For the complete documentation index, see llms.txt. This page is also available as Markdown.

TeaVault v3 Portfolio-GDPS Model

Gradient Decay Profit-Sharing Model

Teahouse's TeaVault v3 Portfolio implements its Gradient Decay Profit-Sharing (GDPS) Model, a vault-level high watermark-like performance fee mechanism. The purpose of this model is to fairly govern the collection of profits.

If you are interested in reading more about why Teahouse designed this model, please see the “Background” tab on this page.

How the Gradient Decay Profit-Sharing model works

Teahouse’s proprietary GDPS model is designed to redefine the dynamics of profit-sharing for strategy managers/providers.

Design

When a strategy successfully generates profits, the smart contract locks the performance fees destined for the manager. The innovative aspect of the GDPS model lies in the gradual release of these performance fees over time, introducing a temporal dimension to the claiming process.

The allocated performance fees (i.e., the manager’s profits) undergo a gradual "unlocking" process, wherein the amount of locked fees the manager can claim gradually decays over time. For instance, the smart contract can be configured so that the manager can claim only 20% of the fees entitled to them after six months. By the twelfth month, they can claim 20% of the remaining profit from the sixth month.

We can use the figure below to illustrate. Let us assume the manager’s initial profits are 100 USDC. In Month 0 (M0), all 100 USDC of the manager’s profits are locked, but by Month 6 (M6), they can claim 20% of their profits, which would be 20 USDC. After the manager has claimed the 20 USDC in M6, 80 USDC in profits will be left (still locked).

By Month 12 (M12), 20% of the remaining profits (80 USDC) will be claimable, which would be 16 USDC, and so on and so forth. This shows that the amount of profits the manager can claim will gradually diminish over time unless they can increase the strategy’s performance to generate more profit.

Upon Loss

In the event of losses, all upfront profits earned by the strategy manager are reinvested back into the strategy pool. This ensures that any losses are absorbed by the manager's initial profits rather than impacting users directly.

Looking at the figure below, we can see that when the profit drops to zero, all previous profits entitled to the strategy manager are returned to zero until the strategy starts to make a profit again.

The GDPS model is a strong deterrent, preventing strategy managers/providers from manipulating short-term gains or engaging in actions harmful to investors. By ensuring accountability and stability, the model safeguards the long-term success of the strategy and its protocol.

Background

The High-Water Mark Mechanism

In traditional finance products, such as hedge funds, the high-water mark mechanism is instituted to ensure a hedge fund manager receives fees only when the fund value surpasses a certain level. This ensures that performance fees are earned solely on new profits generated by the manager and not on profits that merely offset losses from previous periods.

High-water mark example:

Suppose a strategy manager sets a 20% performance fee for profits. (Here, we will overlook other fees charged for simplicity.)

If we set the initial AUM of the vault at 20k USDC, and the manager earned 80k USDC in a three-month period to reach a total vault value of 100k AUM. The manager is then entitled to 100k x 20% = 20k in performance fees.

The next time the manager will be entitled to performance fees will be when the AUM surpasses 100k USDC.

Assuming, in the next five months, the AUM reaches 160k USDC. Considering the previous high-water mark of 100k, the manager would be entitled to (160k - 100k) x 20% = 60k x 20% = 12k in performance fees.

Formula: (Current AUM - Previous high-water mark) x Performance fee %

However, the high-water mark mechanism requires investors to enter/exit a fund at the same time to calculate the high-water mark. In the DeFi world, a permissionless vault's performance can fluctuate as users have the flexibility to enter or exit a vault at any time. This dynamic movement can lead to varying relative levels for each user. Additionally, the trading or repledging of share tokens adds complexity to their calculation, making it extremely difficult to determine their exact value. All of these factors make the high-water mark mechanism infeasible for web3 permissionless vaults.

In light of the challenges in implementing a high-water mark mechanism and the risks associated with strategy manipulation, Teahouse's Team created the GDPS Model to offer users a safer and more transparent DeFi experience. This model promotes accountability of the manager and ensures a harmonious growth journey where both investors and strategy providers mutually benefit.

Last updated

Was this helpful?