This contribution, based on a presentation at the GLOBTAXGOV- St. Gallen University Roundtable of 10 June 2026 on the Inclusive Framework Side-by-Side (SBS) package of 5 January 2026, argues that Pillar Two has a “Janus-faced” character. Rather than imposing a floor on tax competition, Pillar Two restructures it through mechanisms such as QRTCs and QTIs and the recalibration of the effective tax rate calculation, enabling continued, system-aligned competition while producing global asymmetries. At the same time, the incorporation of the SBS package via Article 32 of the EU Pillar Two Directive creates a constitutional Catch-22, raising concerns under EU law regarding delegation, State aid, and the autonomy of EU law. Pillar Two accordingly emerges as a functional but legally and politically fragile framework.
1. Introduction
Professor Irma Mosquera, thank you for the invitation to deliver this introductory contribution here today at the GLOBTAXGOV Roundtable of 10 June 2026 on “The Side-by-Side Agreement – Law, Politics, and the Future of Global Tax,” and in particular on the issues surrounding the newly introduced Safe Harbour for Qualified Tax Incentives (QTIs) on the one hand and Article 32 of the EU Pillar Two Directive on the other.
Today, I have the opportunity to discuss two developments that at first sight appear technical, but which – in my view – reveal a feature of the current direction of international company taxation in the Pillar Two context. On the one hand, this concerns the treatment of Qualified Tax Incentives (QTIs) under the Inclusive Framework’s Side-by-Side (SBS) package of 5 January 2026, and on the other hand Article 32 of the EU Pillar Two Directive and the way in which the implementation of the SBS package on this basis is institutionally embedded within the European Union.
My central proposition is that both developments reveal the dual character of Pillar Two, both in economic-political and in EU-constitutional terms. More sharply, Pillar Two does not so much limit tax competition as it restructures it – and it does so in a way that is simultaneously functional in the short term and constitutionally uncomfortable in the longer term. What at first glance appears to be a uniform global minimum tax standard turns out in practice to encompass a complex system of redistribution, recalibration, and re-labelling of tax incentives, which implementing jurisdictions can mobilise in structuring their competitive positioning. At its core, Pillar Two must now be understood as a system that simultaneously restructures tax competition and obscures the locus of normative authority in the EU context.
2. The Restructuring of Tax Competition under Pillar Two
2.1 QTIs and the Economic-Political Janus Face
Pillar Two was presented in 2021 as a hard floor of 15% effective company taxation and as an instrument that would substantially curb tax competition through effective corporate tax rates. Yet, today we see that such competition has not disappeared. Rather, it has been displaced – and, more importantly, it turns out to be partly embedded in the Pillar Two system itself.
The label ‘economic-political’ in the heading, notably, is not accidental. What appears to be a technical recalibration of tax incentives simultaneously reflects political choices about where and how competition for corporate investment via fiscal interventions is permitted to persist within a global minimum tax framework.
This becomes visible, amongst others, in the treatment of Qualified Refundable Tax Credits (QRTCs). QRTCs effectively reduce the actual corporate tax burden, but within the Pillar Two framework their effect is mitigated, as the effective tax rate (ETR), as calculated for Pillar Two purposes, does not decline proportionally.
The adjustment operates through the denominator of the Pillar Two ETR calculation. This is not merely a technical detail. It reshapes what the system measures and, therefore, what it allows. It implies that the system does not merely measure tax burdens but restructures them through the design of its metric. At the same time, QRTCs introduce a hybrid category in which tax instruments and subsidy-like mechanisms converge, blurring the distinction between tax incentives and direct subsidies. Incentives are not eliminated, rather, they are repackaged within the fiscal domain.
With the introduction of QTIs, this development is further extended. QTIs may be understood as targeted tax incentives – expenditure-based or production-based – that are structured such that their effect is neutralised within the Pillar Two ETR calculation. Operating within capped, substance-based thresholds under the Substance-based Tax Incentive Safe Harbour, these incentives allow a reduction in the actual tax burden without a corresponding reduction in the measured ETR. Whereas QRTCs still operate within the metric, QTIs effectively operate at the level of defining the metric itself.
This move from mitigation via QRTCs to neutralisation via QTIs is crucial. The Pillar Two ETR becomes not merely a measurement tool. It determines which forms of tax relief are visible within the system and which are rendered invisible.
A similar logic appears in the interpretation of the no benefits rule in the Pillar Two system. This rule seeks to prevent jurisdictions from offsetting Pillar Two top-up taxes through compensatory measures, such as tax incentives, subsidies or grants. Yet the Pillar Two system simultaneously accommodates economically equivalent incentives, provided they are embedded within, or filtered through, the design of the metric itself. This creates selective constraint. Some forms of competition are curtailed, while others are restructured and permitted.
Importantly, this tension extends beyond technical design. The interpretation of the no benefits rule, and with that its scope, remains subject to ongoing discussion – it seems – within the Inclusive Framework, developed within a soft-law environment characterised by political negotiation and evolving administrative practice. The result is a system that is mediated not only through formal rules, but also through interpretative practice. Seen in this light, QTIs, QRTCs, and the no benefits rule form part of the same dynamic. Pillar Two seeks to restrict certain forms of competition while redistributing others.
In ongoing research with Irma Mosquera, we describe this as the economic-political Janus face of Pillar Two.[1] Externally, the Pillar Two system presents itself as a minimum tax standard. Internally, it creates space for forms of competition that are no longer visible within the Pillar Two metric. Pillar Two thus functions less as a constraint on tax competition than as a mechanism for its reallocation and recoding.
