Key Takeaways
- ROE is calculated as net profit attributable to ordinary shareholders divided by average shareholders' equity for the period.
- A shift of more than 5 percentage points year-on-year without an identified cause should trigger further investigation under ISA 520 .
- Auditors use ROE in analytical procedures to detect misstatements in both the income statement and the equity section of the balance sheet.
- An entity can inflate ROE by reducing equity through aggressive distributions or share buybacks, which masks declining profitability.
What ROE tells the auditor
Plenty of review notes start the same way: a senior flags that "ROE looks reasonable" and moves on to the next workpaper. That's exactly the kind of vague benchmark ISA 520.5 (c) warns against. If the expectation isn't precise enough to catch a misstatement at performance materiality, the analytical procedure hasn't done its job.
ROE connects two financial statements (the statement of profit or loss and the statement of financial position) into a single ratio: net profit attributable to ordinary shareholders divided by average ordinary shareholders' equity. A sudden change signals that profit or equity (or both) moved unexpectedly. ISA 520.5 requires the auditor, when designing substantive analytical procedures, to determine whether a particular ratio is suitable for the assertion being tested. Because ROE spans two statements, it can flag misstatements that a single-statement test might miss.
ISA 520.7 then requires the auditor to investigate differences that exceed the documented threshold by obtaining audit evidence until the fluctuation is adequately explained. In practice, this means the file should tell a story: prior-year data in, expectation out, deviation measured, cause corroborated. Anything less invites a review point.
Worked example: Rossi Alimentari S.p.A.
Rossi Alimentari is an Italian food production company, FY2025, revenue EUR 67M, IFRS reporter. The engagement team is performing substantive analytical procedures on profitability. Rossi's FY2024 net profit was EUR 4.0M. Shareholders' equity was EUR 28.5M at 31 December 2024 and EUR 26.0M at 31 December 2023. For FY2025, Rossi reports net profit of EUR 5.8M. Shareholders' equity at 31 December 2025 is EUR 31.2M.
Step 1 — Calculate prior-year ROE
FY2024 average equity is (EUR 26.0M + EUR 28.5M) / 2 = EUR 27.25M. ROE for FY2024 is EUR 4.0M / EUR 27.25M = 14.7%.
Step 2 — Calculate current-year ROE
FY2025 average equity is (EUR 28.5M + EUR 31.2M) / 2 = EUR 29.85M. ROE for FY2025 is EUR 5.8M / EUR 29.85M = 19.4%.
Step 3 — Set the expectation and threshold
The team expected ROE to remain within 2 percentage points of FY2024 (range: 12.7%–16.7%), based on stable revenue growth of 4% per year and no material changes to the capital structure. The actual ROE of 19.4% exceeds the upper bound by 2.7 percentage points.
Step 4 — Investigate the deviation
The EUR 1.8M increase in net profit is traced to a one-off insurance settlement of EUR 1.5M recognised in other operating income following flood damage at the Parma warehouse in Q1 2025. Stripping the settlement, adjusted net profit is EUR 4.3M, producing an adjusted ROE of 14.4%, which sits within the expected range. The equity increase is consistent with retained earnings accumulation.
Without that insurance settlement, this workpaper would've ended with an open item and a very uncomfortable manager review. Getting the expectation right at the start (built from specific prior-year data, linked to performance materiality) is what turns a 4.7-point deviation into a 30-minute corroboration exercise instead of a late-stage fire drill.
Where teams get it wrong
Teams set ROE expectations using vague benchmarks ("industry average is around 12%") without documenting why that benchmark applies to the specific entity. ISA 520.5 (c) requires the expectation to be sufficiently precise to identify misstatements that, individually or in aggregate, may cause the financial statements to be materially misstated. A generic industry average without adjustment for the entity's capital structure doesn't meet that threshold.
Auditors also sometimes calculate ROE using closing equity rather than average equity. When equity changed materially during the period (through a capital increase or large dividend, for example), closing equity distorts the ratio. The distortion can suppress a genuine fluctuation or create a false alarm, sending the team off ticking and bashing on the wrong balance entirely.
ROE vs. return on assets
| Dimension | ROE | ROA |
|---|---|---|
| What it measures | Profitability relative to shareholders' equity only | Profitability relative to total assets (funded by both equity and debt) |
| Sensitivity to leverage | Highly sensitive: more debt reduces equity, inflating ROE even if profit stays flat | Low sensitivity: total assets absorb both debt-funded and equity-funded resources |
| When the auditor uses it | Detecting misstatements in equity or profit attributable to shareholders; evaluating distributions policy | Detecting misstatements across the full balance sheet; evaluating asset efficiency |
| Limitation | Can be artificially high when equity is thin due to buybacks or accumulated losses | Can be artificially low when asset base is inflated by revaluations or large intangible balances |
An entity with stable ROA but rising ROE is signalling a change in its financing structure, not a change in operating performance. Auditors who track only ROE may conclude that profitability improved when the real movement is in liabilities. Tracking both ratios in parallel (and documenting the relationship) catches leverage-driven distortions that a single ratio misses.
Related terms
Related tools
Related reading
Frequently asked questions
How do I set a threshold for ROE in analytical procedures?
Base the threshold on the precision needed to detect a misstatement at performance materiality. If performance materiality is EUR 400,000 and average equity is EUR 30M, a 1.3-percentage-point change in ROE equates to that amount. ISA 520.5(c) requires documentation of why the chosen threshold produces the needed precision for the assertion being tested.
Does ROE apply to group audits?
Yes. At the group level, calculate ROE using consolidated equity attributable to shareholders of the parent, excluding non-controlling interests. At the component level, ROE calculated on the subsidiary's standalone equity can signal entity-specific profitability issues that the group-level ratio masks. ISA 600.31 requires the group engagement team to evaluate the appropriateness of analytical procedures applied at both levels.
What happens when equity is negative?
ROE becomes mathematically meaningless when equity is negative, because dividing a profit by a negative denominator produces a negative ratio that suggests poor performance even when the entity is profitable. If the entity has negative equity (for instance, from accumulated losses exceeding share capital), ISA 520.5(a) requires the auditor to determine whether the analytical procedure remains suitable. In this case, switch to an alternative ratio such as return on assets.