Home > Topics > Financial Analytics > Introduction to Credit Risk Modeling

Introduction to Credit Risk Modeling

To manage credit risk quantitatively, we need to measure the Expected Loss (EL). The Expected Loss is composed of three key parameters.

The Formula

Expected Loss (EL) = PD * LGD * EAD

1. Probability of Default (PD)

  • Definition: The likelihood (0% to 100%) that the borrower will default within a specific time horizon (usually 1 year).
  • Source: Credit Ratings (AAA has low PD, CCC has high PD) or Statistical Models (Logistic Regression).

2. Loss Given Default (LGD)

  • Definition: If default happens, what percentage of the loan will you lose?
  • Formula: LGD = 1 - Recovery Rate.
  • Example: If a borrower defaults on a 100 Rs loan, and you sell their collateral for 60 Rs, your Recovery Rate is 60%. Your LGD is 40%.

3. Exposure at Default (EAD)

  • Definition: How much money does the borrower owe at the actual moment of default?
  • Relevance: For a Credit Card or Line of Credit, the borrower might max out their limit right before defaulting.
Note

Unexpected Loss (UL): EL is the "cost of doing business" (you expect some defaults). Banks price this into the interest rate. Unexpected Loss (UL) is the volatility of loss (e.g., a massive sudden wave of defaults). Banks hold Capital to cover UL.

Loading quiz…