Credit Risk

The Probability of Default (PD) is the estimated likelihood that a borrower will fail to repay a loan within a specific time horizon. It is a forward-looking metric, typically expressed as a percentage or a credit rating.

• Core Input: A primary component for calculating Expected Loss.
• Modeling: Derived from historical data, financial ratios, and macroeconomic factors.
• Example: A corporate bond with a 2% annual PD has a 1 in 50 chance of defaulting in the next year.

Loss Given Default (LGD) quantifies the portion of the exposure that will be lost if a default occurs, accounting for recovery from collateral or bankruptcy proceedings. It is expressed as a percentage of the exposure at default (EAD).

• RecoRate: LGD = 1 - Recovery Rate.
• Collateral: Secured loans have lower LGD than unsecured loans.
• Example: A $1M loan with a 40% recovery rate has an LGD of 60%, implying a $600,000 loss.

Exposure at Default (EAD) is the total value a lender is exposed to when a borrower defaults. For revolving credit (e.g., credit lines), this includes both drawn amounts and a estimate of future drawdowns before default.

• Credit Conversion Factor (CCF): Used to estimate the undrawn portion that may be utilized.
• Time Horizon: Snapshot of exposure at the moment of default.
• Example: A borrower with a $100k credit line who has drawn $70k and has a 50% CCF would have an EAD of $70k + (30k * 0.5) = $85k.

Expected Loss (EL) is the anticipated average loss over a given period, calculated as the product of its three core components. It represents a cost of doing business that should be covered by provisions or pricing.

• Formula: EL = PD * LGD * EAD.
• Provisioning: Banks set aside capital based on EL calculations.
• Example: For a loan with PD=2%, LGD=60%, EAD=$1M: EL = 0.02 * 0.60 * $1,000,000 = $12,000.

Credit Migration Risk (or downgrade risk) is the risk that a borrower's creditworthiness deteriorates, leading to a lower credit rating, even if default does not occur. This impacts the mark-to-market value of debt instruments.

• Spread Widening: A downgrade typically causes the credit spread to increase, reducing the bond's price.
• Regulatory Capital: Affects risk-weighted asset calculations for banks.
• Example: A corporate bond downgraded from 'A' to 'BBB' will lose value as investors demand a higher yield for the increased risk.

Concentration Risk arises from excessive exposure to a single borrower, industry, geographic region, or asset class. It violates the principle of diversification, amplifying potential losses from a single adverse event.

• Name Concentration: Large exposure to one counterparty.
• Sector Concentration: Overexposure to a cyclical industry (e.g., real estate).
• Mitigation: Managed through exposure limits, securitization, and credit derivatives like CDS.

Derivative protocols that allow users to hedge against or speculate on credit events like a borrower's default. Key concepts:

• Credit Default Swap (CDS): A buyer pays a premium to a seller for protection against a default.
• Cover protocols (e.g., Nexus Mutual): Provide discretionary coverage for smart contract failure, which is a form of counterparty risk. These instruments create a market for pricing and transferring credit risk.

The integration of traditional credit assessment with blockchain for Real-World Asset (RWA) tokenization. This involves:

• Off-chain legal enforcement paired with on-chain settlement.
• Asset-backed securities (e.g., tokenized mortgages, treasury bills).
• KYC/AML compliant pools that restrict access to verified participants. These frameworks manage credit risk through hybrid legal-technological structures, connecting DeFi with regulated finance.

The risk that the other party in a financial contract will default. In DeFi, this is often decentralized but not eliminated. Key examples include:

• Lending Protocols: A borrower failing to repay an undercollateralized loan.
• Derivatives & Options: A counterparty failing to deliver on a futures contract or option payout.
• Cross-Chain Bridges: The validating entity or custodian on the other side failing to release funds.

The risk that the assets locked as collateral lose value or become illiquid, triggering undercollateralization and potential liquidation losses. Factors include:

• Volatility: A sharp price drop can make a loan undercollateralized before liquidation executes.
• Liquidity: Illiquid collateral cannot be sold during liquidation, causing bad debt.
• Oracle Failure: Incorrect price feeds can show false collateral values, preventing timely liquidations.

The risk that a lending or money market protocol becomes insolvent, meaning its liabilities (user deposits) exceed its assets (loans + collateral). This creates bad debt that is socialized or covered by insurance reserves. Causes are:

• Mass liquidations failing during market crashes.
• Oracle manipulation leading to undercollateralized borrowing.
• Smart contract exploits that drain protocol reserves.

The automated process of selling a borrower's collateral to repay their debt, which itself carries execution risks.

• Liquidation Incentives: Liquidators must be incentivized with a discount (liquidation bonus) to participate.
• Slippage & MEV: During high volatility, liquidations can cause high slippage and be front-run by MEV bots, reducing recovered funds.
• Auction Design: Poorly designed Dutch or English auctions can fail to attract liquidators, increasing bad debt.

Methods used to manage and reduce credit risk exposure.

• Overcollateralization: Requiring collateral value greater than the loan value (e.g., 150% LTV).
• Risk Parameters: Protocols set Loan-to-Value (LTV) ratios, liquidation thresholds, and reserve factors.
• Insurance & Coverage: Using protocols like Nexus Mutual or Unslashed Finance to hedge against smart contract failure or custodial risk.
• Credit Delegation: Allowing trusted entities to borrow against a delegator's collateral, shifting risk assessment.

The risk that the failure of one protocol or large participant triggers cascading failures across DeFi. This amplifies credit risk.

• Composability: Protocols are built on top of each other; a failure in a lending primitive can impact derivative and yield aggregator protocols.
• Collateral Rehypothecation: The same asset (e.g., stETH) used as collateral across multiple protocols can create a single point of failure.
• Contagion: A major liquidation event on one platform can cause market-wide price impacts, affecting collateral values everywhere.

