In the ever-evolving landscape of DeFi, risk management remains a paramount concern. The ongoing success and sustainability of DeFi protocols often hinge on the careful calibration of various market parameters. Among these parameters, the Supply Ceiling, Global Ceiling, Daily Borrow Limit, Collateral Factor, Liquidation Factor, Liquidation Incentive, Liquidation Fee, and Minimum Debt Amount hold critical roles in shaping the performance and security of lending platforms.
This master sheet serves as an insightful exploration into the methodologies employed by Inverse Finance's Risk Working Group (RWG) to determine these market parameters within FiRM, Inverse's fixed-rate lending protocol. Through an understanding of the methodologies behind these market parameters, stakeholders can gain valuable insights into how FiRM strives to maintain a delicate balance between user empowerment, platform efficiency, and risk mitigation. There are additional security features built into FiRM that interact with the various market parameters and, in doing so, play a crucial role in maintaining the elevated security posture that characterizes our lending protocol.
Setting parameters for new markets and changes to existing markets on FiRM are the end result of recommendations presented in “Risk Assessments” authored by the RWG. These recommendations serve as a starting point for an informed conversation amongst core contributors and community members alike of Inverse Finance DAO. By providing clear parameter recommendations, the RWG ensures that new markets added to the FiRM protocol are appropriately risk-managed and able to operate in a safe and sustainable manner. However, these are intended solely for informational purposes and do not constitute financial or investment advice. The assessments do not guarantee the outcome of the proposed initiative.
Definitions
The Supply Ceiling helps manage risk by limiting the total amount of DOLA that can be supplied to a specific market.
The Global Ceiling refers to the overall maximum limit set on the total supply of DOLA that can be lent out across different collateral assets. This parameter restricts the total exposure of the protocol to DOLA loans and helps manage risk.
The Daily Borrow Limit is the total amount of DOLA that may be borrowed within a 24-hour UTC day index and is designed to balance user demand for borrowing against a collateral asset with the same collateral’s capacity to handle large sell orders without excessively impacting the underlying price. The Daily Borrow Limit acts as a guarded launch mechanic and risk exposure limiter for collaterals on FiRM. Limiting borrows over 24-hour periods allows the RWG to observe new DOLA collateral backing in a controlled environment.
The Collateral Factor is the maximum value of a DOLA loan relative to the collateral staked against it. The Collateral Factor helps ensure the protocol remains adequately collateralized, even during market downturns.
The Liquidation Factor determines the maximum amount of debt a liquidator can repay in a single liquidation transaction. The Liquidation Factor plays a pivotal role in ensuring that loans remain sufficiently collateralized, especially in rapidly changing market conditions.
The Liquidation Incentive is an essential parameter in the FiRM protocol, as it encourages third parties to participate in the liquidation process, maintaining the protocol's health. The Liquidation Incentive must strike a balance between attracting liquidators and not excessively penalizing borrowers.
The Liquidation Fee is an important parameter in the FiRM protocol, as it is charged when undercollateralized loans are liquidated. This fee is awarded to the operator of the lending market (Inverse Finance treasury) and contributes to the protocol's revenue. It's important to clarify that the Liquidation Fee is different from the Liquidation Incentive. The Liquidation Fee is a charge imposed when liquidating undercollateralized loans, while the Liquidation Incentive is also designed to attract third parties to participate in the liquidation process.
The Minimum Debt Amount establishes the lowest amount of DOLA a user may borrow in a given market, which ensures that borrowers transact with a sufficient amount of debt to attract liquidators in the event of a liquidation. It is therefore a vital parameter in the FiRM protocol, and it serves to maintain operational efficiency during environments of high transaction costs.
Considerations
Determining the appropriate parameters involves a multifaceted approach that encompasses qualitative evaluations of market behavior, quantitative analysis from historical records, diligent collection of relevant data, and scrutinizing on-chain activities to fully understand the current and future states of the collateral being studied. Some of these are detailed below.
Supply Ceiling
Liquidity Analysis: The first step in determining the Supply Ceiling is to analyze the on-chain liquidity depth of the collateral asset. Only same-chain liquidity is considered, with an understanding that liquidity on other chains can play a supportive role but cannot be relied on due to bridge timing considerations. The same consideration holds for off-chain (CEX) liquidity. Higher liquidity assets are more suitable for higher Supply Ceilings.
