IFRS 18 MPMs: Five ways adjusted profit differs from IFRS

IFRS 18 will bring non-GAAP metrics, now called management performance measures or MPMs, inside the audited financial statements from 2027. A recent IFRS tentative agenda decision, regarding MPMs that include hypothetical amounts, has prompted us to ask how far such a measure can and should depart from IFRS.

We set out six different kinds of adjustment that a management performance measure may contain. Only one of them matches what we think many investors expect. The others range from a change of accounting policy through to including amounts for events that never happened.

An IFRS 18 management performance measure (MPM) is a subtotal of income and expenses that a company uses in public communications to convey management’s view of an aspect of its financial performance. The purpose is not to replace IFRS but to supplement it with additional measures that add further insights into performance. The IASB has recognised that companies routinely publish alternative measures (with labels such as adjusted, underlying or non-IFRS) and that investors benefit from knowing more about them. IFRS 18 requires MPMs to be disclosed in a single note, reconciled to the most directly comparable IFRS subtotal, and explained. In addition, the effect of each reconciling item on tax and non-controlling interests must also be disclosed. The new standard (and the MPM requirements) becomes effective in 2027.

We think many investors assume that adjusted profit is simply IFRS profit with certain IFRS expenses removed – such as non-recurring items and perhaps some amortisation of acquired intangibles. These are certainly the most commonly found adjustments, but they are not the only ones you will find, or that IFRS 18 permits.

Extending an MPM to include adjustments that substitute a different measurement or recognition basis is also permitted by IFRS 18. Where a company calculates an MPM other than by using the accounting policies applied in its financial statements, it must say so and explain the calculation. Where the calculation also differs from the accounting policies required or permitted by IFRS, that additional fact must be stated and any terms used explained.

Does IFRS 18 permit MPMs that reflect hypothetical transactions or events?

The IFRS Interpretations Committee¹ was recently asked whether IFRS 18 permits such a measure to go still further and include income and expenses that a company has not recognised and never will recognise, including for events or transactions that never actually happened.² The tentative decision (at the time of writing it is still subject to finalisation) is that this is indeed permitted. That decision prompted this article.

The Interpretations Committee decision has proved controversial, with some saying these hypothetical measures go too far, are open to abuse, and may confuse investors. While we agree that there can be ‘bad’ MPMs that are a poor reflection of performance and which investors are best advised to ignore, we do not think it is possible or desirable for IFRS 18 MPMs to be constrained in the way some advocate.

We explain more about the Interpretations Committee decision, and the implications for investors, below. First, we consider the range of adjusted metrics you may come across, and that will be subject to the MPM presentation and disclosure requirements in IFRS 18 from next year.

Six kinds of adjustment

We think it helps investors to separate the adjustments that may appear in a management performance measure into six categories. They differ in how far they depart from IFRS, and in what an investor can do about them.

Removing all or part of an item

Removing recognised income or expenses: The adjustments commonly include restructuring charges, impairment losses, amortisation of acquired intangibles. The measure is IFRS profit excluding specified items, with the MPM containing a subset of income and expense, each measured as IFRS requires.

Excluding part of a recognised item: An IFRS amount is split into components on a basis the company defines, and only one of these is excluded. Neither the component excluded nor the one retained is measured in a way IFRS specifies. The example we show below is a fair value change split into an expected return and a residual.

Applying a different accounting policy

Applying an accounting policy that IFRS permits but the company has not adopted: The amounts included in the MPM are recognised and measured on a basis available under IFRS, just not the one used in the financial statements.

Assuming an accounting policy election that was available at the time but was not made: Hedge accounting is the obvious case because it requires formal designation at inception and cannot be applied after the event. While it may appear that the MPM is the same as if IFRS had been properly applied, you cannot be sure due to the risk of selective application.

Applying a measurement or recognition basis that IFRS does not permit: Apart from the first category above, this is perhaps the most common source of adjustments. Examples include lease costs on a cash basis, inventory on a last-in first-out basis, or recognising the whole grant-date value of a share option in the year of grant rather than over the vesting period.

Adjusting for hypothetical transactions or events

Including income and expenses for transactions or events that did not occur: This is the subject of the Interpretations Committee decision. The examples they looked at include profit as though a pandemic had not happened, or as though an acquisition had taken place at the start of the year. There is some overlap here with our second category – one could argue that the expected return on investments also reflects something that never happened.

These categories are merely our guide to help you understand the adjustments. Our descriptions are brief, the categories may not be mutually exclusive, and nor are they used in IFRS 18. Nevertheless, we think it is a useful way to illustrate to investors the breadth of adjustments IFRS 18 MPMs (and existing adjusted profit metrics) can cover.

