Summary
- Utilize YB available to Curve DAO fully for bribing for crvUSD stablecoin pools (mostly to crvUSD/USDC and crvUSD/USDT, crvUSD/pyUSD, but also can include crvUSD/frxUSD, crvUSD/USDf etc if those collaborate for partner rewards). This stream is dynamic, up to 360k YB per week.
- Raise PegKeeper limits from current $108M crvUSD in total to $300M crvUSD in total.
- Give $1B crvUSD allocation for Yield Basis, giving a capacity of $500M to Yield Basis.
(1) and (2) are necessary economic prerequisite before raising Yield Basis capacity. It is also highly recommended for Yield Basis to prepare admin fee distribion switch to be turned on before raising the capacity to have YB token price immediately reflecting the economic value of the system.
Effect on peg stability and required crvUSD liquidity
Stability of crvUSD is required not only for the health of crvUSD itself but also fot the health of Yield Basis. The presence of yb-BTC Curve pools affects the peg in the following way.
For example, when WBTC is being sold - yb-WBTC pool temporarily gets more WBTC and less crvUSD. Most likely, the trader exchanges to USDC or USDT. So, crvUSD goes to crvUSD/USDC stablecoin pool in this exchange, and the stablecoin pool becomes imbalanced, holding a bit more crvUSD. This makes price of crvUSD less than 1.0. Similarly, when one buys WBTC - crvUSD price becomes bigger than 1.0.
In Yield Basis, two things can be out of balance: one is debt of the releverage AMM (equilibrium fraction of the debt being 50%), and the other is balance of Curve cryptopool, the equilibrium fraction of crvUSD also being 50%.
Debt amount in YB releverage AMM appears very stable even in extreme market volatility:
This is very good because it means that the position is almost ideally 200% collateralized at all times.
Ratio of crvUSD in Curve cryptopool, on the other hand, is more volatile:
If we exclude MEV, crvUSD fraction in cryptopool varies between 40% and 60%, which is deviation from matching the debt by 10% of the value of the Curve cryptopool.
At the moment, YB Curve cryptopools have 100M TVL each (WBTC, cbBTC, tBTC), so the amount of crvUSD which has to be absorbed is 30M.
Stablecoin pools have 50% of them working as a supply sink for crvUSD, and they can probably afford to be misbalanced by another 25%. Thus, the TVL required in Curve cryptopools would be:
Stablecoin pool TVL = 30M / 0.75 = 40M.
This is correct only for current TVL of the cryptopools being 300M which corresponds to YB TVL 150M.
Yield Basis revenues
Revenues which Yield Basis is making should be ideally significantly bigger than incentives to crvUSD pools we are about to make, even if we want to pay those incentives out from YB token inflation designated for “Curve licensing fee”. Even though, YB pools are to be modified with a new version, fundamental underlying earnings and AMM stay the same. Currently, underlying returns of Yield Basis look like this:
This corresponds to fundamental (not counting any token part) APYs for the part shown on the picture being:
- WBTC market: 15.8% APY;
- cbBTC market: 21.1% APY;
- tBTC market: 13.0% APY.
This averages out to 0.042% per day, or 63K USD worth per day / 442K per week. These are going to be split between liquidity providers and admin fees, and we need crvUSD liquidity pools to consume much less incentives at equilibrium, even if given in “free” YB tokens.
Sustainability of crvUSD pool incentives
Typical APR of stablecoin pools accepted by the market seems to be around 10% APR for crvUSD. For the necessary 40M USD TVL, that means $4M per year, or $11k per day = $77k per week. This is MUCH SMALLER than returns of YB which come down to $442K per week - good!
Now, practically, Curve DAO has a stream of up to 75M YB inflation in total. At today’s YB prices, it comes to up to $158k YB worth of YB per week, as opposed to the required $77k per week. We further amplify it by buying votes, making up to $237k worth of CRV per week, assuming bribing efficiency of 1.5: 3 times lager than required.
It means, if we use ALL YB emissions designated to Curve DAO for bribing votes for crvUSD stablecoin pools (crvUSD/USDC, crvUSD/USDT, crvUSD/pyUSD etc) - we have enough capacity to raise caps of Yield Basis by factor of 3, e.g. to TVL of $500M (1B crvUSD allocation).
PegKeeper capacity
Bribing for crvUSD will create crvUSD pools of a size of 120M, creating a 60M supply sink for crvUSD. If we are looking at raising caps of YB to 500M, we have to prepare PegKeepers to absorb at least +$133M more in crvUSD capacity. Therefore, I propose to increase PegKeeper allowances to be able to absorb in total of around $300M crvUSD, up from the current $108M crvUSD PegKeeper capacity.