TVL Consideration: In cases where the studied collateral is a vault receipt token or a Liquidity Provider Token (LPT), the TVL metric provides similar insight to the liquidity analysis detailed above. It reflects user demand for that asset. A higher TVL may suggest that users are interested in supplying more of that collateral, which can influence the Supply Ceiling.
Historical Analysis: Historical data on liquidity, TVL, and price stability of the collateral asset can provide valuable insights. It helps assess how the asset has performed in the past, including protocol-specific stress tests, periods of extreme market volatility and potential black swan events, and whether it is suitable for a higher Supply Ceiling.
Risk Assessment: A risk-centric study is conducted via the asset scoring model. This framework evaluates the relative “risk” of the collateral asset, using wETH as a benchmark, by considering six essential factors: market capitalization, trading volume, price volatility, token distribution, project fundamentals; and token utility. The model assigns a weighted score for each factor, producing an overall risk score that can be used to rank tokens according to their risk profiles. Higher scoring collaterals warrant higher a Supply Ceiling.
Liquidation Cascade Consideration: As a precautionary measure, the RWG models liquidations on the assumption that all borrows in a market originate from a single wallet. In doing so, the Supply Ceiling has to be set as such so that the theoretical maximum price impact originating from a liquidation sell order is less than the liquidation factor, and thus does not threaten to cause a liquidation cascade. Such a scenario puts the protocol at elevated risk of incurring bad debt.
Diversification: The protocol aims to maintain a diversified pool of collateral assets to reduce systemic risk. The determination of the Supply Ceiling should consider the overall diversification strategy to ensure that the protocol is not overly exposed to a single asset.
Daily Borrow Limit
Transaction Volume Analysis: To establish the appropriate Daily Borrow Limit, the protocol must analyze daily transaction volumes for each collateral asset. This analysis helps gauge the demand for borrowing and provides insights into how much borrowing activity the protocol can safely support.
Risk Management: The Daily Borrow Limit is not a one-size-fits-all parameter. It is adjusted based on the risk associated with each collateral type. Assets with higher volatility or lower liquidity may warrant lower Daily Borrow Limits to mitigate potential risks.
Monitoring User Behavior: The behavior of users within the protocol is essential to consider. Some users may engage in strategies that involve frequent borrowing and repaying of small amounts. This behavior can have implications for the Daily Borrow Limit, as it can lead to excessive gas costs and operational inefficiencies.
Market Conditions: Like other parameters, the Daily Borrow Limit is periodically reviewed and may be adjusted to adapt to changing market conditions. High volatility or declining on-chain liquidity may necessitate temporary adjustments to the Daily Borrow Limit.
Collateral Factor
Risk Assessment: A risk-centric study is conducted via Inverse’s asset scoring model. This framework evaluates the relative “risk” of the collateral asset, using wETH as a benchmark, by considering six essential factors: market capitalization, trading volume, price volatility, token distribution, project fundamentals; and token utility. The model assigns a weighted score for each factor, producing an overall risk score that can be used to rank tokens according to their risk profiles. Higher scoring collaterals, particularly those with lower price volatility, warrant a higher Collateral Factor.
Market Liquidity: On-chain liquidity of the collateral asset is assessed. Collateral assets with higher liquidity are generally more suitable for higher Collateral Factors, as they are easier to liquidate in case of under-collateralization.
Stress Testing: The Risk Working Group conducts stress tests to evaluate how each collateral asset would perform under extreme market conditions. Stress testing helps determine the appropriate Collateral Factor to ensure the protocol remains secure even during severe market downturns.
Black Swan Event Consideration: The analysis should take into account the potential impact of black swan events, which are rare and unexpected events that can cause significant market disruptions. The Collateral Factor should provide a buffer against such events.
Liquidation Cascade Consideration: As a precautionary measure, the RWG models liquidations on the assumption that all borrows in a market originate from a single wallet. In doing so, the Collateral Factor has to be set conservatively in such a way that the price impact ensuing from a sell order originating from the maximum seizable collateral from an undercollateralized position (determined by the liquidation factor) is not large enough to cause a liquidation cascade. Such a scenario puts the protocol at elevated risk of incurring bad debt.
Adjustments: Collateral Factors should ideally be static, meaning recommendations on the parameter should be forward looking/future facing. The protocol should have the flexibility to adjust these factors as market conditions change, but should seldom act.
Liquidation Factor
Minimum Debt Amount Analysis: The analysis begins with the determination of the Minimum Debt Amount. This is the minimum debt size that ensures liquidations remain economically viable, even in high gas environments. It discourages trivial borrowing and spam loans by setting a threshold that borrowers must meet.
