Alchemix
([REVIEW] [GF: PHASE 1 CYCLE 7 PROPOSAL] Alchemix)
Recommendation
We broadly support Alchemix’s proposal as written but advise halving their ask generally by time and size – there’s simply no good argument for running a grant past six months, and even that amount of time is a stretch.
They’re in the minority of proposals to date that have offered a bottom-up approach in outlining the size of their ask and explaining the additionality of the grant, and we’re inclined to agree with their view.
Another key piece is that Alchemix’s proposal is rare in that they are explicitly committing to move resources over to Optimism themselves — that is, they’re not just trying to draw users and capital, they’re moving the deployment of their own capital to encourage more use on Optimism. This goes beyond the simple idea of “incentive matching” — it’s a reflection that a grant can beget follow-on activity, and it’s exactly the kind of behavior we should be encouraging.
We think there is still some work to be done to determine KPIs, demonstrate migration from Mainnet, and discuss potential integrations with other ecosystem protocols, but these are nits in an otherwise well-presented proposal that seems advantageous for Optimism.
Background
Alchemix is a unique yield protocol that facilitates highly-touted self-repaying loans. The system works like this: deposit some amount of collateral and then mint up to 50% of that collateral value as a defi-composable synthetic derivative, effectively trading your expected collateral yield for immediate liquidity.
The thinking is: your initial collateral earns yield and is pegged to your loan, and over time, that yield pays back the interest and principal on your loan, progressively unlocking your collateral while ostensibly removing liquidation risk. This allows you to get the most out of your money without having to stake or sell it. It’s been a popular proposal that earned the protocol almost a billion in TVL in 2021.
In addition to being a draw for users, what’s compelling about Alchemix is the stimulus it provides the rest of the ecosystem. Specifically, it magnifies velocity on the capital that’s deposited; collateral is deposited in yield protocols like Yearn, and derivatives are minted against them, which can be swapped into other stables (thus creating more demand for them) or deposited into other protocols for more general defi use. Users benefit, protocols benefit, activity increases.
In the past year and a half, Alchemix has legitimately established itself as a protocol with real PMF, utility, and mindshare — and, compellingly, no analog on Optimism.
Asks
250,000 OP to vault incentives.
Estimated to incentivize $25m of deposits (at an approximated market-determined boost of 1% APR on AAVE strategies)
Alchemix “expect at least 25% of AAVE Optimism TVL in aWETH and stablecoins to migrate to Optimistic Alchemix.” - we’re not entirely sure what this means here, whether this is referring to the existing base of WETH and stables on AAVE Optimism (which are in the hundreds of millions) or specifically vault pools on transmuter assets from Mainnet.
Although we recognize the long-term bias of these deposits, we believe that there ought to be something of an understood cap on grant durations due in part to the quickly shifting crypto ecosystem, the need to iterate on how we distribute OP to protocols, and our desired reorientation toward more grants for more tightly scoped objectives.
250,000 OP for Velodrome bribes for one year on alETH and alUSD pools.
Estimated $3m additional liquidity at 10% APR (Alchemix currently incentivizes $2.5m of liquidity on Velodrome)
Liquidity is indeed a key need for Alchemix, as alTOKEN liquidity is a legitimate rate limiter for the uptake of deposits and increased money velocity. In a vacuum we would generally discourage this type of liquidity focus, but we’re in favor of this particular because of Alchemix’s unique need. In addition, there’s an opportunity to drag a lot of stablecoin liquidity into these systems by giving Alchemix a headstart in autonomously generating more liquidity.
In an ideal world, we would reorient this grant to hasten alOP development to allow for a potentially highly appealing alOP-OP pool. We’d almost have prioritized this over bribe subsidies in the hope that this could get built before demand really scales.
What this does for Optimism
The question is: does this grant bring power users from Mainnet to Optimism? We know that this activity is likely stickier than the average defi activity; users are depositing to take out a loan generally to term. That is use that sticks without resorting to locking or similar mechanisms. What isn’t clear to us is whether existing users on Mainnet will be incentivized or even able to migrate their capital. We’d love to see some clear expectations in this regard (versus merely sliding some TVL from AAVE Optimism over).
Even if there is some layer cannibalization, we see real opportunity for alUSD and other synthetics to establish a foothold on Optimism and for Alchemix to begin to explore more yield solutions on Optimism, which are quickly growing in number. With AAVE taking on such a large proportion of Optimism’s TVL almost immediately, it seems clear that Alchemix could almost unilaterally lead a small ecosystem to spring up around it. We’d love to hear more about what they have in mind on the yield and composability side.
Another key piece is that Alchemix’s proposal is rare in that they are explicitly committing to move resources over to Optimism themselves — that is, they’re not just trying to draw users and capital, they’re moving the deployment of their own capital to encourage more use on Optimism. This goes beyond the simple idea of “incentive matching” — it’s a reflection that a grant can beget follow-on activity, and it’s exactly the kind of behavior we should be encouraging.
One way or another, we think this is a good time to be supporting Alchemix’s initiatives, as they target large players and protocols, entities that are still active in today’s market.
The problem is that these are questions that should have been broadly addressed during the public review session when instead it devolved into questions about self-delegation. We don’t fault Alchemix for not having gone more into this, but we do appreciate the thoughtful direction @marc offered.