Single Collateral Factor vs Tiered Collateral Factors

One uniform risk parameter for all assets (e.g., 80% LTV). This creates a predictable, easy-to-audit system. It's ideal for protocols like early Compound, where operational simplicity and user clarity are paramount.

Reduced on-chain computation for borrowing power calculations. This minimizes gas costs for users and simplifies the smart contract logic, reducing attack surface. Best for high-throughput, cost-sensitive environments.

Asset-specific risk parameters (e.g., 85% for ETH, 65% for a volatile altcoin). This allows protocols like Aave to safely list a wider range of assets by precisely managing liquidation risks, protecting the protocol's solvency.

Higher borrowing power for blue-chip assets. Users can borrow more against safer collateral, improving capital efficiency. This attracts larger TVL and can create more competitive lending yields, as seen with MakerDAO's multi-collateral DAI system.

Single risk model for all assets: One collateral factor (e.g., 75%) applies universally, drastically reducing governance overhead and smart contract complexity. This matters for newer protocols like Euler Finance (pre-hack) or Radiant Capital seeking rapid deployment and easy user comprehension. Audits and parameter updates are streamlined.

Forces a lowest-common-denominator approach: Volatile assets like MEME coins must be assigned the same conservative factor as stablecoins like USDC, locking up capital. This matters for maximizing Total Value Locked (TVL) and user borrowing power. Protocols like early Compound faced this limitation before introducing risk tiers for certain assets.

Granular asset classification: Enables high factors for blue-chips (e.g., 80% for wBTC, WETH) and low/zero for volatile assets. This matters for established protocols like Aave V3 and Compound V3, which use multiple tiers to optimize capital efficiency across $10B+ in TVL without proportionally increasing systemic risk.

Increased governance and oracle dependency: Each asset requires individual risk assessment, frequent re-evaluations (e.g., LUNA collapse), and precise price feeds from Chainlink or Pyth. This matters for security budgets and operational load. A misconfigured tier can create isolated but significant bad debt, as seen with MIM on Abracadabra.

Uniform risk assessment: All assets share a single Loan-to-Value (LTV) ratio, like Aave's 80% for major assets. This simplifies audits, governance, and user understanding. It's ideal for protocols prioritizing security and composability, as it reduces attack vectors from complex risk models.

One-size-fits-all limitation: Low-volatility, high-liquidity assets (e.g., stETH, wstETH) are underutilized. They could safely support higher borrowing power but are capped by the blanket LTV. This leaves $B+ in capital idle, reducing protocol revenue and user leverage options compared to tiered systems.

Risk-adjusted efficiency: Protocols like Compound V3 and MakerDAO's Spark use multiple tiers (e.g., 92% LTV for stETH, 75% for UNI). This unlocks 20-30% more borrowing power for premium collateral, directly increasing protocol fee revenue and attracting high-value users seeking efficient leverage.

Increased governance and oracle dependency: Each tier requires precise, ongoing risk parameter management. A mispriced asset in a high-LTV tier can lead to rapid insolvency during volatility. This model demands sophisticated risk oracles (e.g., Chainlink, Pyth) and active DAO oversight, increasing operational overhead.