Default Rate

Default rate measures defaults within a defined loan population and period, using account counts, exposure amounts, or a point-in-time defaulted balance.

The default rate measures how much of a defined credit population enters or remains in default over a stated period. A useful default rate must identify the default trigger, population, time horizon, unit of measurement, and denominator; otherwise, two rates with the same percentage may describe different credit behavior.

The rate can be based on the number of accounts, the amount of exposure, or the balance already in default at a reporting date. These versions are not interchangeable.

Key Takeaways

  • A cohort default rate measures new defaults among accounts or exposure initially at risk during a period.
  • A point-in-time defaulted-balance rate measures the stock of defaulted exposure at one date.
  • Count-weighted and exposure-weighted rates can diverge when larger loans behave differently from smaller loans.
  • A default rate is not automatically a probability of default, delinquency rate, or charge-off rate.
  • The result depends on default, cure, re-default, sale, write-off, and denominator rules.
  • Annualizing a short-period rate can be misleading when defaults are seasonal or correlated with economic stress.

Default Rate Formulas

For account-count incidence over a defined horizon:

$$ \text{Count Default Rate} = \frac{\text{Accounts entering default during the period}}{\text{Eligible accounts at the start of the period}} $$

For exposure-weighted incidence:

$$ \text{Exposure Default Rate} = \frac{\text{Exposure entering default during the period}}{\text{Eligible exposure at the start of the period}} $$

For a point-in-time stock measure:

$$ \text{Defaulted-Balance Rate} = \frac{\text{Defaulted balance at the reporting date}}{\text{Relevant balance at the reporting date}} $$

The numerator and denominator should use the same scope. For example, a numerator based on committed exposure should not be divided by a denominator based only on drawn principal unless the methodology expressly requires it.

Incidence, Prevalence, Count, and Exposure

MeasureNumeratorTypical useMain limitation
Count incidence rateNumber of accounts newly defaultingConsumer vintages and behavior analysisGives a $1,000 loan the same weight as a $100,000 loan
Exposure incidence rateExposure associated with new defaultsCredit-cost and concentration analysisSensitive to exposure-at-default and draw assumptions
Point-in-time defaulted-balance rateBalance currently classified in defaultPortfolio status at a reporting dateAffected by cures, sales, charge-offs, and time spent in default
Cumulative vintage default rateDefaults accumulated since originationComparing origination cohortsNewer vintages have less time to season

A report should say which measure it uses rather than label all four simply as default rate.

Worked Example: Loan Cohort

Assume a lender begins the year with 2,000 eligible loans totaling $40 million. During the year:

  • 50 loans enter default.
  • Their exposure at default totals $1.8 million.
  • At year-end, $1.1 million remains classified as defaulted.
  • The total year-end portfolio balance is $39 million.

The count incidence rate is:

$$ \frac{50}{2{,}000} = 2.5\% $$

The exposure incidence rate is:

$$ \frac{\$1.8\text{ million}}{\$40\text{ million}} = 4.5\% $$

The point-in-time defaulted-balance rate is:

$$ \frac{\$1.1\text{ million}}{\$39\text{ million}} \approx 2.82\% $$

The 4.5% exposure rate exceeds the 2.5% count rate because the defaulting loans were larger than average. The 2.82% stock rate is lower than the exposure incidence rate because some defaulted exposure cured, was repaid, was sold, or was charged off before year-end. The three results answer different questions.

Default Rate vs. Probability of Default

An observed default rate is a historical outcome for a defined sample and period. Probability of default (PD) is a forward-looking estimate assigned to a borrower, grade, segment, or exposure. Historical default rates often help calibrate or validate PD models, but the two should not be treated as identical.

Differences can arise from:

  • point-in-time versus through-the-cycle calibration;
  • one-year versus lifetime horizons;
  • borrower-level versus facility-level default;
  • portfolio mix and underwriting changes;
  • exclusions, data gaps, cures, and re-default treatment;
  • seasoning and economic conditions.

Default Rate vs. Other Portfolio Rates

MetricEvent or status measuredPosition in the credit cycle
Delinquency RatePast-due accounts or balancesEarly to intermediate stress
Default RateNew or existing defaults under a stated definitionSerious credit deterioration
Charge-Off RateAmounts recognized as uncollectible, often net of recoveriesRealized loss recognition
Recovery RateValue recovered after default relative to a defined baseWorkout outcome

A high default rate does not necessarily produce the same charge-off rate. Secured loans may default frequently but recover well, while unsecured defaults may produce more severe losses.

Time Horizon and Annualization

State whether the rate covers a month, quarter, year, or life of a vintage. Multiplying a one-month rate by 12 assumes a pattern that may not exist. Defaults can cluster by season, loan age, payment resets, underwriting vintage, or recession. A cumulative lifetime rate also cannot be compared directly with a one-year rate.

For a portfolio with frequent originations and repayments, analysts may use average balances, beginning balances, person-time, or survival methods. Each denominator answers a different question. The method should remain stable or be reconciled when it changes.

How to Evaluate a Reported Default Rate

  1. Read the exact default definition, including days-past-due and unlikely-to-pay triggers.
  2. Confirm whether default is recorded at the borrower, facility, or account level.
  3. Identify the numerator date: first default, any default, or defaulted at period-end.
  4. Match the denominator to the eligible population and exposure basis.
  5. Check treatment of originations, repayments, cures, modifications, sales, and charge-offs.
  6. Segment by product, vintage, risk grade, geography, collateral, and borrower type.
  7. Compare observed rates with prior periods, forecasts, and underwriting changes.
  8. Avoid ranking lenders until definitions and portfolio mix are comparable.

Common Mistakes

  • Dividing defaulted balances by account counts or otherwise mixing units.
  • Using all current loans in the denominator when only a starting cohort is eligible for the numerator.
  • Counting the same loan again after every delinquency episode without a re-default rule.
  • Comparing a cumulative vintage rate with an annual portfolio rate.
  • Treating a point-in-time stock rate as the period’s new-default incidence.
  • Annualizing a short period through simple multiplication without testing seasonality.
  • Assuming the rate measures loss severity or cash loss.
  • Ignoring large-loan concentration hidden by a count-weighted average.

Risks and Limitations

Default rates are backward-looking observations unless embedded in a forecasting model. They can be distorted by policy changes, servicing practices, restructurings, loan sales, portfolio growth, seasoning, and incomplete recovery periods. Sparse defaults create volatile rates, while broad portfolio averages can conceal severe stress in one segment.

This page is educational and is not accounting, regulatory, lending, model-validation, investment, or personalized financial advice.

Authoritative Sources

FAQs

What is a good default rate?

There is no universal good rate. Interpretation depends on product, borrower risk, collateral, pricing, vintage, economic conditions, and the exact numerator and denominator.

Should default rate use loan count or loan balance?

Use the version suited to the question and label it. Counts describe how many accounts default; exposure weighting better reflects the amount at risk.

Can a default rate fall while credit losses rise?

Yes. Fewer but larger defaults, weaker recoveries, or higher loss severity can increase realized losses even when the count default rate declines.

Is a quarterly default rate multiplied by four an annual rate?

Only under restrictive assumptions. A rolling 12-month or cohort calculation is generally more informative when defaults are seasonal or portfolio composition changes.
Browse Credit and Lending