LGD Calculator – Loss Given Default

How to use and interpret loss given default

Loss given default, usually abbreviated LGD, estimates the portion of a credit exposure that is not recovered after a borrower defaults. This calculator converts a recovery assumption into both a percentage loss and a dollar loss. It is useful for loan analysis, credit portfolio monitoring, allowance modeling, stress testing, and comparing the severity of different secured or unsecured exposures. LGD does not estimate whether default will occur; it estimates the financial impact conditional on default.

What each input means

Rate entered lets you choose the form of the assumption you already have. Select Recovery rate when your estimate describes the share of exposure expected to be collected through collateral, restructuring, guarantees, or bankruptcy proceeds. Select Loss severity when your estimate directly states the share expected to be lost. Switching between the two converts the current rate because recovery rate and loss severity are complements.

Recovery rate or loss severity is required and must be between 0% and 100%. A higher recovery rate lowers LGD dollar-for-dollar as a percentage of exposure. A higher loss severity raises LGD. For example, an 80% recovery rate equals 20% loss severity. Common mistakes include entering 0.80 when the field is intended to mean 0.80%, using a recovery estimate before liquidation costs, or applying a rate from a different collateral class without adjustment.

Expected exposure at default is the amount expected to be outstanding when default happens. For a term loan, it may resemble projected principal plus accrued amounts. For revolving credit, it can include expected additional drawings before default. The input is required and cannot be negative. Increasing exposure increases recovery and LGD amounts proportionally but does not change the recovery or loss percentages.

How the model works

The model assumes the entered rate already incorporates the expected timing and net value of recoveries. In institutional credit models, recovery cash flows may be discounted and reduced for collection, legal, servicing, and asset-disposal costs. The Basel Framework discusses LGD within credit-risk capital rules, while the IFRS 9 overview places credit-loss measurement in a broader accounting context.

How to read the results

Loss given default is the primary dollar result. A zero value means the assumptions imply full recovery; a value equal to exposure means no recovery. A negative LGD is not permitted because neither rate nor exposure may be negative. A high LGD can indicate weak collateral coverage, costly enforcement, junior priority, or low expected resale proceeds, but it does not by itself indicate a high probability of default.

Loss severity is LGD expressed as a percentage of exposure. It makes exposures of different sizes easier to compare. Recovery amount is the expected portion collected after default. Recovery rate expresses that recovery as a percentage. Exposure check confirms that recovery amount plus LGD equals total exposure, which is an important cross-foot for detecting inconsistent assumptions.

The donut chart uses the same model values as the result cards and shows the relative shares recovered and lost. The sensitivity table changes only the recovery-rate assumption. Moving down the table toward higher recovery reduces loss severity and LGD. The change-versus-baseline column isolates the dollar impact of each nearby assumption, making it easier to see how much model risk is concentrated in the recovery estimate.

Practical assumptions and common tradeoffs

• Use exposure and recovery assumptions from the same valuation date and scenario. Mixing a stressed exposure with an unstressed recovery rate can distort the result.
• Consider lien priority, collateral type, jurisdiction, workout duration, guarantor strength, and direct recovery costs. Secured senior claims may recover differently from unsecured or subordinated claims.
• Do not confuse LGD with expected credit loss. A simplified expected-loss framework also requires probability of default and, often, the timing of exposure and recoveries.
• Stress the recovery rate rather than relying on a single point estimate. The sensitivity table provides a compact first step, but a full model may use separate economic, collateral-price, and cure-rate scenarios.

For governance, document how assumptions were selected, how often they are refreshed, and which realized recoveries are used for back-testing. The Federal Reserve's model risk management guidance emphasizes validation, controls, and ongoing monitoring. The FDIC credit-risk manual provides additional supervisory context for credit administration. This calculator is an educational estimation tool and not personalized accounting, regulatory, legal, or investment advice.