The most common application of credit risk analysis. Lenders use credit scores (like FICO) to assess an individual's likelihood of default based on their credit history, debt-to-income ratio, and payment behavior. This determines:

• Loan approval and credit limits
• The interest rate (APR) offered
• Risk-based pricing, where higher-risk borrowers pay more

Example: A borrower with a 720 FICO score may receive a 5% APR on a mortgage, while a borrower with a 620 score might be offered the same loan at 7.5% to compensate for the higher perceived risk.

In capital markets, credit risk is directly priced into corporate bonds. The difference in yield between a corporate bond and a risk-free Treasury bond (the credit spread) reflects the market's assessment of default risk.

• Investment-grade bonds (BBB- or higher) have narrow spreads.
• High-yield (junk) bonds have wide spreads, offering higher returns to compensate for higher default probability.

Rating agencies like Moody's and S&P provide credit ratings (e.g., AAA, Baa2, CCC) that formalize this risk assessment for investors.

Decentralized Finance (DeFi) protocols manage credit risk without credit scores by enforcing overcollateralization. A user must deposit crypto assets worth more than the loan they take out.

• On MakerDAO, to mint 1000 DAI stablecoins, a user might need to lock $1500 worth of ETH as collateral (150% collateralization ratio).
• If the value of the collateral falls too close to the loan value, the position is liquidated to repay the debt.

This mechanism eliminates counterparty risk for lenders but introduces liquidation risk for borrowers.

Emerging DeFi protocols are experimenting with under-collateralized or credit-based lending, which introduces traditional credit risk models on-chain.

• Protocols like Goldfinch and Maple Finance assess the creditworthiness of institutional borrowers (Pool Delegates) off-chain.
• Lenders deposit funds into pools that are allocated to these vetted borrowers, accepting the default risk in exchange for higher yields.
• This enables real-world asset (RWA) financing, such as lending to fintech companies or crypto-native institutions.

A derivative instrument used to transfer and hedge credit risk. The buyer of a CDS makes periodic payments to the seller and, in return, receives a payoff if a credit event (like default) occurs on a reference entity (e.g., a company or country).

• Acts as insurance against default.
• The CDS spread (cost of the swap) is a direct market indicator of perceived credit risk.

Notoriously, the widespread use and mispricing of CDS on mortgage-backed securities was a central cause of the 2008 financial crisis.

Credit risk exists even at the highest levels of finance. When commercial banks deposit reserves with a central bank, they are exposed to the sovereign credit risk of that central bank's government.

• This risk is typically considered negligible for major economies.
• However, it becomes apparent during sovereign debt crises or with central banks in unstable regimes.

Similarly, participants in Real-Time Gross Settlement (RTGS) systems face intraday credit risk to the central bank, which is often mitigated by collateral requirements.

The Collateralization Ratio is a core risk metric representing the value of collateral relative to the loan value. It is calculated as (Collateral Value / Loan Value) * 100%. A higher ratio indicates a safer loan.

• Overcollateralization: CR > 100%. Standard in DeFi to protect against price volatility.
• Liquidation Threshold: The minimum CR before a position is automatically liquidated.
• Example: A $150 ETH deposit backing a $100 DAI loan has a 150% CR.

Liquidation is the enforced closure of an undercollateralized loan position by the protocol to repay lenders. It is triggered automatically when the collateral value falls below a predefined liquidation threshold.

• Process: The protocol sells a portion of the borrower's collateral, often at a discount, to cover the debt plus a liquidation penalty.
• Purpose: Protects lenders from losses by ensuring loans remain overcollateralized.
• Liquidators: Third-party actors who execute liquidations for a profit incentive.

An On-Chain Credit Score is a quantitative assessment of a wallet address's creditworthiness derived from its blockchain transaction history. Unlike traditional scores, it is based on transparent, immutable on-chain data.

• Data Points: Includes repayment history, collateral management, wallet age, transaction volume, and protocol interactions.
• Use Case: Enables under-collateralized or uncollateralized lending by assessing borrower risk.
• Providers: Projects like Cred Protocol and Spectral Finance generate these non-transferable reputation scores.

Default occurs when a borrower fails to meet their legal obligations to repay a loan. In overcollateralized DeFi, technical defaults are rare due to automated liquidations. The concept is more relevant to undercollateralized lending models.

• In DeFi: Often synonymous with a position being liquidated. The lender is made whole via collateral sale.
• In TradFi/Institutional DeFi: Reflects a failure to repay, leading to loss for the lender and potential legal recourse.
• Loss Given Default (LGD): The amount lost if default occurs, a key metric in risk modeling.

Probability of Default (PD) is a statistical estimate of the likelihood that a borrower will default on their debt obligations within a specific time horizon. It is a fundamental input for credit risk models.

• Calculation: Modeled using historical data, credit scores, market conditions, and collateral volatility.
• Use in DeFi: Drives risk-based loan-to-value (LTV) ratios, interest rates, and capital requirements for lending pools.
• Expected Loss: Calculated as PD * Loss Given Default (LGD) * Exposure at Default (EAD).

A Debt Position (or CDP - Collateralized Debt Position) is a user's specific loan instance within a lending protocol. It is a smart contract vault that holds collateral and manages the associated debt.

• Components: Contains the locked collateral amount, the borrowed asset amount, and the current health factor/collateral ratio.
• Management: Users can add/remove collateral or repay debt to manage their position's health.
• Examples: Opening a Vault in MakerDAO to mint DAI, or supplying collateral to Aave to borrow another asset.