I started a draft of a proposal to reward borrowers below. It is still raw and needs work. As mentioned in the “timing” section and above by travisformayor, this should not be a top priority, but I still thought it was worthwhile sharing.
To-dos:
- Solicit and integrate community feedback
- Calculate an appropriate proposal for reward weight
- Implement proof-of-concept reward calculation and merkle distributor
- Consider other options for distribution mechanism
[DRAFT] Incentivizing Debt/Borrowing on Alchemix
Intro & Rationale
Incentives via governance token emissions are a powerful tool to bootstrap growth for DeFi protocols. Alchemix has already shown how effective these can be -- the existing incentives have created a $130M backstop for the alUSD peg, $300M of liquidity for alUSD, as well as deep liquidity for ALCX.
Currently, use of the platform to take out a loan (and let it pay itself off) is not directly incentivized. It is important for Alchemix to have a healthy balance of outstanding loans, as this is the primary source of revenue for the protocol. It is for this reason that I propose creating an incentive for outstanding debt via ALCX emissions.
A Note on Timing
The scarcest resource Alchemix has right now is bandwidth and mindshare of the core team, so it is important to prioritize appropriately. While it is positive to discuss the merits/drawbacks of this proposal, incentivizing debt should not be considered until after at least:
- Audits are completed
- v2 plans are well understood
- Any other high priority items the core team has planned
Prior Art
Like Alchemix, Compound’s revenue depends on borrowers using the protocol. Users are rewarded with COMP tokens based on how much they borrow and how long they borrow for. Similarly, Alpha Homora rewards ALPHA tokens to users who borrow ETH and leverage their LP positions. Both programs have been effective at attracting and keeping end-users.
Incentive Allocation
The purpose of these incentives is to generate real usage, add TVL to the Alchemist contract, and keep it there where it can earn yield and generate protocol revenue. Therefore the reward should be allocated at a constant rate relative to outstanding debt.
Every block, each debt holder would accrue an award amount that is a function of a) the user’s amount of outstanding debt b) the total amount of outstanding debt and c) the ALCX emissions designated for this incentive.
reward = (userDebt / totalDebt) * emissionRate * emissionWeight
Incentive Magnitude
To ensure a sustainable model for user acquisition and growth, unit economics must be considered when determining the incentive amount for borrowers. We want to show that these incentives could feasibly be reduced in the future to the point where they are no longer needed. In the long-term, Alchemix should not pay more in incentives than the revenue that those incentives would generate.
I propose a 5-10% annualized reward. Given the current $100M debt ceiling, and at current ALCX price and emission rate, that would be X-Y% [note: still need to calculate these values] of emissions.
I would also propose that the current alUSD pool reward weight be reduced to zero and be re-allocated to this new incentive. Through self-liquidation, it is possible to get the benefits of that pool without keeping an outstanding loan.
Incentive Distribution
Reward could be distributed in much the same manner as Alpha Homora currently distributes their weekly ALPHA tokens. A calculation is made each week by the multisig to determine reward amounts for each borrower, and a Merkle distributor contract is deployed where users can make claims.
If we choose this method, it is vital that the Alchemix UI can show approximate/pending rewards so as to meet user expectations.
This has the advantage that none of the core Alchemix contracts would need to be modified. However, it does create a weekly chore for developers that can take time away from other tasks. We should strongly consider other proposals for distribution that require less overhead.
Pros
- Incentivizes use of the protocol to grow TVL and revenue
- Rewards can be reallocated from current alUSD stakers
Cons
- Adds weekly chore to developers unless new distribution method proposed
- Could create too strong/unsustainable incentive to borrow
- We may not need to incentivize borrowing at all, especially if version 2 of the transmuter attracts more users on its own