The main points:
The main points:
veToken economics encourage liquidity providers to become long-term stakeholders of the DEX protocol. This is achieved by giving more incentives and governance rights to those who lock the base token.
Curve and Convex create a powerful flywheel that will see token prices and TVL soar in the second half of 2021. This hyped the veToken model and brought supporters and detractors. This article wishes to examine these criticisms in detail.
We developed an alternative approach to the veToken model, leveraging our extensive research on the Curve/Convex relationship. This alternative is to account for a hypothetical token discussed in this article through a program called GOV.
first level title
Curve and veToken economics
The rise of cryptocurrencies and tokenization has opened up the design space of protocol economics, allowing projects to incentivize specific behaviors. However, many DeFi protocols reward all token holders equally regardless of the value they add to the project. More often, this requires very little lock-up in exchange for governance rights and protocol benefits. We believe that the best design dilutes employed capital and speculators, while rewarding long-term oriented participants.
One model we think encourages this behavior is through the implementation of vote escrow introduced by Curve. This model allows CRV holders to lock their tokens in voting escrow CRV (veCRV) for a maximum period of 4 years. The longer a user locks their CRV, the more veCRV they receive. As a way of taking liquidity risk and eliminating CRV supply swaps (showing their long-term commitment to the platform), veCRV holders are entitled to 3 main benefits:
Prorated for Curve fees incurred.
Increased CRV rewards for liquidity supply positions (up to 2.5x).
Governance and measurement of weighted voting power. Importantly, the measurement weights determine the distribution of the Curve for future versions.
The above-mentioned series of incentives have had a positive effect. The more long-term users (as measured by their veCRV), the more rewards they will receive. This flywheel is one of the most important drivers of the veToken model, which we discuss in detail in the next section.
Curve Our long-term value proposition is to be the stable asset and the most liquid place for transactions in the deepest way. The goal is not necessarily to become the main source of daily exchange DEX, but to empower stablecoin projects to build large liquidity reserves. This means that their main users on the supply side are liquidity providers.
Many failed income farming schemes were destroyed due to unsustainable token issuance and the proliferation of employment capital. Heres an example of what veToken economics are designed to avoid:
The initial liquidity mining program kicks off, then hires capital to deposit large amounts of money for the sole purpose of mining, and then sells the rewards.
At first, there will be a reflex rise. As the TVL and hype around the project increase, the Token price will also increase. This can further increase the rate of return on deposits, as the rate of return is paid in the native token, and the cycle repeats.
Eventually, an inflection point is reached and token incentives start to diminish, while hired capital continues to sell the tokens they mined. This sell-off resulted in lower yields for liquidity providers as token issuance declined.
Finally, a reflexive down cycle kicks in as liquidity exits the DEX making it essentially worthless. Lower liquidity means DEXs cannot support as much trading volume, which leads to lower costs. This has led to a further depreciation of fundamentals and a drop in the price of native tokens, further reducing LP yields.
The veToken model avoids this by aligning liquidity providers with the long-term interests of the protocol.
In order for Curve Liquidity providers to receive a 2.5x reward, they must hold a certain amount of veCRV corresponding to their liquidity. Not selling tokens motivates LP We:
Buy... CRV on the open market and lock.
Lock in all or part of the CRV liquidity incentives they earn.
It also encourages Curve liquidity providers on the platform to become long-term stakeholders and supporters of Curves success — rather than adopting a mine and sell strategy.
Furthermore, this not only motivates LP We to lock in 2.5x growth rewards, but also encourages them to accumulate as much veCRV as possible. This is because veCRV sets weights that allow holders to direct token releases to pools where they have funds (adding veCRV = more influence on token releases).
We have seen the accumulation of CRV not only between large LPs, but also at protocol level Convex. The agreement is to use Convex to hold a large amount of veCRV and accumulate CVX to stimulate their capital pool and ensure long-term liquidity.
Through the combination of CRVs slender distribution plan (more than 300 years, releasing... 15% per year) and veToken economics, it has been able to motivate a group of very sticky liquidity providers and token holders. This can be illustrated by the lock time between stkAAVE, xSUSHI and veCRV /Holdings.
first level title
Market reaction: The essence of liquidity is collateral and derivatives Convex
Given veCRVs enormous utility and its obvious disadvantage of being illiquid, the market has passed Convex Respond. Convex offers a non-callable, liquid veCRV collateralized derivative called cvxCRV. If holders want to maintain liquidity, there is no way for them to take ordinary CRV, because all governance rights and fees belong to veCRV. Convex takes advantage of this to allow users to lock their CRV in Convex in exchange for cvxCRV. cvxCRV is a token with strong liquidity and high returns. Its income comes from the basic veCRV cost, Convex cost and CVX Release platform the cost of.
