Where the diluted EPS calculation breaks
Open any working paper file for a listed IFRS reporter with share options and convertible debt outstanding. Odds are good that the diluted earnings per share (DEPS) number is correct. Odds are equally good that the sequencing test is missing. The team dumped every potential ordinary share into the denominator at once, got a number that looked reasonable, and moved on. In our experience, at least a third of the DEPS files we review during quality monitoring skip the IAS 33.41 -44 ranking step entirely.
That shortcut works most of the time. It fails when a single anti-dilutive instrument hides among several dilutive ones, because including it in the pool drags the denominator up without a proportional numerator adjustment, and the resulting DEPS is higher than it should be. The investor reads a number that understates dilution. Nobody catches it because the figure passes the ticking and bashing. It reconciles to the spreadsheet. The spreadsheet is just wrong.
I think the root problem is that IAS 33 prescribes a mechanical calculation, but applying it correctly depends on a completeness assertion that lives outside the spreadsheet. You need every potential ordinary share outstanding at the reporting date. Options, warrants, convertible bonds, contingently issuable shares. Miss one instrument and the DEPS figure is wrong before you even start the arithmetic. That completeness question sits with management, not with the formula, and it is the piece that ISA 540.13 (a) asks the auditor to evaluate.
Key points
- DEPS includes only instruments that reduce EPS. Anti-dilutive instruments are excluded from the calculation entirely under IAS 33.41 .
- IAS 33.36 requires the treasury stock method for share options and warrants, assuming proceeds buy back shares at the average market price (not the closing price).
- Instruments must be sequenced from most dilutive to least dilutive and added one at a time. Dumping them all in at once can mask an anti-dilutive instrument.
- Entities on a regulated European market present both basic and diluted EPS on the face of the statement of profit or loss, not in the notes.
What the standard actually requires
IAS 33.30 defines DEPS as basic EPS recalculated on the assumption that all dilutive potential ordinary shares outstanding during the period were converted. Both the numerator (profit attributable to ordinary shareholders) and the denominator (weighted average ordinary shares) are adjusted.
Two methods handle the adjustment, depending on the instrument type.
For convertible instruments (convertible bonds, convertible preference shares), IAS 33.33 requires the numerator to add back the after-tax interest or dividend that would have been avoided on conversion. The denominator adds the shares that would have been issued. For share options and warrants, IAS 33.45 applies the treasury stock method: the calculation assumes exercise at the start of the period (or grant date, if later), and treats the proceeds as if used to repurchase shares at the average market price. Only the incremental shares enter the denominator. No numerator adjustment.
What actually happens
At firms like ours, the convertible bond adjustment is usually correct because somebody has to actively compute the after-tax interest add-back, which forces attention. The treasury stock method is where errors cluster. Teams pull the year-end closing price instead of the average market price, producing the wrong notional buyback count. On a volatile stock the difference can be material. I reviewed one file where using the closing price instead of the average flipped the options from dilutive to anti-dilutive, changing reported DEPS by four cents per share.
The sequencing requirement under IAS 33.41 -44 is the second failure point. Each potentially dilutive instrument gets tested individually, ranked from most dilutive to least dilutive, then included one at a time. You stop when the next instrument would be anti-dilutive. This prevents an anti-dilutive instrument from masking the dilutive effect of others higher in the ranking. In our experience, teams skip this step because they already have a DEPS number from the spreadsheet and cannot see why sequencing would change it. It changes it when the population includes a mix of instruments with very different dilution profiles.
Worked example: Schäfer Elektrotechnik AG
Schäfer Elektrotechnik AG is a German electronics company, FY2025, revenue €310M, IFRS reporter listed on the Frankfurt Stock Exchange. Net profit attributable to ordinary shareholders: €18.6M. Weighted average ordinary shares outstanding: 12,000,000. The corporate tax rate is 30%.
Schäfer has two potentially dilutive instruments: 600,000 share options granted to senior management with an exercise price of €24 each, and €10M in convertible bonds carrying 3% annual interest, convertible into 800,000 ordinary shares. Average market price during FY2025: €40.
Step 1: basic EPS
€18.6M divided by 12,000,000 shares. Basic EPS is €1.55.
Step 2: treasury stock method for share options
The 600,000 options at €24 generate notional proceeds of €14.4M. At the average market price of €40, the entity notionally repurchases 360,000 shares. Incremental shares: 600,000 minus 360,000 equals 240,000. No numerator adjustment for options.
Step 3: convertible bond adjustment
Annual interest avoided on conversion: €300,000 (3% of €10M). After-tax saving: €210,000. Denominator increase: 800,000 shares.
Step 4: sequencing and final DEPS
Options produce zero numerator effect and 240,000 incremental shares (incremental EPS per share: nil). Convertible bonds produce €210,000 numerator effect and 800,000 shares (incremental EPS per share: €0.2625). Options rank as more dilutive, so they enter first.
After options: €18.6M / 12,240,000 = €1.5196.
After convertibles: (€18.6M + €0.21M) / 13,040,000 = €1.4432.
€1.4432 is below €1.5196, so the convertible bonds remain dilutive. Both instruments pass. DEPS is €1.44.
