Pillar Two in practice: what the top-up tax means for mid-market groups

The OECD's Pillar Two rules, the Global Anti-Base Erosion, or GloBE, framework, establish a 15% minimum effective tax rate for large multinational groups. The headline is simple enough; the mechanics are not. For groups whose finance functions were built for a domestic or lightly cross-border footprint, the first year of application tends to expose an uncomfortable gap between what the rules assume you can produce and what your systems actually hold. This article sets out who is caught, how the charging mechanisms interact, and what a proportionate preparation programme looks like for a mid-market group.

Who is actually in scope

The in-scope test is a consolidated revenue threshold: EUR 750 million in at least two of the four fiscal years immediately preceding the tested year. That figure is measured at the level of the ultimate parent entity's consolidated financial statements, not at the level of any individual operating company. This is the point most frequently misread. A profitable business with, say, EUR 200 million of its own turnover can sit comfortably within scope because it is a constituent entity of a larger group, a private-equity portfolio held under a consolidating structure, a family-controlled holding group, or a subsidiary of a listed parent, whose aggregate revenue clears the threshold.

So the relevant question is rarely 'are we a large multinational?' It is 'whose consolidated accounts do we appear in, and what is the group's total revenue?' Where the parent prepares consolidated statements, that consolidation is the boundary of the group. Certain entities are excluded, governmental entities, international organisations, non-profits, pension and investment funds, and real-estate investment vehicles that are ultimate parents, but the exclusions are specific and should be confirmed rather than assumed. For most mid-market groups the practical conclusion is that scope is determined above them, and they will need to comply regardless of their own size.

Three charging mechanisms, in a deliberate order

Pillar Two collects top-up tax through three mechanisms, and the order in which they apply matters a great deal to where the liability actually lands. The first is the Qualified Domestic Minimum Top-up Tax (QDMTT). A jurisdiction that has adopted a QDMTT collects any top-up on locally under-taxed profits itself, before any other country can reach them. Because a QDMTT ranks first, its adoption across most implementing jurisdictions means the tax is frequently paid at home rather than exported, which is convenient conceptually but does nothing to reduce the compliance obligation.

The second is the Income Inclusion Rule (IIR), which operates top-down: the ultimate parent (or an intermediate parent, where the structure requires it) brings into charge the top-up tax on low-taxed constituent entities elsewhere in the group. The third is the Undertaxed Profits Rule (UTPR), a backstop that reallocates any residual top-up, typically where a low-taxed entity sits under a parent in a jurisdiction that has not adopted an IIR. The UTPR generally denies deductions or makes an equivalent adjustment to recover the amount. For a well-structured mid-market group the QDMTT and IIR will usually resolve the position; the UTPR is the safety net you hope not to meet, but must still model for.

The real burden is data, not tax

The recurring surprise is that the compliance effort is largely uncorrelated with the tax finally payable. A group can conclude, correctly, that it owes no top-up tax and still spend considerable effort proving it. The GloBE effective tax rate is computed on a jurisdictional basis, blending all constituent entities in a country, using GloBE income and covered taxes derived from, but not identical to, the accounting figures. A series of prescribed adjustments sits between the trial balance and the GloBE numbers: the treatment of deferred tax at a recast 15% rate, the substance-based income exclusion for payroll and tangible assets, adjustments for stock-based compensation, purchase-accounting effects, and the allocation of taxes such as controlled-foreign-company charges.

That analysis then has to be reported on the GloBE Information Return, a standardised, data-intensive filing running to hundreds of data points per jurisdiction. Much of what it demands, entity-by-entity covered taxes, deferred tax movements analysed by category, payroll and asset carrying values by location, is not held in a form that consolidation systems readily surrender. Filing deadlines extend to fifteen months after year-end (eighteen months in a transition year), which sounds generous until the first attempt reveals how much of the underlying data must be sourced, reconciled and, in some cases, created from scratch.

Transitional CbCR safe harbours: relief, if you qualify

The transitional Country-by-Country Reporting safe harbour is the single most valuable tool for reducing early-year effort, and it should be assessed jurisdiction by jurisdiction as the first analytical step. Where a jurisdiction satisfies one of three tests, a de minimis test (revenue below EUR 10 million and profit below EUR 1 million), a simplified effective-tax-rate test (meeting a rate that rises from 15% to 16% and then 17% across the transition period), or a routine-profits test (profit at or below the substance-based income exclusion), the top-up tax for that jurisdiction is treated as nil, and the full GloBE calculation is not required for it.

Two cautions apply. First, the safe harbour relies on qualifying Country-by-Country Report data, so the quality and consistency of your CbCR filings now directly affects your Pillar Two position; sloppy or inconsistent CbCR is no longer merely a transfer-pricing risk. Second, the relief is transitional, it applies to fiscal years beginning on or before 31 December 2026, and does not apply to any fiscal year that ends after 30 June 2028 (broadly, the first three years). A group that leans entirely on the safe harbour without building the underlying capability is deferring the problem, not solving it. The sensible posture is to use the safe harbour to buy time while standing up the full calculation in parallel.

What to do now

A proportionate programme has a clear sequence. Confirm scope by reference to the ultimate parent's consolidated revenue and identify every constituent entity and its jurisdiction. Run the transitional safe harbour test across all jurisdictions to isolate those needing full computation. Perform a data-gap assessment against the GloBE Information Return's requirements, focusing on deferred tax, covered taxes and substance-based inputs. Establish which mechanism, QDMTT, IIR or UTPR, governs each part of the group, and confirm filing and any domestic registration obligations, which vary by jurisdiction and sometimes fall due independently of the return itself. None of this is intractable, but it rewards starting early and treating Pillar Two as a data and process project rather than a year-end tax calculation.

Key takeaways

• Scope turns on the ultimate parent's EUR 750m consolidated revenue in two of the prior four years, so mid-market businesses are routinely caught through group membership rather than their own size.
• QDMTT applies first (tax paid locally), the IIR second (top-down at the parent), and the UTPR is the backstop, the ordering determines where any liability lands and must be mapped per jurisdiction.
• The dominant cost is data and the GloBE Information Return, not the tax itself; groups often owe nil yet still face substantial computation and reporting effort.
• Use the transitional CbCR safe harbours to reduce early-year work, but note they rely on clean CbCR data and apply only to fiscal years beginning on or before 31 December 2026 (and to no year ending after 30 June 2028), build the full capability in parallel.