Cost to Liquidate Position: The next step involves calculating the cost to liquidate a borrower's position. This includes estimating the gas fees, transaction costs, and any other expenses associated with the liquidation process. The cost to liquidate should be considered relative to the collateral value.
Data-Driven Analysis: The Risk Working Group conducts data-driven simulations to estimate the potential costs and outcomes of liquidations for various collateral assets. These simulations consider different scenarios, including varying levels of market volatility and gas prices.
Liquidation Incentive Alignment: The Liquidation Factor is set to align with the Liquidation Incentive. This incentive is essential for encouraging third parties to participate in the liquidation process. The Liquidation Factor should ensure that liquidators have a reasonable chance to earn the incentive while covering their costs (gas and slippage).
Market Conditions Consideration: The protocol's Liquidation Factor should be adaptable to changing market conditions. During periods of high volatility or congestion on the Ethereum network, a more conservative Liquidation Factor may be warranted.
User Experience: The Liquidation Factor should strike a balance between security and user experience. It should provide borrowers with some leeway to manage their positions and avoid premature liquidations, while also safeguarding protocol interests.
Liquidation Incentive
Alignment with Industry Standards: The Liquidation Incentive should be set to align with industry standards and practices. It's essential to offer liquidators an incentive that is competitive with other protocols to attract participation.
Liquidity Assessment: The liquidity of the collateral asset is considered when setting the Liquidation Incentive. Less liquid assets may require a higher incentive to attract liquidators, as it can be more challenging to sell these assets quickly.
Volatility Consideration: The volatility of the collateral asset is another important factor. Highly volatile assets may require a higher Liquidation Incentive to compensate liquidators for the increased risk associated with price fluctuations.
Liquidation Fee Alignment: The Liquidation Incentive should work in tandem with the Liquidation Fee (in cases where both are present for a market), ensuring that both parameters promote responsible market user behavior and market stability.
User Experience: While incentivizing liquidators is essential, the Liquidation Incentive should not excessively penalize borrowers. It should strike a balance that encourages borrowers to manage their positions effectively while safeguarding protocol interests.
Liquidation Fee
Revenue Generation: The primary purpose of the Liquidation Fee is to generate revenue for Inverse. Fees collected from liquidations contribute to the sustainability and operations of the lending platform.
Market and Competitor Analysis: The protocol may conduct market and competitor analysis to assess the fees charged by other lending platforms in the DeFi space. This analysis helps ensure that the Liquidation Fee remains competitive and attractive to users.
User Experience: While the Liquidation Fee is a revenue-generating parameter, it should not be set so high that it discourages borrowers from using the platform. Striking a balance between generating revenue and maintaining a positive user experience is crucial.
Alignment with Liquidation Incentive: The Liquidation Fee should be set in coordination with the Liquidation Incentive. Both parameters should work together to promote responsible market user behavior and market stability.
Minimum Debt Amount
Data-Driven Analysis: The determination of the Minimum Debt Amount is based on a data-driven analysis. This analysis considers several key factors including Collateral Factor, cost to liquidate, and gas fees.
User Behavior Patterns: Historical user behavior patterns are analyzed to understand how borrowers interact with the platform. This analysis helps identify trends in borrowing behavior and informs the setting of the Minimum Debt Amount.
Spam Prevention: The Minimum Debt Amount acts as a spam prevention measure, deterring users from repeatedly borrowing and repaying tiny amounts of assets. This prevents the platform from being overwhelmed by trivial transactions.
User Accessibility: While discouraging trivial borrowing, the Minimum Debt Amount is set to ensure that the platform remains accessible to a wide range of users. It should strike a balance between accessibility and operational efficiency.
Formulas
Correlations
Rules
The relationship between Collateral Factor (γ) and Liquidation Factor (η) is a delicate balance. When γ is set high, borrowers can access more liquidity, but the risk of a larger undercollateralization (in $ terms) increases, necessitating a higher η. Conversely, a lower γ reduces borrowing capacity but allows for a lower η, ensuring that smaller repayments can restore the loan's health.
The sum of Collateral Factor (γ) and Liquidation Incentive (x) can never exceed 100% as it creates self profiting liquidations and puts the protocol at risk of bad debt.
The Daily Borrow Limit is our “guardian” parameter and sets a theoretical maximum to bad debt accrual within in a single block.
More to come...