The IFRS 18 reconciliation will always take you from the MPM back to an IFRS subtotal. What differs is the nature of the reconciling item. In the first category it is an identifiable income or expense measured as IFRS requires, so you can judge each item on its merits. In the other five it is a management determined part of an IFRS item, the difference between two ways of measuring or timing the same thing, or even an amount with no transaction behind it at all. Each of these you can accept or reject, but we think the adjustments are more difficult to analyse.

Two kinds of faithful representation

The Interpretations Committee’s reasoning is based on the concept of faithful representation. A measure using a non-IFRS measurement basis for a particular item can faithfully represent management’s view of performance, provided it does what management says it does, and the required disclosures are given. However, in this context, faithful representation says nothing about whether a measure is a good one or a bad one. IFRS 18 applies no restriction on what can be an MPM, however egregious the adjustments might seem to investors.

Faithful to management’s definition is not necessarily faithful to the economics

Whether the measure faithfully represents the underlying economics is a different question. For example, suppose a company decides that for its adjusted profit metric it will measure the stock-based compensation expense using the intrinsic value rather than fair value of stock and options granted. This would not result in removing the expense entirely, as some companies do (a first category adjustment). Instead, it would mean recognition of the value of shares granted, but assign little or no value to the grant of options, given that most are issued at-the-money. This is an accounting policy that differs from IFRS requirements – our fifth category.

We think this alternative stock-based compensation accounting policy is an incomplete and difficult to analyse partial recognition of the economic expense. The time value of an option is real, it is valuable, and it is most certainly a cost to shareholders. Any measure that omits it cannot be a faithful representation of the economics of employee compensation, whatever it faithfully represents about management’s definition. Omitting the expense entirely is also problematic, but at least investors know exactly what they are getting. Nevertheless, both these approaches would qualify as MPMs under IFRS 18.

Excluding part of an item – Aviva

Some IFRS gains or losses are required to be split by a standard on a specified basis, with an IFRS profit subtotal that includes only one component. For example, IAS 19 splits the remeasurement of a pension obligation into the net interest accretion on a pension deficit or surplus, which is reported in profit or loss, and other remeasurements which are included in other comprehensive income. However, if a company itself splits a single IFRS required amount, the basis becomes the company’s choice and, if they provide a subtotal that includes just one part, it becomes an MPM.

Insurers are a good example. Aviva measures its investments at fair value through profit or loss and does not use the IFRS 17 option to take part of its insurance finance expense to other comprehensive income. Its main non-GAAP performance measure, group adjusted operating profit, includes an expected return on investments less the unwind of the discount on insurance liabilities. Variances from the expected asset return and the effect of changes in economic assumptions on the liability are excluded. IFRS does not specify either split.

Aviva 2025 group adjusted operating profit – definition

…..

Aviva plc Annual Report and Accounts 2025 (extract)

The first two adjustments in the table above are those that reflect the split of an IFRS amount. The others are the exclusion variety – the first category in our list. Aviva provides extensive explanations about methodology and assumptions, only part of which we show above.

An expected return set too high becomes income never earned

The reliability and comparability of an MPM that includes these second category adjustments depends on how the retained component is determined. A discount unwind at a stated rate for a liability is close to mechanical, whereas an expected asset return requires decisions about methodology and setting of assumptions. If an expected return is set too high the excluded variances will be persistently negative and the measure will include income never earned, which is close to our sixth category. The IASB removed the expected return on pension plan assets from IAS 19 in 2011 for similar reasons.³

Despite the limitations, we think this kind of disaggregation can be useful, particularly where it separates the effect of assumption or market-driven changes that would otherwise be combined with a recurring component, and where only the latter has predictive value.⁴ For example, suppose a company isolates the effect of changes in assumptions within its stock-based compensation expense and excludes only that element as non-recurring, leaving the amortisation of grant-date value in the measure. The two parts still add up to the IFRS expense, but the disaggregation may help investors understand and forecast the expense.

One measure, several types of adjustment – TotalEnergies

A further example shows how several of the categories can sit inside a single measure. TotalEnergies reports adjusted net income, the measure most widely used to compare the performance of the oil majors. It has adjustments that reach three of our six categories.

TotalEnergies 2025 adjustment items – basis of preparation

TotalEnergies Universal Registration Document 2025, note 8 to the consolidated financial statements (extract)

The special items described above are our first category of MPM adjustments. Recognised income and expenses – gains and losses on asset disposals, restructuring charges, impairments – are taken out of the measure, but each is measured as IFRS requires. In 2025 special items reduced reported net income by $1,185m.