Convex Base veCRV all provides power, governance and measurement weights for the platform and CVX holders as an exchange for providing high-yield liquidity tokens.
Convex Our basic product allows Curve LPs to stake their LP tokens on the platform, and depositors can get higher returns (2.5 times as high), in return, Convex collects CRVs mining cost percentage (10%), and then pays to cvxCRV holder. In addition, 5% of the CRV mined by Convex is used with cvxCRV in the form of CVX Pledgor. This forms the flywheel in the image below.
Convex also gives CVX holders access to underlying veCRV governance and weighting. CVX holders can vote on governance and receive incentives provided by third-party systems that guide them to vote to specific weight releases (also known as tickets, voting rewards, bribes, etc.). If you want to know more about the relationship between Curve/Convex, please see our previous article.
image description
An Alternative to veToken Economics
The remainder of this report discusses the implications for veTokens: a common economic criticism, and how they can be addressed by an alternative implementation. For simplicity, well use a hypothetical token called GOV for illustration.
This hypothetical GOV token would have the following main design features, each of which would draw from Curves experience in incentivizing alignment, while hopefully avoiding a single, dominant, Convex Proxy-like voting system.
It will use three token structures combined with Curve and Sushiswap-like governance features:
GOV represents the basic governance token and is not staked, similar to CRV. It provides no voting rights and no smart contract system fees.
xGOV obtains transferable governance tokens by staking GOV, similar to xSUSHI.xGOV provides mobile GOV risk and is entitled to a certain percentage of smart contract system fees and governance power.
vxGOV refers to the non-transferable governance token obtained by locking GOV, similar to veCRV. vxGOV has the right to obtain a larger proportion of smart contract system costs and governance rights.
It will avoid the need for a proxy voting system (similar to Convex entities) managed by a whitelist.
It will allow long term staking positions (vxGOV) to terminate early with penalties. (ie, it will allow exiting from the DAO in the protocol).
It will provide a voting incentive (also known as a bribe) mechanism within the protocol.
secondary title
1. Reduced investor base due to long-term lock-in
One of my arguments against veTokens is that long-term lockups make it difficult for entities such as index funds and trust products to incorporate these tokens into their offerings, which reduces the potential of the protocols governance foundation. While arguably this may reduce the participation of some entities, we believe this is a case of quality over quantity. The main use of trust and index products is speculation. In terms of governance, these entities will become indifferent stakeholders with no added value.
In our opinion, it is better for a community to have a small number of highly active stakeholders than a large number of passive and disinterested stakeholders. The whole point of veToken economics is to unleash the power of governance from short-term thinkers to long-term oriented players.
Another argument is that “governance will be much less secure during ownership transitions when large holders stop relocking tokens for sale”. However, this argument is based on the assumption that no other governance actor will fill the void. Furthermore, the framing of governance takeover does not appear to be accurate. Those doing the takeover, because of the locked veCRV holdings, are the largest stakeholders in the game, so they are naturally incentivized to make positive changes to the protocol.
Our position is that by directing governance to the parties that invest the most in the platform, the incentives far outweigh the disadvantages of excluding speculative entities from governance. Also, shifting governance away from large stakeholders who want to stop migrating tokens to new stakeholders who fill in the gaps is a healthy move.
In addition, we believe that the GOV model guarantees equivalence by providing liquidity within the protocol, and xGOV greatly alleviates this problem.
vxGOV (the veCRV version of the model) is designed to be most attractive to liquidity providers and long-term oriented investors, which makes sense since vxGOV holders are the most important stakeholders in the ecosystem.
On the other hand, this does not mean that other players cannot bring value and should be excluded from participation. And thats exactly what xGOV does.
xGOV is a pledged GOV position with liquidity. It can share the costs incurred by the smart contract system while gaining governance capabilities. This allows potential value-added stakeholders to continue to participate and contribute to the associated ecosystem, with the long-term and most dedicated players holding vxGOV retaining the majority of power and fees. The exact cost split between the two, and the resulting APR, will depend on many factors. But vxGOV will enjoy more benefits than xGOV.
secondary title
2. Mortgage derivatives
Another common argument against veTokens is that the resulting liquidity is collateralized derivatives. These staked derivatives allow token holders to forego governance in exchange for short- and medium-term soft-correlated fee and liquidity benefits. This again has two potential negative effects:
The fragility of the soft hook.
It is possible to generate a dominant proxy voter holding a large percentage of governance power, especially if there is a whitelist.
We have studied the relationship between Curve/Convex before. Around the sustainability of Convex, our main concern is the vulnerability of liquid mortgage derivatives linked to basic governance tokens.
