How should we measure the 20 percent we did not buy?

One acquisition to work through.

An example group reporting under IFRS buys 80 percent of Company B on 1 January 2026 for 8,000,000 dollars in cash. B’s identifiable net assets have a fair value of 7,000,000. A valuer puts the fair value of the 20 percent the founders kept at 1,800,000. That is less than 20 percent of the 10,000,000 the group implicitly paid for the whole company, because a minority stake does not carry control. B then earns 1,000,000 a year and pays no dividends.

The controller has to put a number on the founders’ 20 percent in the opening consolidation. The answers that come up first, such as Grant Thornton’s IFRS 3 note, explain both methods well and stop at day one, describing the later effects in words. The later effects are where the choice really costs or saves, so we followed the numbers for two years.

What the standard asks you to choose.

We read IFRS 3 in the Australian compiled text, which carries the same wording and paragraph numbers. Paragraph 19 sets the choice for ordinary shares held by outside holders:

“For each business combination, the acquirer shall measure” them at fair value or at their proportionate share of the acquiree’s identifiable net assets.IFRS 3, paragraph 19

The words “for each business combination” matter. The election is made one acquisition at a time, so two subsidiaries in the same group can be measured differently. Paragraph B45 explains why fair value per share can be lower for the outside stake than for the controlling one: a discount for lack of control. Paragraph 32 then measures goodwill as the consideration plus the non-controlling interest, less the identifiable net assets.

Groups that also report under U.S. GAAP have one answer. Deloitte’s IFRS and U.S. GAAP comparison notes that ASC 805 requires the noncontrolling interest at fair value, which matches the IFRS fair value option.

Day one: 400,000 dollars of goodwill apart.

Proportionate share puts the founders at 20 percent of 7,000,000, which is 1,400,000. Fair value puts them at 1,800,000, and the extra 400,000 becomes goodwill that belongs to them. After that the two methods move in parallel. IFRS 10 paragraph B94 attributes 20 percent of B’s profit to the founders each year, so both balances rise by 200,000 a year, and the 400,000 gap simply stays there.

Two years later, the smaller goodwill tests harder.

Suppose B is its own cash generating unit and at the end of 2027 its recoverable amount is 10,500,000. Its net assets are now 9,000,000. Under fair value the test is simple: 9,000,000 plus 2,800,000 of goodwill is 11,800,000, so the loss is 1,300,000, and IAS 36 paragraph C6 splits it 80 and 20 like profit. The parent bears 1,040,000.

Under proportionate share, the text of IAS 36 asks for a gross up first. Paragraph C4 says goodwill attributable to the outside holders is in the recoverable amount but not in the books, so the entity “shall gross up the carrying amount of goodwill” before comparing. The 2,400,000 grosses up to 3,000,000, which is the parent’s price applied to 100 percent. Now the carrying amount tested is 12,000,000, the loss found is 1,500,000, and paragraph C8 recognises only the parent’s 80 percent of it: 1,200,000.

So the method with less goodwill books a smaller total loss, as the usual summaries say, but the parent’s own share of the loss is larger, and the test fails at a recoverable amount of 12,000,000 instead of 11,800,000. The cause is the discount for lack of control. Without it the two methods would give the same gross figure.

Buying the rest is an equity transaction.

Now suppose there is no impairment, and in January 2028 the group buys the founders’ 20 percent for 2,600,000. We read IFRS 10 for this. Paragraph 23 says changes in ownership that keep control are equity transactions, and B96 removes the non-controlling interest at its carrying amount and takes the difference from the price straight to equity attributable to the parent. Goodwill does not move.

Under fair value the founders’ balance is 2,200,000, so parent equity falls by 400,000. Under proportionate share it is 1,800,000, so parent equity falls by 800,000. The extra 400,000 is the same goodwill that proportionate share left off the balance sheet on day one, now showing up as a larger charge to equity.

A way to decide, deal by deal.

• If the group, or a parent above it, reports under U.S. GAAP, use fair value. ASC 805 has no proportionate option, so one measurement serves both sets of books.
• If a put, call or forward over the remaining shares is part of the deal, or buying the rest is the plan, lean towards fair value. The buyout charge to parent equity is smaller by the goodwill you recognised for the outside holders.
• If the outside holders are likely to stay for good and there is no market price for their shares, proportionate share saves a valuation and keeps goodwill lower.
• If the subsidiary has little headroom in its impairment test, model both methods first. When the fair value of the outside stake carries a discount for lack of control, the grossed up proportionate test fails at a higher recoverable amount.

Both methods are allowed, and each suits a different kind of deal. The useful habit is to make the election on purpose, before the opening consolidation is posted, with the later events in mind.

Where Rivane holds the answer.