A measurement basis IFRS prohibits – adopted to match US peers

The inventory valuation effect is our fifth category. In the financial statements, petroleum product inventories are measured on a first-in first-out basis, consistent with IAS 2. In the adjusted results of the Refining & Chemicals and Marketing & Services segments they are measured at replacement cost, which the company says approximates last-in first-out. IAS 2 prohibits LIFO, so this is a basis no company reporting under IFRS may use. The stated reason for the adjustment is comparability with the segments’ main competitors – largely US companies for which LIFO remains available under US GAAP. The effect reduced 2025 net income by $610m. There is a good argument that a replacement cost basis for inventory helps to produce more relevant performance metrics when commodity prices are volatile, even though the precise method by which it is implemented is specific to the company and may not necessarily be comparable across the sector.

The ‘changes in fair value’ component contains three separate adjustments. IFRS requires trading inventories to be measured at fair value using period-end spot prices, whereas the internal measures of performance use forward prices – a different measurement basis, so our fifth category again. The company also enters into storage contracts whose future effects are carried at fair value in its internal economic performance, and states that IFRS precludes recognising that fair value effect. The contracts are real and the difference unwinds as they settle, but this too is an accounting policy IFRS does not permit. Derivatives used to manage the risk on operational contracts is the third adjustment. Deferring a derivative gain (so that it is recognised in the same period as the item it was taken out to hedge) is what cash flow hedge accounting achieves, which seems to put it in our fourth category.⁵

TotalEnergies 2025 adjustment items to net income

TotalEnergies Universal Registration Document 2025, alternative performance measures (extract)6

Together the three lines add back a net expense of $2,460m, giving adjusted net income of $15,587m, almost a fifth higher than the IFRS figure. Three lines, and behind them one exclusion and four separate departures from IFRS measurement or recognition, with nothing in the reconciliation to tell them apart. However, none reaches our sixth category, which is what the tentative agenda decision covers.

The financial statements are prepared and audited on the IFRS basis, the adjusted measures are labelled as non-GAAP, and each reconciling item is disclosed. We offer no judgement about the usefulness of the adjusted profit metric in this case. Our point is about what a reader can do with the reconciliation, not about whether the accounting is right. Nevertheless, we expect that IFRS 18 will prompt welcome improvements by many companies – more explanations, greater clarity in the reconciliations, and the additional disclosure of tax and NCI effects per adjustment.

Amounts that were never earned or incurred

The tentative IFRS Interpretations Committee agenda decision concerns our sixth category. The Committee was asked whether a measure containing income and expenses that a company has not recognised, and never will recognise, can meet the definition of a management performance measure. It has tentatively concluded that it can – the decision will be confirmed or amended after consultation. The view is that IFRS 18 places no restriction on how a subtotal is calculated,⁷ and the requirement that a measure communicate management’s view of an aspect of financial performance means that, provided management does indeed use the measure, it is acceptable irrespective of how hypothetical it is and whether or not it is useful to investors.

The examples put to the Committee are all hypothetical measures, but in different ways.

Annualised property rental income, calculated by multiplying the rent recognised in the final month by twelve: The underlying leases are real, but the adjusted figure is not a re-measurement of the rent the portfolio earned; it is the rent the portfolio would have earned had the final month applied all year, which is why the Committee treats it as hypothetical and why we place it in the sixth category.

Pro forma gross profit calculated as though an acquisition had happened at the start of the year: This recreates something that partly resembles the old pooling of interests or merger accounting, but it is still hypothetical.

Profit excluding the effects of a major crisis, such as a pandemic: The company estimates the income and expenses it would have generated had the crisis not occurred.

All three might have some use for investors, although it does seem odd to include these completely hypothetical measures of profit within the financial statements, even if they are described as management performance measures.

Many of those responding to the IASB felt these adjustments go too far, including the securities regulators, and the CFA Institute, the only investor group that expressed a view. IOSCO reports that most of its members have concerns, ESMA recommends that the IASB amend IFRS 18 to exclude such measures, and the CFA Institute asks for the decision to be withdrawn and the matter referred to the IASB.⁸

Those objecting clearly think there are ‘bad’ MPMs that should not be presented in audited financial statements. Doing so risks legitimising measures that may mislead more than they inform. However, there is a valid counter argument. A measure that fails the definition of an MPM probably does not disappear as a result. It may be published anyway, but outside the financial statements, probably with less explanation, no reconciliation, no attribution of the tax and non-controlling interest effects, and no audit. Keeping these more extreme deviations from IFRS outside of the MPM definition would arguably reduce transparency rather than improve it.

Defining a bad MPM is probably impossible – the judgement therefore falls to investors

Trying to define what is an unacceptable adjustment is also very difficult. For example, the expected asset return disaggregation used by Aviva, which we think is perfectly acceptable (although of course you should scrutinise the assumptions), could arguably be categorised as a hypothetical measure that ESMA is questioning. If the value of investments falls, but the adjusted profit includes an ‘expected’ positive return, the measure arguably becomes hypothetical and based on an event that never occurred. The same is true of same-store sales and profit at constant currency, both widely used and both arguably hypothetical.