The link between cvxCRV/CRV aims at maintaining the liquidity of the Curve stablecoin exchange pool through mining incentives. This pool facilitates low-slippage trading of both assets within tight ranges. If you want to understand the difference between these pools and traditional X*Y=K AMMs, we recommend reading Curve’s stablecoin pool paper. To illustrate this, lets dig into some data.
At the time of writing, Curves pool consists of 17.43 million CRVs (29.76% of the pool) and 41.13 million cvxCRVs (70.24% of the pool). Despite this imbalance, cvxCRV leaves only a 2% discount on CRV on trades of 1 million CRV.
It is worth noting that of the 1.88 billion cvxCRV in circulation, only about 4.1 billion are in this stablecoin exchange pool, and most of them (~76.87%) are pledged for protocol rewards.
If there is no way for everyone to redeem the base CRV with cvxCRV, we believe that this soft link will continue to deteriorate over time. This will inhibit new CRV locking, because it is cheaper for users to buy cvxCRV with Curve Pool than 1:1 conversion on Convex platform.
While this is a structural issue, lets see what Convex can do to help strengthen the hook.
Currently, Convex returns 10% of all CRV mining rewards to cvxCRV pledgers. This percentage can be increased up to 15%. This will increase the revenue generated by these mining rewards on CVX.
Convex DAO Protocols can use part of veCRVs shares, and staying in the cvxCRV/CRV pool is a measure of sustainable voting weight to motivate LPs.
Fees that flow into the protocol treasury are currently set at 0%. This can be raised up to a hardcoded 2% limit and used to buy cvxCRV, Help hook.
secondary title
3. Centralized governance
Regarding the centralization of governance, we agree that it is not healthy for one protocol to control more than 50% of the governance power. However, we see it as a special case of Curve/Convex rather than the overall veToken economics. And the novelty of Convex when it was first launched and its impact on the Curve whitelist, Convex can occupy more than 50% of veCRV. Without a first-mover advantage and the ability to block competition through a whitelist, Convex will have stiff competition from the start. After some time, we urge those protocols that intend to adopt veToken economics to launch without a whitelist. This will facilitate competition among several similar Convex protocols and maximize decentralization.
In addition to centralizing voting power by owning veCRV, the current Convex architecture also relies heavily on multiple identities of administrators to convert vlCVX votes into Curve DAO proposals and weight them. This poses another important issue for centralization, as a 3/5 multisig can effectively control more than 50% of veCRV voting power.
In conclusion, we would like to highlight some proposals that can reduce the risks associated with governance centralization as it relates to Curve/Convex.
Partial vote:
Before partial voting, if 51% of vlCVX votes for a certain proposal, all controlled veCRVs of the Convex will vote for it in Curve DAO China. In this case, after partial voting is implemented, only 51% of veCRVs controlled by the protocol will vote yes, and the other 49% will vote no.
Incentive adjustments:
Convexs success largely depends on Curve. One way to value CVX is to calculate the present value of all future bribes that CVX releases from its influence on CRV. Given Convexs dependency on Curve and the CRV token price, it seems unlikely that Convex stakeholders will act maliciously against Curve.
So far, the relationship between Curve and Convex has been mostly symbiotic. This can be extended to other protocols participating in the Battle of Curve, such as Frax, Terra, Redacted, Olympus, etc. are actively accumulating more CVX protocols. The veToken model and its impact on future CRV The ability to issue has combined the top full space level Agreement to jointly accumulate CRV/CVX and further adjust the incentive mechanism.
Proxy voting with minimal trust:
secondary title
4. Bribes and short-term incentives
Bribes have been making waves in the DeFi space as a way to manage liquidity — more precisely, to steer future token offerings. See a snippet from the MakerDAO forums. As long as the bribe income (Bribespaid / CRV Reward) is higher than the percentage of CRV locked, veCRV holders can still gain benefits by diluting the unlocked holders.
One of our counter-arguments to this argument is that when it comes to accepting bribes in exchange for weight votes, token dilution happens whether you accept the bribe or not. In our opinion, the word bribe is a bit of a poor nomenclature here.
secondary title
5. veToken Economics increases the cost of defending against governance attacks
To mount an attack through the governance veToken system, an attacker must purchase and lock their voting rights. This greatly increases the cost of the attack, as purchased tokens cannot be recovered and sold. This is the basis behind the veToken economic incentives.
However, veToken economics critics see this as a double-edged sword, as it also increases the cost of defending against attacks. For example, if an attacker decides to buy and lock enough vxGOV to gain 51% of the voting power, the defender will have to re-lock vxGOV or buy more GOV to increase its voting power to counter the malicious proposal.
