[PRE GIP-X] Gearbox Revenue Model Revamp

The purpose of this revamped model is to streamline, identify, and quantify Gearbox’s revenue model in a manner that enables us to become profitable while ensuring that yield levels are optimal to compete with the rest of the market without feeling excessive.

Sources of revenue

We are going to focus on three key sources of revenue with an outlook to move to two as liquidity stabilizes. Further, the goal of these changes will be to move away from liquidation fees as the primary source of revenue to a model that’s more sustainable:

  1. Liquidation fee: Monetise user collateral

Liquidation fees are common in DeFi and are a source of revenue generation. They can further be utilized to set up an insurance fund over time.

Having a leverage protocol where funds can further be levered up or used in adapters entails increased risk. In particular, during times of high volatility, it is possible that the value of some accounts will drop below zero before they can be liquidated. Should these “underwater” accounts occur, they must be handled promptly in order to ensure the solvency of the system as a whole.

In order to address “underwater” positions, we require creation of an insurance fund that enables us to restore equilibrium to the system. To maintain the same levels, I propose we allocate 1% of our liquidation fee towards the insurance fund and another 1% as protocol revenue. The remaining 5% to be allocated to the liquidator.

Protocol Comp DyDx MakerDAO Aave Gearbox
Liq Fee 8% 13% 13% 5% 7%

As seen above, the liquidation fee is in-line with industry average and thus a reduction at the moment is not needed

  1. Funding fee spread- monetise utilized TVL

As a protocol that’s not defined it’s PMF yet, the best way to maximize revenue is to utilize our overall position size and the time that it has been open into a constant income stream. One strategic advantage we have over peer-to-peer based funding fee model is that the spread between supply APY is significantly lower than annualized funding fee traders pay on exchanges.

Referring to Binance blog in the appendix, the annualized APY for funding fee is 10.95%. Industry rates have gone as high as 13% annualized (add appendix reference). DeFi supply APY ranges between 0.3-3%, assuming average at 2%, it translates to 0.0018% funding fee every 8 hours.

We can monetise this by adding 0.0018% on top, which will lead to a monetisation of 2% commission on the daily average utilized TVL.

The ideal way to represent this will be to maintain lender APY as annualized percentage

While funding fee is represented at an 8 hour interval

This spread based model enables us to have an effective daily funding rate of 0.0108% daily, which is 64% lower than the lowest funding fee paid on CEXs.

Even if APYs in DeFi go 3X, our borrow rates will still be 38% lower than CEXs making this a great area of monetising TVL without excessively charging users. Even DyDx is 0.03% daily and this will be a massive competitive advantage.

This model also incentivises the DAO to make sure that liquidations are limited, positions are able to last the longest and in turn makes $GEAR as universal collateral a direct option to safeguard our continuous earning on utilized TVL.

  1. Withdrawal fee on TVL

Currently, there is a 1% withdrawal fee, but it poses as a barrier to add liquidity, considering users make that much APY in 6 months at current average APYs). We should be looking at significantly lowering this to about 1 month’s APY, 0.167%, as that should be deterrent enough to withdraw but won’t pose as a barrier to entry. Lender’s don’t usually pay withdrawal fees on most protocols.

Post V2, we should aim to completely remove the withdrawal fee.

Useful Links:

  1. Liquidation fees
    there we should remember that decreasing liquidation fee leads that small accounts don’t be liquidated (non-profitable for liquidators). so decreasing liquidation fee means that we should increase min collateral size for credit account. I don’t think it’s worth increasing barriers to entry yet - let’s focus on PMF first, understand who are our real users and then taki decisions on best value of liquidation fee.

  2. Interest rate spread
    We discussed on Discord funding rate <> interest rate spread correlation. as Interest rate spread is used in Gearbox, let’s put numbers here - what’s your suggestion for interest rate spread?
    Now it’s 10%.

  3. Withdrawal fee
    Agree, this should be decreased…

1 Like
  1. There’s no decrease in liquidation fee. 5% for liquidators as is now, 1% to DAO treasury, 1% to insurance fund to set up.

  2. It’s a combination of both, interest rate for lenders, funding rate for borrowers. We can charge 2% annualised with 8 hour periods, as is the norm, and still be the cheapest protocol wrt to funding fee. A spread on current interest rate equates to 0.05% annually which is extremely low and not sustainable

On liquidation fees and their uses:
Generally, I am very much in favour of building a solid insurance fund balance. The massive insurance funds of the established centralized exchanges played a major role in their professionalization and the establishment of trust in the absence of regulation akin to legally mandated deposit insurance (FDIC in the United States; similar insurance schemes in most other advanced economies). Binance’s insurance fund, for example, has a current balance of > USD 300 million (see screenshot below). Obviously, Gearbox caters (and likely always will) to a more sophisticated and less risk averse group but adding these backstops increases our potential user base and lowers users’ required rate of return threshold in their decision whether to open a credit account via Gearbox as they demand a lower risk premium.

The current liquidation fee does indeed seem fine. We should keep in mind here that we have the dual objective of (i) remaining attractive to protocol users (comparison most appropriate with other platforms enabling the same kinds of strategies) and (ii) liquidations remaining profitable for liquidation bots (liquidation fee covers slippage, gas fees, difference between oracle and market price etc) to ensure timely liquidations and avoid bad debt.

On interest rate spreads / funding fees:
We could also calibrate this to be a certain share of average return achieved by credit account owners (say, 10%) and translate that to a certain interest rate spread parameter. I believe these recurring fees should, together with liquidation fees, be the main revenue source for the protocol as they are proportional to the actual usage and do not introduce additional friction as is the case for the withdrawal fees.

On withdrawal fees:
I agree that the current figure should be lowered substantially (0.1% max). As described above, I would prefer it if this was not a major source of protocol revenue (prefer monetization via the other two routes); a small withdrawal fee may be fine here but this may act as a deterrent and drive potential users away. For the sake of transparency as well, I believe we should very clearly communicate any such fees prior to the credit account opening so as to avoid any bad user experiences due to seemingly hidden fees and costs associated with protocol usage.


I think the confusion is that the liquidation fee is written as “I propose”, implying a change. Perhaps just re-word that first part to clarify, maybe “I do not propose a change to the liquidation fee. It shall remain as …”