Transfer Pricing Adjustments Under Pillar Two: How They Interact

A transfer pricing adjustment does two things under the GloBE rules that most groups still budget for separately. First, it resets the income or expense a constituent entity reports for the GloBE computation, not just its local taxable income. Second, if the correction lands after the group’s Pillar Two return is already filed, it can force a recalculation of that entity’s effective tax rate (ETR) for a year everyone assumed was closed — and generate additional top-up tax on the back of it. Transfer pricing and Pillar Two teams that run as separate workstreams are, in effect, each only seeing half of the same transaction.

The clearest public account of how this mechanism actually works is the UK’s implementation of the OECD GloBE Model Rules, transposed into Finance (No.2) Act 2023 and documented line by line in HMRC’s Multinational Top-up Tax manual. It is detailed enough to verify rather than approximate, so it anchors everything below.

Where transfer pricing enters the GloBE computation

Every constituent entity’s GloBE income or loss starts from its own financial accounting profit, adjusted under a fixed set of rules. Two of those adjustments exist specifically because related-party transactions are priced — and sometimes re-priced — between group members.

Cross-border transactions: the rule follows the high-tax side

Income, expenses, gains and losses from a transaction between a constituent entity and another group member are, as a general rule, reflected in adjusted profit exactly as each entity recorded them (HMRC, MTT21120). Transfer pricing complicates that the moment the two tax authorities involved don’t agree on the price: one side assesses a transfer pricing adjustment, the other may or may not grant a corresponding one.

HMRC’s manual sets out the fix for the asymmetric case directly (MTT21125). Where a cross-border transaction creates a permanent difference — a mismatch between the tax and accounting treatment of an amount that never reverses, so it never produces a deferred tax asset or liability — for both counterparties, each entity’s adjusted profit is restated to reflect its own taxable income. But where only one counterparty is a “high-tax member” and the permanent difference arises only on that side, both entities’ adjusted profits are restated to reflect the high-tax member’s position. A high-tax member, for this purpose, is one whose nominal tax rate exceeds 15% in the period, where the standard members in its territory also cleared a 15% effective rate in at least one of the two preceding periods.

In practice: if a unilateral transfer pricing assessment increases taxable income in the high-tax jurisdiction without a matching downward adjustment in the low-tax one, GloBE does not let that one-sided correction sit unmatched. It pulls the low-tax entity’s adjusted profit into line with the high-tax entity’s position — which is the opposite of what a group under TP audit in the high-tax country might hope for.

Same-territory transactions: arm’s length isn’t automatic, but it can be required

Transactions between group members in the same territory are usually left alone, because they don’t change the aggregate income taxed in that territory. HMRC’s manual lists four specific conditions under which an arm’s length adjustment is required anyway (MTT21230): an asset transfer between same-territory group members booked as a charge without an arm’s length basis; a transaction between a minority-owned member and a non-minority-owned member without an arm’s length basis; a transaction between an investment entity and a non-investment entity without an arm’s length basis; or — simplest of all — the same transaction recorded at different values in each counterparty’s own accounts. Any domestic restructuring, minority-investor joint venture, or investment-entity carve-out inside a single territory needs to be checked against this list before anyone assumes same-country transactions are automatically out of scope.

What happens when the correction arrives after the GIR is already filed

Transfer pricing corrections are rarely contemporaneous with the year they relate to. An audit assessment, a mutual agreement procedure (MAP) settlement, or a group’s own self-initiated year-end true-up can all land twelve, twenty-four or more months after the relevant accounting period closed — and after the GloBE Information Return for that period has already been filed.

HMRC’s manual treats this as a distinct event: a post-filing adjustment of covered taxes (MTT25300), implementing section 217 of Finance (No.2) Act 2023. If a group determines that a member’s covered tax liability for a prior period increased or decreased, that change is dealt with in the current year’s return — but the path it takes depends entirely on its size. The manual is explicit that a transfer pricing adjustment is one of the events that can trigger this: “a transfer pricing adjustment may also give rise to a post-filing adjustment of covered taxes.”

The €1 million threshold that decides whether you recalculate

This is the detail that turns a routine TP true-up into a multi-period recalculation exercise, or lets it pass through unnoticed.

An increase in covered tax liability for a prior period is simply added to the current period’s covered tax balance — no recalculation, no reopening of the prior year’s ETR.

