I re-ran the collateralization-analysis
for USDT and USDC and here are the results:
Start analysing usdt...
The suggested liquidation threshold is 110%
The suggested premium_redeem threshold is 115%
The suggested safe_mint threshold is 155%
Start analysing usdc...
The suggested liquidation threshold is 105%
The suggested premium_redeem threshold is 115%
The suggested safe_mint threshold is 165%
This actually means that we should adjust the safe_mint
UP by 5-10%. Agreed, USDT and USDC themselves have been very stable during downturns. But it's the USDT/BTC and USDC/BTC ratios that would cause a vault to be liquidated. DOT and vDOT vaults have lower safe_mint
thresholds because as crypto assets they're more correlated to Bitcoin.
There definitely were several premium redeems during the last downturn. The big liquidations I saw were on Kintsugi where perhaps there wasn't as much activity. I think the 5% is sufficient still and perhaps more premium redeems didn't occur because people hadn't been able to accumulate iBTC ahead of time.
I don't think enough BNC and HDX holders will deem it worth their risk for <20% APY.
We as Interlay gov/community simply enable the capabilities and then individual operators can make that decision themselves. I would be willing to run an HDX vault and I'll just set a high enough collateralization level to avoid large price swings causing a premium redeem.
Edited
I am in favor of this for it will allow vault runners more options to choose over what asset they wish collateralize
i think this is an interesting proposal - the collateral ratios should be higher at the start imho to account for the volatility and limited venues to complete liquidations. So, start slow and grow.
One thing that I would want to see here is a DAO-DAO agreement where BNC and HDX commit to contributing back to Interlay. For example:
The ultimate outcome here would be for DAOs to pay Vaults staked in their native token to secure BTC that DAO acquires as a diversification via iBTC - think, decentralized BTC custody. Rewards could flow eg via staking emissions in the native asset
Edited
I understand the motivation behind this is to increase the total capacity of vault operators and bring more iBTC into the ecosystem. While I do believe that adding HDX and BNC may very well achieve this in bull cycles, I worry about the impact such high-beta coins may have on Black Swan days, especially given that we have seen that premium redeems are ineffective on such days. Only a handful of vaults saw premium redeems during the last big drop in April. iBTC holders either do not want to part with their iBTC for a mere 5% or they do not act quickly enough to come to the aid of struggling vaults. We saw two large vaults get liquidated as a result. Already, the volatility of DOT and VDOT and their liquidation risk profiles are barely worth the vault rewards. I don't think enough BNC and HDX holders will deem it worth their risk for <20% APY.
I suggest lowering safe_mint thresholds for our stablecoins instead. They have proven to be resilient on large swing days. They, in fact, have the opposite effect, stabilizing our ecosystem when the price of BTC plunges and allowing users to quickly wrap more. We currently have around $650k in USDC and USDT. Lowering safe_mint from 150% to 130% would currently only add 1 iBTC in capacity, but could convince some vault operators to open up more stablecoin-backed vaults and it would be the equivalent of adding $125k in BNC and HDX collateral.
Edited