SBR gauge weights held every week. Users can lock SBR for up to four years in order to vote on which Saber pools receive more rewards in a fashion similar to Curve, and projects are definitely participating.

It will be interesting to see if history can repeat itself.

Keep in Touch

Website | Twitter | Telegram | Discord | Email | Reddit