What it means
Every decision to lend, invest or insure carries some chance that things go wrong. POF rates put a number on that chance, so that decision-makers can compare different risks on the same scale.
A 1% rate suggests that about one in a hundred similar cases is expected to fail over the stated period. The rates are usually built from history.
An analyst counts how many companies, loans or machines in a group failed over a given time, divides by the number in the group and adjusts for current conditions. Credit teams also model rates from financial ratios, payment behaviour and industry trends.
POF rates feed directly into pricing and provisioning. A lender charges a higher interest rate for a group with a 5% rate than for one with a 0.5% rate, and it sets aside more money to cover expected losses.
Insurers, investors and project sponsors use the same logic. The definition of failure must be clear.
It might mean missing a payment by 90 days, entering insolvency, a project overrunning its budget by a set margin or equipment stopping working. The same group can show very different rates under different definitions, so the definition should always be stated alongside the number.
POF rates are estimates, not guarantees. They depend on the quality of past data, they can change when the economy turns and they describe groups rather than individuals.
A borrower in a 2% group either fails or does not, so the rate is best used for portfolios and policy decisions. Segmenting the data usually improves accuracy.
A single average rate hides the fact that new businesses, highly indebted firms and cyclical industries often fail at very different rates from established, cash-rich ones. Splitting the population into sensible groups, and checking that each group has enough cases to be meaningful, produces rates that are far more useful for pricing.
In practice
Real-world examples.
Example
A bank sorts its business customers into risk grades and finds that the weakest grade has a POF rate of 6% a year. It prices loans in that grade at a higher interest rate and requires collateral.
Example
A manufacturer records that one type of machine has failed in 3% of cases within its first year. It uses the rate to decide how many spare units to hold and how much to budget for warranty repairs.
Example
A venture investor reviews a group of early-stage start-ups and estimates that 40% will fail within five years. She sizes each investment on the assumption that a good share of the money will be lost.
Formula
Calculation
POF rate = Number of failures / Number of cases observed x 100%
Expected loss = Exposure x POF rate x Loss given failure (the share of the exposure actually lost when a failure occurs)
A lender reviews 2,000 small-business loans over one year and 50 of them fail. POF rate = 50 / 2,000 = 0.025, or 2.5%.
The portfolio exposure is $20,000,000, and on average 60% of the money is lost when a loan fails.
Expected loss = $20,000,000 x 0.025 x 0.60 = $300,000.Case study
Seen in the real world.
Northgate Lending is a fictional finance company that offered equipment loans to small contractors. Its early pricing used a single flat rate because the team had no view on failure rates by customer type.
After a year, the credit analyst grouped loans by contractor size and sector and found that sole traders had a POF rate of about 7% while larger firms were nearer 1.5%. In this illustrative case, the company moved to risk-based pricing and cut its annual credit losses by roughly a third, even though it declined a few of the riskiest applicants.
The analyst also warned that the groups with only a handful of loans could not support firm conclusions, so those were pooled with a neighbouring group until more data arrived. Each quarter she compared the predicted number of failures with the actual number, and the gap stayed within a few loans. That backtest gave the board enough confidence to rely on the rates for next year's budget.
Watch out
Common mistakes.
- Quoting a POF rate without saying what counts as failure and over what period.
- Applying a group rate to a single borrower as if it were a certainty about that person.
- Using last year's rates in a changed economy without adjusting them.
Questions
People also ask.
Is a POF rate the same as a probability of default?
They are closely related, but POF can cover more than loans, including projects and equipment, while probability of default refers specifically to missing debt payments.
How often should POF rates be updated?
At least annually, and sooner when the economy, the industry or the lender's own customers change in a meaningful way.
What is a good POF rate?
There is no universal answer, because the right level depends on the pricing, the loss if failure occurs and the lender's appetite for risk.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related