We found this argument to be rather weak because it works both ways. Given the attack, economic protocols adopting veTokens need to buy and lock to maximize impact, which we see as a net positive.
Provide important governance authority for the lock-up, which can prevent attackers from borrowing the protocols governance tokens and malicious voting. In this case, the attacker can only be affected by the borrowing interest, but not the long-term adverse effects of wrong decisions. Heres an old example of an attacker trying to influence... the governance of MakerDAO with flash loans. While this is an extreme case facilitated by flash loans, this still allows attackers to borrow tokens and vote for malicious proposals without significant price risk.
first level title
Learning to make GOV models from Curve/Convex
We formulate our hypothetical GOV model using experience from extensive studies of the Curve/Convex relationship.
secondary title
1. No whitelist
Convex on Curve I have > 50% Lack of governance One of the main reasons is lack of competition. While Convex our design is the first of its kind, it has the opportunity to rack up a lot...before any competitor discovers veCRV. At this point, it is too late for other competitors, because Convex can use veCRV to control the Curve whitelist, basically preventing further competition.
Now the market has realized the pattern of Convex. Therefore, new projects should announce the veToken model transition with a considerable lead time. This has many market players preparing for its launch. Since it was the result of competition from the start, we believe this will encourage two things:
Makes the governance distribution ratio curve fairer, as many projects have the opportunity to accumulate base GOV Tokens. This makes it very unlikely that a single entity will accumulate > 50% governance.
secondary title
2. Rage Quit
We believe that the attractiveness behind cvxCRV is not only its high APR, but more importantly its liquidity. So we came up with a few different ways to preserve the benefits of the veToken Design while still providing depositors with potential liquidity when needed.
The first option is to make vxGOV your position into a transferable NFT (such as Solidly). Although this makes it less attractive for you to give up GOV Token and go to a similar Convex agreement, the pricing of these NFT positions will be difficult. We can also assume that sellers will be willing to discount these positions on the secondary market, which may make locking GOV tokens less attractive.
If vxGOV positions could be purchased at any time, it would be a fair question for potential GOV buyers to ask why lock up GOV if I can just buy the NFT at a discounted price? With this in mind, we thought another better option would be to introduce the ability to rage quit vxGOV locations, with heavy penalties for quitters. Various strategies can be implemented with this project, here are some hypotheses to give us a potential idea of vxGOVs implementation of angry exits:
If you lock up as much vxGOV as possible and want to exit, you will get 50% of GOV tokens and the other 50% will be distributed back to the community. The penalty for exiting in rage will decrease as your lock time decreases. The benefits of angry quitters can be distributed in a few different ways.
Diamond Hands: Collected GOV is maximally re-locked and distributed proportionally based on each account controlling vxGOV. This has the benefit of incentivizing diamond hands and long-term commitment to the protocol, as these rewards will benefit the stakeholders who are most invested in the protocol.
Treasury: This program requires some active financial management. For example, the funds could be used to buy similar stakes in other Convex or Redacted Cartel A, or to build a pool of less related assets.
The protocol is liquid: as an extension of #2, in Treasury GOV can be regularly deployed to the liquidity pool of GOV tokens. This will help build a strong liquidity base for GOV, and is protocol-ready liquidity, which will increase GOVs sustainability in the long run.
secondary title
3. Built-in bribery
We believe that built-in bribes may increase the attractiveness of staking vxGOV alone, rather than giving up Convex etc., especially for small investors.
Taking Curve as an example, before the emergence of platforms like bribe.crv.finance, it was difficult for small veCRV holders to directly measure the release ability of Monetize.
If there is no market, bribery must be done privately through the OTC. This is bad for three reasons:
About the DAO Come on, it makes more sense to contact a single veCRV whale than to contact 1000s of smaller holders. This is good for whales, not small users - thus, not great.
Since these contacts typically do not occur on-chain, there is some counterparty risk involved. If the DAO bribes early, the veCRV whales may not keep their word and vote for the agreed standard. On the other hand, if veCRV Whales vote before receiving payment, the DAO may not be able to deliver on its promise to pay.
Off-chain bribery happens behind the scenes and prices are opaque, resulting in an inefficient pricing market. Having markets on-chain promotes healthier markets for all participants.
Summarize
Summarize
Our results show that Curve, created by veToken Economics (thanks to Michael Egorov and team), remains one of the best incentive alignment mechanisms to date. We hope that our analysis and suggestions can inspire the implementation of veToken model projects. Our core goal here is to reduce centralization while still maximizing the core attributes of veToken: rewarding the longest and most dedicated participants, at the same time, it does not exclude that we cannot afford the illiquidity brought about by the contributor economics of traditional veToken sexual risk.
