The Pillar Two 2026 playbook for mid-market multinationals

Why Pillar Two is no longer just a Big Four problem

For three years, GloBE has been treated by mid-market tax functions as a problem for groups with familiar headquarters in Paris, Munich and Stockholm. That assumption is now wrong. With Belgium, the United Kingdom, South Korea, Vietnam and Australia all running their first full filing cycles in 2026, and the EU Directive (2022/2523) finishing its member-state implementation, the question for any group approaching the €750 million revenue threshold is no longer whether Pillar Two applies, but which entities are caught and which top-up mechanism reaches them first.

We see the same pattern across our mid-market client base: a group that crossed the threshold in 2024 or 2025 expected to be a year away from compliance, only to discover that a Qualified Domestic Minimum Top-up Tax (QDMTT) in one of its operating jurisdictions had already created a filing obligation. The cost of a late start is rarely the tax itself — most well-run groups end up close to the 15% floor anyway. The cost is the data work, the reorganization of the consolidation pack, and the board-level discussion that should have happened twelve months earlier.

This piece is the playbook we wish every CFO and group head of tax in the €750m–€5bn band had on their desk by Q1 of the year they cross the threshold. It is deliberately practical: where to start, what to ignore, and which decisions to escalate.

The €750m threshold — what actually counts

The threshold test is borrowed from country-by-country reporting and seems mechanical, but it is where most mid-market groups initially misread their position. The rule (Model Rules Article 1.1) requires consolidated revenue of at least €750 million in at least two of the four fiscal years immediately preceding the tested year. Three points are routinely missed:

• It is consolidated revenue at the level of the Ultimate Parent Entity (UPE), not segment revenue and not statutory revenue. Equity-method investments are out. Discontinued operations are in until the period in which they are deconsolidated.
• Currency translation matters.A USD-functional UPE sitting at $760m in 2024 may fall below €750m once translated using the OECD's prescribed average rate. Run the conversion every quarter; do not wait for the annual report.
• Acquisitions reset the clock the wrong way. If you acquire a target whose own consolidated revenue exceeds €750m, the combined group is in scope from the next fiscal year — even if the acquirer alone never crossed.

QDMTT first, then IIR, then UTPR — and why that matters

The three Pillar Two charging mechanisms are usually presented as a menu. They are not. They are an ordered cascade, and getting the order right is what determines where your top-up tax actually lands.

1. Qualified Domestic Minimum Top-up Tax (QDMTT).Charged by the jurisdiction in which a low-taxed Constituent Entity is located. If the QDMTT is “qualified” under the OECD peer review, it is creditable against the IIR and UTPR — meaning the revenue stays local and the parent jurisdiction has nothing further to collect.
2. Income Inclusion Rule (IIR). Charged by the parent jurisdiction on its share of the low-taxed profit, to the extent not already absorbed by a QDMTT. Most EU UPEs apply this from fiscal years starting on or after 31 December 2023.
3. Undertaxed Profits Rule (UTPR). The backstop. Where no IIR applies (typically when the UPE is in a non-implementing jurisdiction), low-taxed profit is allocated to UTPR jurisdictions in proportion to tangible assets and employees. Effective for most adopters from fiscal years starting on or after 31 December 2024.

For a mid-market group, the practical implication is that almost every meaningful planning decision happens at the QDMTT level. If your low-taxed entity is in a jurisdiction with a qualified QDMTT (UAE, Switzerland, Hong Kong, Bermuda, the Channel Islands, Barbados, and a growing list of others), the top-up is collected there. The IIR and UTPR are residuals. Treating the IIR as the first-order rule — as much of the early literature still does — is a leftover from 2022 modeling and now distorts decision-making.

The Transitional CbCR Safe Harbour: the lifeline, and the cliff edge

The Transitional CbCR Safe Harbour, available for fiscal years beginning on or before 31 December 2026 and ending on or before 30 June 2028, removes the obligation to perform a full GloBE calculation in a jurisdiction if any one of three tests is met:

• De minimis test.The jurisdiction's total revenue and profit before tax in the qualified CbCR are below €10m and €1m respectively.
• Simplified ETR test. The simplified ETR (computed using qualified financial accounts, not GloBE income) is at least 15% for fiscal year 2024, 16% for 2025, and 17% for 2026.
• Routine profits test. Profit before tax does not exceed the substance-based income exclusion amount.

The safe harbour is the single most useful operational tool a mid-market group has, but two cautions are essential. First, the data must come from qualifiedCbCR — meaning prepared from the group's consolidated financial statements or from local statutory accounts, not from management reporting. Many groups discover during the compliance cycle that their CbCR has historically been built off management data and is therefore not safe-harbour-eligible without rework. Second, the safe harbour ends after fiscal 2026. Groups that lean on it in 2024 and 2025 must still build the full GloBE engine for 2027 onwards; treat the safe harbour as buying time for implementation, not as a permanent answer.

