Materiality, Performance Materiality and Tolerable Misstatement Are Not Interchangeable

Three numbers are set before an audit starts, and candidates routinely treat them as one number with three names. They are not. Overall materiality is a judgement about what would change a reader's decision. Performance materiality is a smaller amount chosen so that the errors nobody finds do not add up to that judgement. Tolerable misstatement is smaller still, and it is not a judgement about readers at all: it is the input to a sample size. Confuse the first with the third and the audit is planned to the wrong number; confuse the second with the first and every sample is too small.

AUD examines the distinction directly, usually as a question about which amount is used for what, and indirectly, through every sampling and evaluation question in Areas II and III. The standards involved are AU-C 320 on materiality, AU-C 450 on evaluating misstatements and AU-C 530 on sampling; their ISA equivalents, 320, 450 and 530, are the ones SOCPA endorses, and the definitions are the same.

The three, defined

Term	Definition	Purpose	Set by
Materiality for the financial statements as a whole	The magnitude of misstatement that could reasonably be expected to influence the economic decisions of users taken on the basis of the statements	The threshold for the opinion: uncorrected misstatements above it mean the statements are materially misstated	Auditor's judgement about users, usually a percentage of a benchmark
Performance materiality	An amount set by the auditor at less than overall materiality, to reduce to an appropriately low level the probability that the aggregate of uncorrected and undetected misstatements exceeds materiality	The threshold for planning procedures and deciding which balances to test	A percentage of overall materiality, reflecting the expected level of misstatement
Tolerable misstatement	The application of performance materiality to a particular sampling procedure; the maximum misstatement in a population the auditor is willing to accept	An input to sample size and to the evaluation of a sample's result	Equal to or less than performance materiality, for each population sampled

A fourth amount sits underneath the three. AU-C 450 lets the auditor set a clearly trivial threshold below which misstatements need not even be accumulated, on the basis that they could not be material individually or in aggregate. It is a small fraction of overall materiality and a fact pattern that gives it is testing whether you know it is not tolerable misstatement.

The order they are set in

The sequence runs from the reader inward, and each amount constrains the next.

1. Overall materiality is set at planning, from a benchmark. The standards prescribe no percentages, but firm methodologies converge on a few: around 5% of pre-tax income for a profit-oriented entity, 0.5 to 1% of revenue or total assets where income is volatile or the entity is not-for-profit, around 1% of equity for a fund. The choice of benchmark is itself a judgement about what the users of these particular statements care about.
2. Specific materiality, where needed, is set lower for particular classes of transactions, balances or disclosures that users are sensitive to below the overall level: related party transactions, directors' remuneration, a regulatory capital figure.
3. Performance materiality is set below overall materiality, commonly at 50 to 75% of it. The lower end is used where prior audits found many misstatements or the entity's controls are weak; the upper end where the population is clean and controls are strong.
4. Tolerable misstatement is set for each population that will be sampled, at or below performance materiality, and feeds the sample size under AU-C 530.
5. The clearly trivial threshold is set last, and low.

All of them are revised if the audit turns up information that would have changed them at planning: a fall in pre-tax income, a discovered fraud, a different user base. AU-C 320 requires the auditor to reconsider, and a revision downward means procedures already performed may be insufficient.

The worked numbers

A trading company in Riyadh has pre-tax income of 20,000,000 and the auditor takes 5% as the benchmark percentage.

Amount	Basis	Figure
Overall materiality	5% of 20,000,000	1,000,000
Performance materiality	75%, controls assessed as strong	750,000
Tolerable misstatement, receivables sample	Below performance materiality, given the population's risk	600,000
Clearly trivial threshold	5% of overall materiality	50,000

Three things follow. A receivable balance of 700,000 is above performance materiality and is selected for testing regardless of any sample. The receivables sample is sized so that the auditor can conclude, at the chosen confidence level, that the population is not misstated by more than 600,000. A misstatement of 30,000 found in a supplier invoice is noted but not accumulated, because it is below 50,000. And at the end of the audit, the aggregate of everything accumulated and uncorrected is compared with 1,000,000, not with 750,000: performance materiality is a planning cushion, and the opinion is formed against overall materiality.

