kBTC Collateral Analysis: KSM
An analysis of KSM as collateral for kBTC. We propose collateral thresholds that should be used for KSM as collateral.
kBTC is the flagship product of Interlay on the Kusama (KSM) network. kBTC is a 1:1 Bitcoin-backed asset on Kusama, maintained by a decentralized network of collateralized vaults. Vaults lock Bitcoin and issue the equivalent amount in kBTC. KSM will be the first of many collateral to secure kBTC.
Recently, we conducted a risk analysis of KSM as collateral for kBTC. We used protocol simulations based on a number of assumptions and challenged KSM pricing. Our findings propose collateral thresholds that should be used for KSM to ensure security.
Here’s a quick overview of how Interlay has done a collateral analysis and proposes to mitigate risks through collateralization thresholds.
Multi-Collateral System of kBTC
Anyone can become a vault and lock KSM collateral on the Kintsugi parachain. The collateral prescribes how much kBTC a vault can issue, through a ratio called the secure collateralization threshold (https://docs.interlay.io/#/vault/overview?id=collateral-thresholds).
If collateralization falls below the liquidation threshold, the 1:1 Bitcoin peg is destabilised as there is now more issued kBTC than physical BTC backing it. To restore the peg, arbitrageurs can burn kBTC in exchange for KSM at a beneficial rate, this is known as the burn event. The liquidation threshold determines how beneficial the burn rate is.
At launch, KSM is the main collateral asset. Soon after launch the goal is to add more assets to create a diversified basket to improve stability.
The goal of our kBTC analysis is to minimize the risk of system under-collateralization (when kBTC can only be redeemed for less than its worth in BTC). We forecast worst-case trajectories for Kusama and Bitcoin using historic price data to stress-test the vaults.
Interlay is proud to have stablecoin and DeFi experts as part of our team. In 2020, our co-founder Dominik Harz co-wrote a paper The Decentralized Financial Crisis, Gudgeon et al, (https://arxiv.org/pdf/2002.08099.pdf). We used Gudgeon et al’s approach in our kBTC analysis, where price trajectories are modelled as a Geometric Brownian Motion (GBM) process. While GBM makes the assumption of constant price volatility, the 100-day forecast period produces a good number of outliers — which helps to identify the tail risk.
Accounting for slippage, the KSM research showed that the larger the debt ceiling in the bridge, the more likely that liquidators can move the market. The results also revealed that a large trade could affect KSM price more than it does BTC.
In a similar work, Gauntlet (https://medium.com/gauntlet-networks/karura-parameter-recommendation-methodology-6ce7fe06cb77) trains a price impact model to measure how quickly the market recovers after large trades.
Given the results of the price slippage analysis, liquidators can “improve” KSM:BTC price with respect to the security of Kintsugi if they start a new vault to mint kBTC. The size of the collateral ceiling does not affect the security of the protocol.
The first step in the simulation is for vaults to issue the maximum amount of kBTC as allowed by their collateral. The simulation generates multiple price trajectories for both KSM:USD and BTC:USD, while preserving historic correlation between these assets. The worst trajectory of the collateral is picked (where KSM price with respect to BTC is lowest). We checked vault collateralization over the 100-day simulated price trajectory. We observed collateralization that was below the liquidation threshold and what would happen with arbitrageurs and burn-redeem events. We tested the system when under-collateralized and investigated how to restore the peg.
We ran conservative simulations on the price of KSM:BTC, assuming a heavily congested Bitcoin network, and worst-case behaviour from vaults and liquidators. We identified a configuration that minimizes the risk of system under-collateralization (where kBTC can only be redeemed for less than its worth in BTC).
After our analysis, the following collateralization thresholds for KSM are recommended:
In the future, we can replace the GBM model with a Jump-Diffusion model, which can accommodate changing price volatility and will be better at producing outliers. We also plan on aggregating order data from multiple exchanges (both centralized and decentralized).
Let’s keep an eye on how the correlation between KSM and BTC changes over time and whether there are changes needed to these parameters.
You can learn how to participate in governance https://docs.interlay.io/#/guides/governance and read more about the role of collateral in Interlay https://docs.interlay.io/#/getting-started/interlay-101.
We invite our community to submit proposals for adding more collateral assets via our Subsquare governance platform https://kintsugi.subsquare.io/discussions.
We envision a future where blockchains seamlessly connect and interact: anyone can use any digital asset on any blockchain, trustless and without limitations. We work with Bitcoin, Ethereum, Polkadot, Cosmos, and others to expand interoperability, capital efficiency, and openness.
Our flagship product is interBTC — Bitcoin on any blockchain. A 1:1 Bitcoin-backed asset, fully collateralized, interoperable, and censorship-resistant, realizing the true free nature of BTC and decentralized finance.
by Daniel Savu (Software Engineer Interlay) and Rowena Wiseman