The Substance-Based Income Exclusion (SBIE) — small today, real over time

The SBIE allows a group to exclude a routine return on tangible assets and payroll from GloBE income before the top-up calculation. The transitional rates for fiscal 2026 are 7.6% on payroll and 7.4% on the carrying value of tangible assets, declining to a steady-state 5% on each by 2033. For services and IP-heavy groups, the SBIE is rarely material. For groups with real factories and large payrolls, it can shift a jurisdiction's top-up amount by tens of basis points of the consolidated ETR.

Two operational points are worth flagging:

• Tangible assets are valued on net book value, not gross. Aggressive depreciation policies erode the SBIE; groups with a heavy capex footprint should make sure the Pillar Two team is consulted before any depreciation policy review.
• Payroll includes independent contractors who work under the day-to-day direction of the entity. A US-style 1099 model, common in mid-market technology subsidiaries, requires careful tagging in the HRIS to capture this correctly.

Where the work actually is: data, not policy

A frequent misconception in the boardroom is that Pillar Two is a tax policy problem. It is not. By the time the rules apply to a mid-market group, the policy is already settled. The real project is a data project, and it has four pillars:

1. Consolidation-pack mapping. Each of the ~60 GloBE adjustments has to be sourced from a specific line in the consolidation pack, traced back to the local GL, and reconciled. Most groups discover that 4–6 of the adjustments cannot be sourced cleanly and require a separate data feed.
2. Tax provisioning rebuild. Deferred tax for GloBE purposes follows special rules (5-year recapture, 15% cap on deferred tax assets). Existing IFRS / US GAAP provisioning logic does not produce GloBE deferred tax without a parallel calculation.
3. Entity-level data quality.The GloBE Information Return requires constituent-entity-level data. Groups with weak intercompany reconciliation routinely produce returns that fail the OECD's standardized validation tests on first submission.
4. Exchange rate and translation discipline. The GloBE rules prescribe how each adjustment is translated. Inconsistent application across jurisdictions is the most common reason an otherwise compliant return is rejected.

Tax leadership tends to underestimate items 1 and 3. Finance leadership tends to underestimate items 2 and 4. The successful programmes we see put a single accountable owner across both functions and spend the first ninety days on data plumbing, not on calculation.

A realistic 2026 timeline

For a calendar-year group that crosses the threshold in fiscal 2025 (in scope from fiscal 2026), here is the timeline we have seen work in practice. It is conservative.

1. Q4 2025. Confirm threshold breach. Map constituent entities. Identify which jurisdictions have a qualified QDMTT and which apply IIR / UTPR. Build the safe-harbour test workbook.
2. Q1 2026. Run safe-harbour tests on prior-year CbCR data. Identify the jurisdictions that fail any of the three tests and scope a full GloBE calculation for those.
3. Q2 2026. Reconcile the consolidation pack to the GloBE adjustments. Confirm deferred tax methodology with the external auditor. Build the GloBE Information Return template.
4. Q3 2026. First quarterly GloBE calculation. Tune SBIE inputs. Identify any QDMTT instalment obligations.
5. Q4 2026. Year-end provisioning. Run the full-year calculation. Pre-clear with auditor before the close.
6. 2027 onwards. First GIR submission (typically due 15–18 months after year-end). Move from safe-harbour reliance to full GloBE calculation as the transitional rules sunset.

What to push back on with auditors

Three points come up repeatedly in audit cycles where mid-market groups have less leverage than their Big Four-advised counterparts. They are worth knowing:

• Overstated “adjustments needed” lists.Out of ~60 listed GloBE adjustments, most groups need fewer than fifteen. If your auditor is scoping all sixty as if each were material, ask them to quantify the threshold below which they will accept “not applicable”.
• Re-deriving deferred tax from scratch. The GloBE deferred tax can be derived from existing IFRS / US GAAP deferred tax with adjustments — it does not require a parallel ledger-by-ledger rebuild for most groups.
• Conservative QDMTT positions in unsettled jurisdictions.For 2024–2025 returns, a number of jurisdictions still have QDMTT legislation that has not been peer-reviewed by the OECD. Auditors sometimes default to treating these as unqualified. That position has tax cost; it is worth challenging if the legislation is plainly modeled on the OECD template.

The bottom line

Pillar Two is not the existential rewrite the 2022 commentary made it out to be. For most mid-market groups, the consolidated ETR moves by a few tenths of a percentage point. The cost is the implementation itself: data plumbing, governance, and the board-level conversation that needs to happen before, not after, the first cycle. Start twelve months before you think you need to. Lean on the safe harbours while you build. And keep the QDMTT as your primary planning lever — the IIR and UTPR are residuals.