Why the cushion exists

Performance materiality is the concept candidates find least intuitive, because it seems to make the audit stricter than the opinion requires. The reasoning is probabilistic. An auditor who plans every procedure to detect misstatements of 1,000,000 will detect most of them and miss some, and will also fail to detect some smaller misstatements in balances that were not tested at all. If the ones found and left uncorrected total 800,000, and the undetected ones total 300,000, the statements are materially misstated and the opinion is wrong. Planning to a lower number leaves room for what the audit did not find. That is all performance materiality is: an allowance for the undetected, set before the audit so the sample sizes reflect it.

Tolerable misstatement takes the cushion one step further, into a single population. A receivables balance, an inventory count, a set of payables: each gets a tolerable misstatement that says how much error in that population the auditor could accept and still be within the overall plan. Because several populations are sampled and their projected misstatements add up, tolerable misstatement for any one of them is often lower than performance materiality, and never higher.

Evaluation: the same three at the end

AU-C 450 requires the auditor to accumulate misstatements, other than clearly trivial ones, and to distinguish three kinds: factual misstatements, about which there is no doubt; judgemental misstatements, arising from management's estimates or policy choices that the auditor considers unreasonable; and projected misstatements, the auditor's best estimate of misstatement in a population from a sample. The projected ones are where tolerable misstatement returns: a sample's projected misstatement, plus an allowance for sampling risk, is compared with the tolerable misstatement for that population to decide whether the population is acceptable.

The aggregate, after management corrects what it will, is compared with overall materiality. Qualitative factors can make a quantitatively small misstatement material: one that turns a loss into a profit, breaches a loan covenant, affects management compensation or masks a change in trend. A question that gives an uncorrected misstatement below the number and asks whether the statements are fairly presented is usually asking about one of those.

The vocabulary across standards

The three frameworks a candidate meets use slightly different words for the same ideas, and AUD occasionally tests the mapping.

Concept	AU-C (AICPA)	ISA (IAASB, endorsed by SOCPA)	PCAOB
Threshold for the opinion	Materiality for the financial statements as a whole	Same	Materiality
Planning cushion	Performance materiality	Performance materiality	Tolerable misstatement, used in AS 2105 in the sense AU-C gives performance materiality
Sampling parameter	Tolerable misstatement	Tolerable misstatement	Tolerable misstatement
Below the accumulation line	Clearly trivial	Clearly trivial	Clearly trivial

The PCAOB row is the trap. AS 2105 does not use the term performance materiality; it uses tolerable misstatement for the amount set below materiality at the account or disclosure level, which is what AU-C calls performance materiality. A question set in an issuer audit that says "tolerable misstatement" may mean either thing, and the surrounding sentence decides.

How AUD asks

Question pattern	Answer
"Which amount is used to determine which balances to test"	Performance materiality
"Which amount is used to determine a sample size"	Tolerable misstatement
"Which amount are uncorrected misstatements compared with at the end"	Overall materiality, with qualitative factors
Pre-tax income falls sharply mid-audit	Revise materiality downward; reconsider the sufficiency of procedures performed
Weak controls, many prior-year misstatements	Performance materiality set at the lower end of the range
A misstatement below the clearly trivial threshold	Not accumulated
Small misstatement that flips a loss to a profit	Material by nature, whatever the number

Where to practise this

The distinction is learned by answering fifty questions that each hinge on one of the three words, until "sample" triggers tolerable misstatement and "opinion" triggers overall materiality without thought. Prepito's AUD course covers materiality, evaluation of misstatements and sampling with practice questions whose explanations name the amount at stake in every option. The SOCPA Audit course covers ISA 320, 450 and 530 as endorsed. For where these fit in the section, see the AUD study guide; for the model the numbers serve, The Audit Risk Model and Internal Control; and for what tolerable misstatement does inside a sample, Audit Sampling: Attribute vs Variables.

---

Sources: AICPA, Statements on Auditing Standards currently effective (AU-C 320, 450 and 530), PCAOB AS 2105, Consideration of Materiality in Planning and Performing an Audit, IFAC profile of Saudi Arabia on SOCPA's adoption of the ISAs. The percentages quoted are common practice, not requirements of any standard; standards are revised periodically, so confirm the current text for a specific engagement.