What are the differences between the ve models of Curve & Plenty?

Plenty's new vote escrow (ve) system is a new decentralized finance (DeFi) protocol design first proposed by Andre Cronje, founder of Yearn Finance. The model iterates on the vote escrow system (“ve”) from protocols such as Curve.
To explain what the differences are let’s first look at the protocol design of Curve.

Curve's “ve” model

Vote escrow(ve) is a popular model in DeFi. It has proven to align the incentives of token holders and liquidity providers which has resulted in, among other things, market outperformance of projects that have incorporated “VeNomics”. Consider this chart by Bankless:
The VeNomics model locks user funds (usually the project’s token, for example CRV) and grants voting rights and trading fees, and other benefits (the veTokens, for example veCRV). The locking period is usually between a week and four years. The longer the lock, the higher the rewards and voting power. Curve‘s CRV tokens are governance tokens. Their function is to incentivize liquidity provision on Curve and to involve more users in the governance mechanism.
Liquidity providers on Curve receive CRV tokens rewards in pools that have been selected by veCRV governance. CRV has three main uses: voting, staking and boosting. These three operations require users to first lock (more precisely, “vote lock”) their CRV tokens to acquire veCRV. veCRV (i.e. vote escrowed CRV) are non-transferable tokens representing CRV locked for a period of time in Curve pools. The longer CRV is locked for, the more veCRV a user receives. veCRV tokens enable users to vote in governance, boost their CRV rewards (by multiplying their liquidity up to 2.5x times) and receive trading fees generated in Curve pools. veCRV holders are also eligible to receive airdrops.
Existing protocols in DeFi, have been competing to get voting power within Curve’s ecosystem because that enables them to select the liquidity pools towards which CRV token emissions are directed. This competition has been referred to as “Curve Wars“. By redirecting CRV token emissions towards a liquidity pool, liquidity providers (LPs) in that pool receive higher rewards. This attracts more LPs and the liquidity in those pools grows further.
Vote locking might be the biggest innovation introduced by Curve Finance. Because of this feature, vote weights and share of rewards are proportionally assigned in accordance with the locking period. Hence, time locking a token increases the long-term commitment from the holders, reduces circulating supply, and removes potential downward price pressure. Considering staking operations on Curve, it is noticeable that 50% of all trading fees are distributed to veCRV holders. This is to align incentives between liquidity providers and long-term token holders (i.e. veCRV holders). Every time a trade takes place on Curve Finance, 50% of the trading fee is collected by the users who have voted locked their CRV. Every week, fees are collected from the pools, converted, and distributed.

Plenty's "ve" model

Plenty's ve design iterates on Curve's ve model. With a focus on addressing the issues related to liquidity mining, the liquidity bootstrapping mechanism employed by the majority of DeFi projects that led to the 2020/2021 DeFi boom (and bust). According to Bankless:
As we know, much of DeFi’s growth over the past year and a half has been fueled by liquidity mining. While it’s often done at the product level, such as with a DEX, many protocols have also incentivized liquidity for their native token through token emissions. While it’s important for a token to have deep liquidity, these programs have often been taken to the extreme to attract yield farmers, resulting in inflation rates that would make Jay Powell blush, and leading to perpetual sell-pressure on the underlying token.
It doesn’t take a PhD in economics to see why DeFi tokens would underperform: They have a massively inflating supply with no demand to help offset this.
The goal of Plenty's ve model is to better align emission of tokens to beneficial actions and solve the problem with current AMM designs where liquidity provision is temporarily subsidized while fees generation, the more sustainable incentives-generating mechanism, is not.
Existing autonomous market makers (AMMs) are primarily designed for LPs and incentivize liquidity depositing into the protocols’ liquidity pools in order to receive a temporary emission of free tokens. Following Cronje Medium articles, “current AMMs need a few modifications to make it easy for protocols to leverage them":
  • Must be able to easily add token incentives to your liquidity.
  • Must be able to easily bribe token emissions onto your liquidity.
  • Must be able to accrue fees from liquidity you incentivize.
  • Must be able to permissionlessly deploy your liquidity.”
The introduction of a new kind of ve model will be accompanied by a new AMM design. This new design will feature:
  • Native support for swaps between closely correlated assets using Arthur's Flat Curve.
  • Native support for swaps between uncorrelated assets.
  • 0.10% fee for correlated swaps.
  • 0.30% fee for uncorrelated swaps.
  • Fees are paid out in base assets, not converted.
  • Permissionless creation of liquidity pools for volatile pairs.
  • Tezos community whitelist for flat curve liquidity pools.
  • Tezos community whitelist for addition of Gauges & Bribes to a liquidity pool.
  • Fees attract incentives instead of manual liquidity incentives.
  • Native support for adding third party incentives.
  • Ve locks accumulate all fees for pools they vote on.
  • Ve locks increase holdings proportional to emission, no dilution.
  • Ve locks are represented as an NFT to allow capital efficiency of locks.
  • Minimal DAO.
Plenty's ve model ensures the totality of fees will be paid to users locking in their assets in the protocol, ensuring a higher level of structural sustainability over time.
The reasoning behind ve is to encourage users to lock up their governance tokens and obtain vePLY NFTs with an aggressive inflation model to ensure that locker’s rights are not diluted.
To encourage locking and voting, the system compensates users with PLY for the inflation risk brought by locking up, and gives locks the share of additional tokens issued in emission according to the corresponding total circulation share to ensure that the corresponding equity share of vePLY will not be diluted.
In this way, users who lock their positions do not have to worry about inflation affecting the value of the tokens in their accounts, while those who do not lock their positions will bear the downside risk of token inflation.

Three economic principles of the ve model of Plenty that deviate from the standard ve rules from Curve

We have modified the math of the original first two principles of ve(3,3) to allow decent incentives for protocols that join later in the future.
1. Weekly emissions are adjusted as a percentage of circulating supply
The real emission that the users receive is related mathematically to a base emission as follows:
Emissionreal=Emissionbase(EmissionbasePLYlockedSupplyPLYtotalSupply0.5)Emission_{real} = Emission_{base} - (Emission_{base} * \frac {PLY_{ lockedSupply}} {PLY_{totalSupply}} * 0.5 )
Meaning, if the weekly base emission is set at 2,000,000. Then, if 0% of PLY is locked for vePLY, the entire 2,000,000 is emitted. If 50% of PLY is locked for vePLY, the weekly emission would be 1,500,000. If 100% of PLY is locked for vePLY, the weekly emission would be 1,000,000.
2. ve locks increase their holdings proportional to the weekly emission
The locked PLY supply is inflated to prevent dilution as:
PLYnewLockedSupply=EmissionrealPLYlockedSupplyPLYtotalSupply0.5PLY_{newLockedSupply} = Emission_{real} * \frac {PLY_{ lockedSupply}} {PLY_{totalSupply}} * 0.5
Assume a 1,500,000 PLYweekly emission, a total_supply of 20,000,000 PLY, and a locked_supply of 10,000,000 PLY. This would mean that 1,500,000 are minted and provided as incentives. Then, according to the math, total locked supply would be inflated by 3,75,000 PLY.
3. vePLY is transferable as an NFT
By tokenizing the lock position we allow a single address to own more than one lock. Lock balances are cumulative and each lock contributes to the overall voting power. This further allows locks to be traded on secondary markets, as well as to allow participants to borrow against their locks in future lending market places. By extending locks into Non Fungible Tokens, the capital inefficiency problem of ve assets is solved, as well as addresses concerns over future liquidity (should it ever be required).

What are the differences between Curve's ve model & Plenty's new ve model?

The differences between Curve's ve model & Plenty's ve model