The goal of this proposal is to:
Therefore, the proposal suggests to:
Background: Interlay’s analysis repository used to estimate the collateral thresholds, applies a value at risk (VaR) approach which uses the analytical method (based on monte carlo simulation) as well as the historical method (based on actual historical data). VaR scales by the square root of time (imagine a horizontal parabola) , such that the result of the analysis would always estimate the largest increment in VaR from the 1:1 peg to the liquidation threshold. These increments would become smaller, such that the relative and absolute difference between the secure mint threshold
and the premium redeem threshold
was relatively small. Hence, vaults have a small margin for price changes of their collateral to BTC and run the risk of quickly receiving premium redeem requests.
Change: This shifted some risk from the protocol towards the vaults and was a conservative approach in order to protect the protocol (iBTC peg) as much as possible. By now we know how swift users are, when it comes to 'selflessly protect' the protocol by requesting premium redeems against vaults or make liquidations. Therefore, it seems to be suitable to give vaults a better margin of safety before getting penalized. This can be done by systematically rearranging the increments of the thresholds in reverse order.
Result: As you can see in the table below, the secure mint thresholds
under the old and new model remain the same, just the distribution of the increments between each threshold are shifted, so that there is more margin for price movements before falling under the premium redeem threshold
. It is important to make clear that this does NOT change the likelihood of the 1:1 peg breaking.
You may have noticed that the thresholds in the table above are different from the current ones for both models. As new price data becomes available, it is sound to re-run the analysis to incorporate these new information into the estimation for the thresholds.
Below are the suggest thresholds and system ceilings for the tokens, based on the results of Interlay’s analysis repository.
Collateral | Liquidation Threshold | Premium Redeem Threshold | Secure Mint Threshold | System Collateral Ceiling |
---|---|---|---|---|
DOT | 125% | 140% | 155% | 2,450,000 |
qDOT | 125% | 140% | 155% | 1,000,000 |
USDT | 135% | 145% | 155% | 2,000,000 |
qUSDT | 135% | 145% | 155% | 600,000 |
vDOT | 140% | 155% | 165% | 500,000 |
Collateral | Liquidation Threshold | Premium Redeem Threshold | Secure Mint Threshold | System Collateral Ceiling |
---|---|---|---|---|
DOT | 105% | 115% | 130% | 2,450,000 |
qDOT | 105% | 115% | 130% | 1,000,000 |
USDT | 105% | 115% | 150% | 2,000,000 |
qUSDT | 105% | 115% | 150% | 600,000 |
vDOT | 105% | 115% | 135% | 500,000 |
SubSquare has proposed a plan to increase the mint capacity for iBTC and make the bridge more efficient. They want to reduce the risk for vaults so that it is cheaper for them to provide capital. The proposal suggests updating collateral thresholds with a new model and new price data to reduce thresholds and increase total mint capacity by up to 15%. They also suggest rearranging the threshold increments to provide more safety margin for vaults without increasing the risk of under-collateralization of iBTC. The proposal does not change the likelihood of the 1:1 peg breaking. The suggested collateral thresholds and system ceilings for the tokens are shown in tables.