- Add a fee credit to user stats, which can be created by “burning” their Drift stake (reappropriating it to protocol-owned stake) according to a stable price, such as an oracle TWAP. This makes staking Drift more attractive and less restrictive compared to just unstaking with a long cooldown period.
Regarding how this fee credit works in practice, I would defer to the protocol engineers. It might be safest to allow redemption from the USDC revenue pool after:
- All or $100 of credit has been “spent.”
- A certain time period since the last credit was earned has passed (e.g., X days).
- Increase the protocol’s share percentage of insurance emission. Currently, most offer a 50% split between the protocol and stakers (of which the protocol is a part). Lowering the staker fraction seems reasonable given the risk/reward dynamics of insurance funds. Developing a system where additional time commitment and/or two tiers of insurance (e.g., Tranche A pays bad debt before Tranche B in exchange for higher rewards) would lead to a more effective insurance fund.
- Ensure an aggressive USDC interest curve and charge a larger borrow rate on USDC. High USDC rates but safe utilization is a win and a good place to migrate more fee collection to. Protocol-owned liquidity will benefit from this approach.
- Turn liquidation into logic. Drift tries to solve other parts of the system with auctions, so why not here? Pairwise spot takeovers versus perpetuals seem complex, but most liquidations come from perpetual leverage. This approach would both reduce user charges for liquidations by compensating liquidators more appropriately and improve liquidation scenarios where the oracle is lagging or too volatile.