We think many MPMs can be problematic, sometimes completely outrageous, and often less useful for investors than IFRS compliant subtotals. Bad MPMs are also not just confined to the hypothetical ones – excluding the stock-based compensation expense, for example, is fully IFRS compliant. However, as we explain above, this produces a poor measure of performance that we think investors should ignore.Nevertheless, we agree with the conclusion of the Interpretations Committee, and with the IFRS 18 approach of not attempting to create a ‘good’ versus ‘bad’ categorisation of MPMs. Screening out a subset of ‘bad’ MPMs from financial statements through accounting standards would be worthy, but it is just not practical.

Our main concern is how investors interpret MPMs rather than what meets the IFRS 18 definition. Putting a measure in the audited financial statements does not mean that it is useful, or that it faithfully represents performance. It simply means management claim they use it, and that certain disclosures will be provided. Investors should remain vigilant and scrutinise companies’ motivation and explanations for the adjustments to IFRS items in MPMs and be ready to ignore measures that are deficient.

Reading the management performance measure note

We think investors should treat the arrival of IFRS 18 MPMs as an improvement in transparency, but it does not mean that every MPM is a better measure of performance than IFRS profit.

For each item in the MPM reconciliation, think about which of the six categories it belongs to. Where management has simply excluded an IFRS expense, the question is whether the item is genuinely unusual and whether excluding it improves your view of recurring performance or provides a better basis for forecasting. Where only part of an item is excluded, examine how the retained part is defined. Where an IFRS measurement or recognition basis has been replaced, go further and ask why the alternative is more relevant and whether it changes the timing of profit systematically. A change of basis appears in the description of how the measure is calculated rather than in the reconciliation, so read that first.

Look for the two statements the standard requires: that accounting policies different from those used in the financial statements have been applied, and that those policies do not comply with IFRS. If either appears, treat the resulting MPM with more caution. Ignore the measure altogether if you think it does not faithfully represent performance, even where the rest of the market relies on it – although widespread use of a less relevant measure is itself worth knowing, since it may affect how the company’s performance is understood by the rest of the market.

MPMs can result in welcome additional disaggregation

MPMs can also contain much useful information, which is why we have previously welcomed this aspect of IFRS 18. An adjusted metric is particularly useful where it supplies additional disaggregation, or where it applies a methodology that may be better than or complements IFRS. Furthermore, the attribution of tax and non-controlling interest effects to each reconciling item lets you build the earnings measure you want rather than accept the one you are given.

MPMs generally have more to offer in assessing how a business has performed than in forecasting and valuation. That is not to say they are irrelevant to valuation – excluding the amortisation of acquired intangibles is a sensible adjustment when computing enterprise value multiples. An adjustment which helps you judge the performance of a business may be the wrong starting point for a forecast, so for forecasting and valuation we would stick closely to the IFRS figures.⁹

Insights for investors

Establish for each adjustment whether an item has been removed, partially removed or re-measured. A removed item can be judged on its own merits; a measurement difference can only be accepted or rejected as a whole.

Where an IFRS 18 management performance measure uses a basis that does not comply with IFRS, that fact must be stated. Look for the statement and evaluate the merits of the policy difference.

Bringing these measures into the audited notes improves disclosure and adds discipline, but it does not make a subtotal that misrepresents performance informative.

A measure can faithfully represent management’s own definition and still misrepresent the economics. Test it against the economics of the transaction, not against the description in the note.

A management performance measure is most useful where it adds disaggregation or shows how management genuinely steer the business. An MPM is usually least useful where management moves away from measurement bases mandated by IFRS, or includes hypothetical transactions or events.

Don’t miss new articles published by The Footnotes Analyst – subscribe here:

No advertising, no spam, just new articles. Unsubscribe anytime.

Send us a question or comment about this article:

For related articles select subject tags or see the suggestions below:

Print or save article as a pdf:

Cookie	Duration	Description
cookielawinfo-checkbox-analytics	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Analytics".
cookielawinfo-checkbox-functional	11 months	The cookie is set by GDPR cookie consent to record the user consent for the cookies in the category "Functional".
cookielawinfo-checkbox-necessary	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookies is used to store the user consent for the cookies in the category "Necessary".
cookielawinfo-checkbox-performance	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Performance".
viewed_cookie_policy	11 months	The cookie is set by the GDPR Cookie Consent plugin and is used to store whether or not user has consented to the use of cookies. It does not store any personal data.

IFRS 18 MPMs: Six ways adjusted profit departs from IFRS