Lending Protocol
Market Design Principles
Curvance markets are built with a focus on risk management, liquidity efficiency, and economic security. The protocol employs a novel market architecture that enables both capital efficiency and strong risk management.
Thesis-Driven Markets
Curvance creates thesis-driven micro-ecosystems. Each Market Manager focuses on a specific financial thesis:
Interest-bearing stablecoins.
Bluechip long market exposure.
Volatile LP tokens for a particular DEX or perpetual platform.
Other specialized asset categories.
This targeted approach allows for customized risk parameters appropriate for each asset class, rather than forcing disparate assets to share the same risk model.
Systemic Risk Reduction
By segregating markets by thesis, Curvance minimizes the contagion risk between asset classes. A volatility event in one market doesn't propagate to unrelated markets, protecting the overall protocol health. Isolated Market Managers are designed to contain risk within their boundaries. If extreme market conditions impact one Isolated Market Manager, other Market Managers remain unaffected, ensuring protocol stability.
Core Market Components
Two-Token System
Curvance employs a dual token model:
Position Tokens (pTokens): Collateral tokens that can be posted as security for borrowing.
Debt Tokens (eTokens): Tokens that can be borrowed against posted pToken collateral.
Both are collectively referred to as Market Tokens (mTokens), with each serving a specific purpose within the ecosystem.
Market Manager
The Market Manager is the central contract that:
Manages risk parameters for all tokens in its market.
Handles collateral posting and borrowing interactions.
Coordinates liquidation processes.
Enforces position health requirements.
Monitors and applies interest rate models.
Position Flow State Machine
Deposit: User deposits underlying assets into a pToken.
Collateral Posting: User posts pToken shares as collateral.
Borrowing: User borrows eTokens against posted collateral.
Repayment: User repays borrowed eTokens.
Withdrawal: User withdraws collateral after repaying debt.
Asset Management
No Rehypothecation
Unlike many lending protocols, Curvance disables rehypothecation of position token deposits. This means:
Collateral assets cannot be re-lent to other users.
Each asset's risk exposure remains isolated.
Risk modeling becomes more accurate and predictable.
Complex collateral chains that could amplify systemic risk are avoided.
Collateral Management System
The collateral system has several key components:
Collateral Caps: Each pToken has a maximum amount of shares that can be posted as collateral, limiting exogenous risk exposure.
Collateral Posting: Assets are posted as shares, allowing collateral caps to grow proportionally with any yield-generating strategies.
Cap Management: Collateral caps can be decreased even if current utilization is above the new cap, which prevents new risk while not forcing position unwinding.
Dynamic Liquidation Engine (DLE)
The Dynamic Liquidation Engine enables more nuanced position management:
Liquidation State Machine:
Healthy Position: Collateral value exceeds required thresholds.
Soft Liquidation Threshold: When collateral/debt ratio falls below soft threshold, partial liquidations begin with base penalties.
Hard Liquidation Threshold: When ratio falls below hard threshold, complete liquidation is permitted with higher penalties.
Bad Debt Threshold: When debt exceeds collateral value, socialized bad debt handling begins.
OEV Liquidation Queue System
Curvance implements an innovative liquidation queue with Optimal Extractable Value (OEV) capture:
Priority Phase: Winning OEV auction liquidators get immediate liquidation rights.
Fallback Mechanism: If OEV auction fails, liquidations enter a queueing process:
Priority window for liquidators who called
queueLiquidation()
.Regular window for any liquidator.
End window when the queue resets for that position.
Risk Modeling
Curvance enhances risk modeling through:
Asset-Specific Risk Parameters: Each asset has customized collateralization requirements.
Three-Tier Liquidation System: Soft, hard, and bad debt thresholds for graduated liquidation responses.
Volatility-Responsive Liquidations: Aggressive liquidations in volatile periods, gentler in stable periods.
Bad Debt Socialization: When a user's debt exceeds collateral, lenders share any shortfall.
The overall architecture provides a powerful framework for managing diverse asset classes while maintaining protocol solvency and capital efficiency.
Last updated