Probability of Default

What Is Probability of Default?

Probability of default, or PD, is the estimated likelihood that a borrower, issuer, or counterparty will default over a specified time horizon. It is a core credit-risk measure used by banks, bond investors, rating models, lenders, and risk managers.

PD does not measure how much money will be lost if default occurs. It measures the chance of default. Loss severity is captured separately through loss given default, exposure at default, recovery assumptions, and collateral analysis.

Basic Credit-Risk Framework

Expected credit loss is often described using probability of default, loss given default, and exposure at default.

PD is the chance of default. LGD is the share of exposure expected to be lost if default occurs. EAD is the exposure amount at default. A loan can have a low PD but high loss severity, or a high PD but strong collateral recovery.

How PD Is Estimated

For consumer credit, PD may be estimated using credit scores, delinquency history, income, debt burden, loan type, and macroeconomic variables. For companies, models may use leverage, interest coverage, profitability, liquidity, industry risk, market spreads, ratings, and default history. Banks using internal ratings-based approaches estimate PD within regulatory frameworks.

PD should always be tied to a time horizon. A 1% one-year default probability is not the same as a 1% lifetime default probability. The horizon matters for pricing, reserves, capital, and risk appetite.

PD Versus Credit Rating

A credit rating is a rating agency's opinion of credit risk. Probability of default is a quantitative estimate. Ratings may imply broad default-risk ranges, but they are not the same thing as a model's exact PD. Two borrowers with the same rating may still have different model-based default probabilities.

Market prices can also imply default risk. A widening credit spread may suggest investors are demanding more compensation for default and liquidity risk, though spreads include more than pure PD.

Where PD Changes Decisions

PD affects loan approvals, credit limits, interest rates, bond yields, allowance for credit losses, stress testing, and capital requirements. A lender charging the same rate to borrowers with very different PDs may be underpricing risk. A bond investor ignoring PD may mistake a high yield for a bargain when it is compensation for real default risk.

For businesses, rising PD can increase borrowing costs and reduce access to capital before an actual default occurs.

Simple Lending Example

Assume a lender has a $1,000,000 exposure, estimates a 2% one-year probability of default, and expects to lose 40% of the exposure if default occurs. The simplified expected loss is $8,000: 2% multiplied by 40% multiplied by $1,000,000. That figure is not a prediction that exactly $8,000 will be lost. It is an average loss estimate across many similar exposures or scenarios.

This is why PD belongs in a broader credit framework. A lender still needs underwriting judgment, collateral analysis, covenants, portfolio concentration limits, and stress tests. The same borrower can also have different PD estimates under base-case and stressed scenarios.

The Bottom Line

Probability of default estimates the chance that a borrower or issuer will default over a defined period. It is useful only when read with loss severity, exposure, collateral, time horizon, and the economic environment behind the estimate.