In Rivane each company in an entity group is a member with an ownership percentage and a consolidation method: full consolidation or the equity method. A member that was acquired also carries its acquisition date, the consideration paid and a goodwill election, so the choice in paragraph 19 is recorded per acquisition, which is how the standard frames it.

For every partly owned subsidiary that is fully consolidated, Rivane measures the non-controlling interest each period as the outside share of the subsidiary’s equity, translated into the group’s currency, and the outside share of its net income, and keeps one ledger row per group, subsidiary and period. An associate on the equity method has no non-controlling interest; its share of profit is shown as one figure below the line rather than in revenue.

# One subsidiary in a Rivane entity group.
# Example company. Shape only.

member              Subsidiary B
consolidation       FULL          (EQUITY for an associate)
ownership           80.00 %
acquisition date    2026-01-01
consideration       8,000,000.00 USD
goodwill election   recorded per acquisition

each period, per partly owned subsidiary
  NCI balance       20 % of B's equity, in group currency
  NCI share of P&L  20 % of B's net income
  -> one nci_ledger row per group, subsidiary, period

associate on EQUITY -> no NCI, share of profit
                       shown below the line

One limit to plan around. Rivane does not yet hold acquisition date fair values: goodwill is calculated against the subsidiary’s book net assets, and there is no input for the fair value of the outside stake. Today its measurement matches the proportionate share method when book values equal fair values. If you elect fair value, or the purchase price allocation steps up assets, carry the difference in a consolidation journal and keep the workings in the deal file below.

What to keep in the deal file.

• The election for this acquisition, the date it was made and who approved it, with the paragraph 19 reasoning.
• The fair value of the non-controlling interest and how it was measured, or a note that proportionate share was chosen and none was needed.
• The goodwill calculation under paragraph 32, tied to the opening consolidation journal.
• A roll forward of the non-controlling interest each period: opening balance, share of profit and other comprehensive income, dividends to outside holders, closing balance.
• For impairment, the gross up workings under C4 when proportionate share was chosen, and the split of any loss under C6 and C8.

Choose with the exit in mind.

The goodwill difference on day one is the visible part. The same difference returns as the parent’s share of an impairment, or as the charge to equity when the outside holders sell. If buying them out is likely, fair value usually costs less over the life of the deal. If they are staying, proportionate share is simpler. Either way, write the election down before the first consolidation.

If you are setting up a group with partly owned subsidiaries and want to see how ownership, the election and the non-controlling interest ledger fit together on your own structure, talk to the Rivane team.

Sources

• AASB 3 Business Combinations, compiled December 2022, which reproduces IFRS 3 with the same paragraph numbers. Downloaded and read in the file for paragraph 19 (the measurement choice for non-controlling interests), paragraph 32 (goodwill) and paragraphs B44 and B45 (fair value of a non-controlling interest and the discount for lack of control).
• AASB 10 Consolidated Financial Statements, compiled November 2024, which reproduces IFRS 10. Downloaded and read for paragraphs 22 and 23 (presentation, and changes in ownership without loss of control as equity transactions), B94 (attribution of profit, including a deficit balance) and B96 (adjusting the carrying amounts when the proportion changes).
• AASB 136 Impairment of Assets, compiled January 2026, which reproduces IAS 36. Downloaded and read for Appendix C, paragraphs C4 (grossing up goodwill when the non-controlling interest is measured at its proportionate share), C6 (allocating the loss on the same basis as profit) and C8 (the share of the loss that is never recognised).
• Deloitte, Roadmap: Comparing IFRS Accounting Standards and U.S. GAAP, section 5.1 Business Combinations. Read in the page for the difference on measuring a noncontrolling interest: a choice for each acquisition under IFRS, fair value under U.S. GAAP.
• Grant Thornton, Insights into IFRS 3: Recognising and measuring non-controlling interests (2023). Downloaded and read as an example of the current answer. It describes both methods with a day one example and notes the effect on impairment and later purchases in words.

A note on how this was researched. The IFRS paragraphs were read in the Australian compiled standards, AASB 3, AASB 10 and AASB 136, which reproduce IFRS 3, IFRS 10 and IAS 36 with the same paragraph numbers, because the consolidated EU text could not be reached on the day we wrote this. Paragraph numbers come only from text we read. The U.S. GAAP point is taken from Deloitte’s comparison, not from the Codification. The company is an example, and every figure was calculated in a script, which you can repeat with a spreadsheet. The Rivane section describes the consolidation engine as built today, checked against the code rather than marketing copy. Confirm the treatment of your own acquisitions with your auditor.

On embedded social posts: we looked for public posts from standard setters or practitioners on measuring non-controlling interests and found none we could verify as current, relevant and authentic, so this article uses ordinary citations instead.