BIP 68: ibBTC Yield Retention Model

Category: ibBTC

Scope: Adjust harvest structure to support ibBTC yield capture on partner strategies

Status: Pending

Objective: Restructure partner strategy (ex. convex) harvests such that the max selling capacity is used to autocompound into ibBTC. In return a portion of the ibBTC earned yield is redistributed to all vault depositors. Both ibBTC and vault depositors get higher yield.

Specifics:

Tl;dr

  • All selling capacity of partner token harvests across all vault depositors is used to support yield in ibBTC
  • In return ibBTC will distribute a portion of its token based yield across all eligible vault depositors. Gross base yield for all vault depositors will go up
  • Performance fees will be collected 100% in vault tokens
  • The portion of yield autocompounded for vault depositors will be variable
  • Vault Depositors have opportunity to earn more additional yield than what is collected for performance fees, potentially becoming fee-negative.

pre-BIP community discussion: Badger

Problem

The highest yielding strategies at badger earn the majority of their yield through native token distributions (i.e. CRV and CVX emissions on convex setts). With the Partner-first approach badger is committed to programmatically selling smaller percentages of these assets (generally 20%). When these strategies are held by ibBTC the token emissions are accumulated by the ibBTC contract with no easy way to pass that yield to ibBTC holders since ibBTC is intended to be used across chains and protocols.

Solution

Use the available total selling capacity (20% currently) to support autocompounding a portion (for example 60%) of the token yield generated by ibBTC. This means that the token yield for ibBTC relative to direct deposit into the badger vaults will be ~40% less but will be 100% denominated in BTC. This allows ibBTC to have extensive utility and liquidity across numerous chains and protocols without dev work being needed to support distributing rewards. With this change and the recent rebalancing, ibBTC will become a more managed product where parameter tweaks and rebalancing can be done periodically to target an attractive yield.

This is also a major improvement for vault depositors. By depositing with badger, if ibBTC grows to be a significant holder of all vault positions, depositors can receive additional yield above and beyond what is paid in performance fees. So beyond the current benefits of autocompounding, and auto-staking reward assets, depositors can now earn more gross yield than they would depositing anywhere else. By accepting volatility in the amount autocompounded, depositors are rewarded with higher yield subsidized by ibBTC.

Parameters

ibBTC Yield Retention % - The percentage of token yield will be used to compound back into ibBTC

ibBTC Yield Model Eligible - Vaults that will operate within this structure

Implementation

Assuming all strategies are harvested at the same time, when a harvest is done….

  1. Send 20% of total token yield to DAO (performance fee)
  2. Sell 20% of all tokens for WBTC
  3. Deposit 60% of the token value harvested for ibBTC held vault positions, or the total amount of WBTC sold in #2 (whichever is smaller) back into ibBTC using WBTC from #2
  4. Distribute all remaining funds (tokens via tree & WBTC autocompounded) to vault depositors (as normal)

This is just descriptive and may be implemented in a different order but directionally is correct.

Implications

For ibBTC

Higher Yield - To have optimal utility with ibBTC we must autocompound rewards into it in BTC terms, distributing tokens across all supported platforms and chains for ibbtc isnt practically feasible. With this plan ibBTC is able to capture much of the yield from strategies where the majority of yield is token based while retaining optimal utility. Badger can continue focusing on using the partner first approach to secure the best yields in DeFi and not worry about ibBTC yield lagging.

Vault Depositors

Vault depositors are those that are directly depositing to the vaults, not holding ibBTC. Currently 10% of the total yield (12.5% of total depositor yield net fees) of all token rewards are sold and auto-compounded for partner token strategies. With this new structure the percentage autocompounded will be variable between 25% (if ibBTC aum were to go to 0) and 0% (if ibBTC grows to be of substantial size). The smaller ibBTC aum relative to eligible vault positions the less extra yield vault depositors get but the higher percentage yield autocompounded. The larger ibBTC relative to eligible vault positions the higher gross yield for vault depositors but the lower percentage autocompounded. Vault depositors still receive all benefits of badger boost and have full control over reward assets once they have been claimed.

DAO Treasury

The DAO will no longer collect BTC denominated performance fees from partner token denominated yield. By optimizing for accumulating strategic assets like CRV and CVX the DAO can continue to push to secure the best yield possible for depositors.