This development has significant implications, particularly for developing countries. Traditional tools to attract investment, such as tax holidays, tax treaty-based relief, and certain investment treaty-linked stabilisation expectations, become less effective under Pillar Two. These jurisdictions are now increasingly compelled to increase effective tax levels while reintroducing incentives through credits and subsidies, leading to complicated “cash carousel” dynamics. These mechanisms require substantial administrative capacity, thereby creating asymmetries between developed and developing jurisdictions, as well as between administrative “capacity haves” and “capacity have-nots” – favouring the first over the latter.
Similar asymmetries arise geopolitically. Under the SBS framework, the United States retains greater flexibility in designing incentive structures, including through the continued application of its own minimum tax regime without being subject to the same structural limitations on tax incentives that apply under Pillar Two. This reflects underlying geopolitical power dynamics. What emerges is thus a politically contingent distribution of tax competition space.
Within the European Union, a comparable dynamic emerges. This is illustrated, inter alia, by the recently leaked European Commission (draft) Omnibus package. The (draft) proposed tax incentives in that package are simultaneously framed as instruments of competitiveness and as components of a coordinated anti-tax avoidance architecture, requiring alignment with the Pillar Two framework to prevent exposure to top-up taxation. In this sense, Member States’ company tax policy space becomes structurally endogenous to an overarching international minimum taxation order. Autonomy is not eliminated, but reconstituted. It persists only insofar as domestic policy choices are rendered compatible with computational logics of effective taxation under the global minimum tax framework.
2.2 Article 32 and the Constitutional Janus Face
Where QTIs demonstrate how Pillar Two restructures tax competition economically, Article 32 reveals a parallel development at the constitutional level. The directive provision obscures the locus of normative authority. It becomes increasingly difficult to determine where the essential elements of the Pillar Two system are determined and to whom they are legally attributable.
Article 32 functions as a channel through which externally agreed safe harbour rules developed within the OECD Inclusive Framework are given legal effect within EU law. Yet this is not a simple act of implementation. It introduces a form of circular or back-delegation. Authority is first centralised at EU level and subsequently re-externalised through Member State consensus outside the EU’s institutional framework.
In joint work with Sam van der Vlugt, building on some academic work of Dennis Weber, we describe the implications of this as a Catch-22.[2] The central issue is the legal qualification of the SBS framework within EU law. Whether it is characterised as Union law, Member State action, or external norm reception determines the applicable constitutional constraints.
If SBS-implementation in the EU is treated as a Union measure, serious concerns arise under the Court of Justice’s Meroni doctrine, as essential elements of taxation would effectively be shaped outside the EU legislative framework. This would involve a delegation of normative authority without the institutional safeguards required under EU law, which the Court of Justice does not permit. If SBS-implementation is instead attributed to the Member States, the analysis shifts to EU State aid law. The Side-by-Side safe harbour may then constitute a selective advantage attributable to Member States, potentially falling within Article 107 TFEU.
The Catch-22 lies in this duality. Any qualification triggers a different body of law, yet both lead to constitutional friction. There is no stable intermediate position. Article 32 thus reflects the constitutional Janus face of Pillar Two. It embodies an external coordination project that depends on an unstable allocation of normative authority between the Union and the Member States. This raises deeper concerns regarding legality of taxation and democratic accountability, as essential elements of the minimum tax system are effectively shaped outside parliamentary processes.
These tensions also engage the principle of autonomy of the EU legal order. To the extent that externally developed rules are incorporated without constitutional safeguards, EU constitutional questions arise akin to those addressed in the Court of Justice’s Kadi case law. Judicial scrutiny, at some point, is therefore unavoidable. Questions about the legal status of the SBS package and its safe harbours, the relevance of Pillar Two instruments under the EU Pillar Two Directive now potentially neutralised under the SBS package, and the legal standing of affected parties before the Court of Justice depend on prior constitutional classification of the SBS package.
Political discourse suggests that these issues have been resolved, for instance through the Commission notice of 12 January 2026 confirming the implementation of the SBS package via Article 32 of the EU Pillar Two Directive. Yet such statements from the Commission are non-binding and cannot determine the legal status of the SBS arrangement. That authority remains with the Court of Justice. The reassurance offered at the political level is therefore constitutionally fragile.
3. Concluding Remarks
Taken together, a clear pattern emerges. Where the SBS package’s QTIs reveal the economic-political Janus face of Pillar Two, the narrative on implementing the SBS package via Article 32 exposes its constitutional Janus face. Where QTIs obscure the true incidence of tax competition by filtering what counts within the Pillar Two metric, Article 32 obscures the locus of normative authority by filtering where the system’s essential rules are deemed to originate.
Pillar Two does not eliminate tax competition but restructures it. It does so by layering multiple objectives, namely global minimum taxation, investment promotion through specified tax incentives, international tax coordination, and jurisdictional competitiveness. These objectives are not reconciled, but managed through increasingly complex design choices that give rise to an ambivalent legal status.
The result is a system that functions, but under conditions that are politically contingent and legally fragile. A minimum tax model that depends on exceptions, reinterpretations, and soft-law developments is inherently under pressure. The question, therefore, is not only whether Pillar Two works, but whether it can remain a sustainable and credible framework in the face of these structural tensions.
[1] Maarten de Wilde and Irma Mosquera Valderrama, ‘Corporate tax and investment incentive policies post-Global Minimum Tax’, forthcoming.
[2] Maarten de Wilde and Sam van der Vlugt, ‘The Side-by-Side Arrangement under EU Law: A Constitutional Catch-22’, forthcoming in European Taxation.