Where it gets complicated
Two weeks before sign-off, the engagement partner notices something. Schäfer granted a second tranche of 200,000 share options in September 2025 with an exercise price of €42. Nobody included them in the DEPS calculation because the exercise price exceeds the average market price of €40. The team assumed "out of the money" means "ignore."
They were right to exclude these options from the DEPS denominator. Under the treasury stock method, notional proceeds of €8.4M buy back 210,000 shares at €40, which exceeds the 200,000 options exercised. The incremental share count is negative. IAS 33.47 treats these as anti-dilutive.
But the team never documented the exclusion. The working papers (WPs) showed two instruments in the DEPS calculation and said nothing about the third. IAS 33.70 (b) requires disclosure of instruments that could dilute EPS in the future but were excluded as anti-dilutive in the current period. That disclosure was also missing from the draft financial statements (FS). The file should tell a story, and this one had a gap: an instrument that existed, was correctly excluded from the number, but left no trail in either the WPs or the FS. The partner sent it back. Adding the documentation took 20 minutes. Finding the omission took two review cycles.
Diluted EPS vs. basic EPS
| Dimension | Basic EPS (IAS 33) | Diluted EPS (IAS 33) |
|---|---|---|
| Denominator | Weighted average ordinary shares actually outstanding | Weighted average shares plus incremental shares from dilutive potential ordinary shares |
| Numerator | Profit attributable to ordinary shareholders | Profit adjusted for after-tax effects of assumed conversion (e.g., convertible bond interest added back) |
| Options and warrants | Ignored | Included via treasury stock method if dilutive |
| Presentation | Face of the statement of profit or loss (IAS 33.66) | Equal prominence alongside basic EPS (IAS 33.66) |
A large gap between basic and diluted EPS signals material dilution risk. Investors use the spread to assess how much equity dilution they face before committing capital. When the gap is narrow, some analysts skip the diluted figure entirely, which is a mistake. A narrow spread this year can widen sharply next year if out-of-the-money options move into the money or contingently issuable shares vest.
A disagreement worth documenting
There is a legitimate split among practitioners on how to handle contingently issuable shares under IAS 33.52 -57 when the contingency condition is only partially met at the reporting date.
One position: if the contingency condition is not fully satisfied at year end, the shares are excluded from DEPS because IAS 33.52 says inclusion depends on conditions being "satisfied." This reading is strict and conservative. It keeps the denominator clean.
The other position: if current performance trends suggest the condition will almost certainly be met (say, a revenue target where the entity is at 95% of threshold with one quarter remaining), the shares should be included because excluding them understates dilution that investors can already see coming. This reading prioritises economic substance over the literal timing of the contingency test.
I lean toward the second position because DEPS exists to show maximum dilution, and excluding shares that are almost certain to vest undermines that purpose. But I have worked on engagements where the partner insisted on the first reading, and the regulator did not challenge it. Both approaches survive review if the WPs explain the reasoning. What does not survive review is silence. Picking a position without documenting why is the fastest way to generate a review finding on an EPS file.
Second-order effects most teams miss
DEPS does not just affect the income statement presentation. It feeds directly into price-to-earnings ratios that analysts compute, which means a DEPS error can move the entity's perceived valuation multiple. A four-cent overstatement on a stock trading at €40 shifts the diluted P/E ratio by roughly 0.7x at an EPS around €1.50. That sounds small until you realise that analysts use P/E bands to set buy/sell recommendations, and a 0.7x shift can move a stock from "hold" to "buy" in a screen.
The audit implication: DEPS errors are rarely material in absolute euro terms relative to profit. They become material through the ratio channel. At firms like ours, we test DEPS materiality against both the profit figure and the per-share impact, because a misstatement that is immaterial to net income can still mislead the primary users of the FS when they convert it into a ratio. ISA 450.A16 supports this approach by noting that the auditor considers the effect of uncorrected misstatements on ratios or trends used by financial statement users.
Related terms
Related tools
Related reading
Frequently asked questions
How do I document diluted EPS in the audit file?
List every potential ordinary share outstanding at the reporting date (options, warrants, convertibles, contingently issuable shares) with the terms of each instrument, then document the sequencing test from most dilutive to least dilutive. IAS 33.70(b) requires disclosure of instruments that could dilute EPS in the future but were excluded as anti-dilutive in the current period. The audit file should mirror that disclosure with the supporting calculations.
Does diluted EPS apply to interim financial statements?
Yes. IAS 33.4 requires entities that present earnings per share under IAS 33 to calculate diluted EPS for every period for which a statement of profit or loss is presented, including interim periods reported under IAS 34. The weighted average share count and the average market price for the treasury stock method are calculated for the interim period, not annualised from the full year.
What happens to diluted EPS when all options are out of the money?
When the exercise price of share options exceeds the average market price for the period, the treasury stock method produces a negative incremental share count (the notional buyback exceeds the shares issued). IAS 33.47 treats these options as anti-dilutive, and the entity excludes them from the diluted EPS calculation entirely. Diluted EPS equals basic EPS in this scenario if no other dilutive instruments exist.