A decrease is where the threshold bites. If the aggregate decrease across the low-tax or mixed sub-group of a territory for the period is below €1,000,000, the reporting entity can elect, territory by territory and year by year, to treat it as insignificant: it simply reduces the current period’s covered tax balance. Above that threshold — or if the election isn’t made — the group must recalculate the ETR and top-up tax amount for the affected prior period, for every other member in that territory, and carry the correction forward through every subsequent period’s adjusted profit and ETR. Where the decrease stems from a reduction in adjusted profit rather than covered tax itself, HMRC’s guidance adds a backstop: the adjusted profit for the prior period is only corrected to the extent needed to stop the ETR from falling — not simply restated to match the new accounting figure.

Direction and size of the correction	Treatment
Increase in covered tax, any amount	Added to the current period’s covered tax balance. No recalculation.
Decrease in covered tax, aggregate below €1m per territory, election made	Treated as insignificant. Reduces the current period’s covered tax balance only.
Decrease in covered tax, aggregate at or above €1m per territory, or no election	Mandatory recalculation of ETR and top-up tax for the prior period and every subsequent period affected.

The practical trap is the word “aggregate.” The threshold is tested per territory, not per transaction or per entity. A transfer pricing true-up that looks immaterial at the level of a single intercompany invoice can combine with unrelated covered-tax corrections elsewhere in the same jurisdiction — a tax credit adjustment, a deferred tax rate change — to clear €1 million in aggregate without any single item looking large enough to flag on its own.

Why this has to be a joint call between tax and transfer pricing teams

Three things follow from the mechanism above, and none of them are optional once a group is inside Pillar Two’s scope.

First, a transfer pricing adjustment needs to be dated and sized against the specific GloBE period it affects, not just booked against the current year’s tax charge the way it would be outside Pillar Two. The Pillar Two team needs that mapping to run the €1 million aggregation test correctly — and TP teams rarely produce it unprompted, because their own documentation is built around the tax return, not the GloBE return. A buyer’s checklist for transfer pricing documentation tooling that actually tracks this link is covered in our transfer pricing documentation software guide.

Second, a self-initiated TP true-up booked late in the filing cycle — common where a US parent adjusts intercompany pricing to target a specific operating margin before year-end — is no longer purely a transfer pricing decision once Pillar Two applies. Making the adjustment in one period rather than another can change which side of the €1 million line a correction falls on, and whether it triggers a multi-period GloBE recalculation.

Third, the high-tax-member mechanics above mean a TP position that looks defensible in isolation can still pull an unrelated low-tax entity’s GloBE income into line with it. Any multinational still treating a jurisdiction’s safe harbour status as fixed for the year should revisit it in light of this — see our breakdown of the six Pillar Two safe harbours currently in force — because a post-filing TP adjustment reaching back into a prior period can move the numbers a safe harbour test relied on.

None of this is tax advice for a specific company’s position; it is a map of the mechanism, built from the one jurisdiction that has published it in enough detail to verify.

FAQ

Does a transfer pricing adjustment automatically change GloBE income?
Not automatically. Same-territory transactions are generally left alone unless one of four specific conditions applies (an asset transfer without an arm’s length basis, a minority-owned/non-minority-owned pairing, an investment-entity/non-investment-entity pairing, or mismatched booked values). Cross-border transactions are matched between the two entities, with the high-tax counterparty’s position controlling when only one side carries a permanent difference.

What happens if a transfer pricing adjustment is settled after the GloBE Information Return has already been filed?
It is treated as a post-filing adjustment of covered taxes. An increase simply adds to the current period’s covered tax balance. A decrease is tested against a €1 million per-territory threshold: below it (and with an annual election), it’s absorbed in the current period; at or above it, the prior period’s ETR and top-up tax must be recalculated, with the correction carried forward through subsequent periods.

Is there a minimum size below which a transfer pricing correction can be ignored for Pillar Two?
Only for decreases in covered tax liability, and only if the aggregate decrease across the relevant territory for the period is below €1 million, tested in aggregate rather than per transaction. Increases in covered tax liability have no such threshold.

Can a transfer pricing adjustment in one entity affect a completely different, unrelated entity’s Pillar Two position?
Yes, in two ways: a cross-border adjustment can force a counterparty’s adjusted profit to follow the high-tax side’s position even if the counterparty itself made no TP change, and a post-filing decrease is aggregated across every member in the same territory, so unrelated corrections can combine to cross the €1 million recalculation threshold together.