By the Numbers

These are all roughly approximated numbers to illustrate the behavior of the model directionally. As the ibBTC ratio (ibBTC AUM / Total eligible AUM) increases, gross yield increases for depositors and the DAO while the BTC component of vault depositors decreases. ibBTC retention is set to 60%. The first column of numbers shows approximately what the current state would be if it were to be implemented now.

ibBTC Rewards Redistribution [PUBLIC]

Rollout

As stated this change is a move towards ibBTC being a more actively managed product. To retain the ability to monitor the market response and react as needed it makes sense to leave some wiggle room in the parameters. For governance purposes a Yes vote on this BIP would approve the core team to:

  1. Include all convex vaults in this model (but not necessarily require them all to be included day 1) and add more vaults with council approval.
  2. Implement and adjust the ibBTC yield retention parameter in a range of 50%-75%

Vote:

Implement this model as designed (Team will choose eligible vaults and initial retention parameter and communicate at least 1 week before deployment)
  • Yes
  • No

0 voters

GM ser,

Does this reverse our Partner First approach by selling a higher proportion of their tokens overall?

Many thanks,

1 Like

No, not at all. We still sell just 20% total on aggregate. Right now 10% is sold as performance fee for the DAO (so 50% of the total 20% perf fee) and 10% is sold and autocompounded into the vaults (12.5% of the total earned by vault depositors).

With the proposed model up to the full amount of that 20% that we can sell will be used to drive yield in ibBTC, the amount thats used is driven by the amount of rewards earned by ibBTC each harvest. If there is BTC left over that flows to all other vault depositors. And the DAO no longer receives any BTC as part of performance fees.

1 Like

Thank you for re-explaining for me.

We’ve always had high hopes for ibBtc as a product for Badger. Have you thoughts on how to generate dao income from it in the future if it reaches its potential?

1 Like

We take small fess on mint and redeem, so once it is very liquid and highly traded, one would expect us to make money on arbitrage.

1 Like

GM. yes and as Tritum stated we get some mint/redeem fees. also ibBTC holds badger vaults which are all generating performance fees. We want ibBTC to be the largest vault depositors because

  1. Its locked in TVL and perf fees, i imagine more people will trade out of ibBTC vs redeem
  2. We dont distribute any BADGER to it (more for direct depositors)
  3. With this new model it juices yield for direct depositors.
1 Like

can you help me unwind some of the complexity around calculating the split between the WBTC going to ibBTC/depositors (top green note)?

i understand:

  • performance fee: 20%
  • yield retention: 60%
  • thus leaves 1-.2-.6=20% to sell for autocompounding [orange]

so for ibbtc:
ibbtc share = harvest * ibbtc harvest share
harvest = 10000 units * (1 - performance fee) = 8000
ibbtc harvest share = 8000 * .15 = 1200 [where 1200 is in WBTC]

and then for depositors:
depositor harvest share = 10000 * .6 = 6000
depositor autocompounding share = autocompounding - ibbtc harvest share
depositor harvest share = (10000 * .2) - 1200 = 800
depositor share = 6000 + 800 [where 800 is in WBTC]

done right? [1200+800] + 6000 + 2000 = 10000

i think i am confused by the ibbtc yield retention of 60%, which is not the same as the 6000 tokens going to depositors? if so, how are the 15% harvest share and 60% yield retention related to each other? cant those two parameters be brought back to one?

Hey!
So we will always sell the full 20% of the full harvest for BTC. So this means 2000 tokens in the example
60% of the total value of the harvest going to ibBTC goes to ibBTC. so in this example its 10000*.8*.15*.6
.8 - Net 20% performance fee
.15 - ibBTC harvest share (ibBTC holds 15% of the total supply of this vault token in this example)
.6 - Yield retention variable
This gets you to 720 tokens sold for btc and going back into ibbtc
since the max selling capacity is 2000 then that means 1380 will be sold and autocompounded into the vaults for everyone else.

Does that clear it up?

Aaaah I missed the fact that the 15% is the position of ibBTC in that vault! It all makes a lot more sense then, and I understand where the 720/1280 split comes from now.

1 Like

@jonto what is the status of this. overwhelmingly supported here and via snapshot vote 2 weeks ago. thx

delayed a bit with all of the focus on getting our locked CVX vote weight lined up for today, hopefully getting this back on top of the priority list next week

1 Like