## Deciphering the GloBE in a Low-Tax Jurisdiction — Working Paper No. WP/2024/064

## Source details

**Canonical URL:** [Deciphering the GloBE in a Low-Tax Jurisdiction — Working Paper No. WP/2024/064](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024064-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024064-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024064-print-pdf.pdf.json)

---

### Introduction and overview
- Central message:
  - The QDMTT is not the solitary reaction; ideally it is a backstop/top-up tax rather than the general tax system. A general tax system should target only economic rent, but the GloBE rules have key implications for choices of tax rate and base.
- Focus:
  - Implications of (and possible reactions to) Pillar Two from the standpoint of a zero- or low-tax jurisdiction.
- Key questions:
  - Can the minimum corporate tax rate be set so that all countries, including low-tax jurisdictions, are better off?
  - What responses (to a binding minimum tax) can zero- or low-tax jurisdictions consider given the specificities of the Pillar Two rules?
- Key observations:
  - The QDMTT gives the source country the right to collect any minimum tax, but legal design and policy flexibility issues make adoption decisions non-trivial—especially where initially there is no corporate income tax (CIT).
  - There is an opportunity to endorse a broader tax reform that can be beneficial for low-tax jurisdictions, subject to design constraints implied by the Pillar Two rules.
- Illustrative empirical note preserved:
  - Tørsløv et al. (2023) estimate that 36 percent of multinational profits (around $600 billion) are shifted to low-tax jurisdictions in 2015.

### Where we are right now: profit shifting, Pillar Two mechanics, and revenue impacts
- Existing problems:
  - Separate accounting and the arm's length principle; tax treaties allocating taxing rights.
  - Arm's length principle increasingly complex and difficult to enforce.
  - Outcomes: (i) prevalence of shifting of profits by multinationals to low-tax jurisdictions; (ii) tax competition between countries over profits and investment.
- Pillar Two adoption and scope:
  - 140 subscribing jurisdictions as of November 2023.
  - Applies to multinational groups with global revenues above €750 million, with some exceptions.
- GloBE effective rate and top-up tax mechanics:
  - GloBE effective rate = ratio of ‘covered taxes’ to GloBE income.
  - GloBE income = accounting net income (or loss), adjusted for specific items.
  - If GloBE effective rate < 15 percent, a top-up tax applies.
  - In a zero-CIT country, the GloBE effective rate for any in-scope entity is zero.
  - Top-up tax calculation example preserved:
    - Suppose the covered tax is zero, the GloBE income is 100, and the SBIE is 20. The resulting top-up tax is 12 (that is, the effective rate on excess profit is 15 percent but the average paid tax is 12 percent).
- Allocation of top-up tax:
  - If the source country adopts a QDMTT, the top-up tax will be collected by the source country.
  - Otherwise, the headquarters country collects the top-up tax through an IIR.
  - If neither rule is implemented or fully captures the top-up tax, other countries along the chain can collect via the UTPR.
- Revenue impact estimates preserved:
  - IMF (2022) earlier estimate: 5.7 percent (global impact).
  - OECD (2024) estimate: between 6.5 percent and 8.1 percent of the global CIT revenues.
  - Distribution of global revenue gain depends on country responses and multinationals’ responses.

### Theoretical prediction: can low-tax countries be better off?
- Conceptual setup:
  - Consider a binding minimum corporate tax with a rate set between initial unconstrained equilibrium rates of high-tax and low-tax countries.
- For high-tax countries:
  - Higher tax abroad protects domestic tax base and increases revenues but may lower shareholder incomes via denial of profit-shifting.
  - Presumption: reduction in profit shifting will dominate and high-tax countries will be better off under the minimum tax.
- Effects on low-tax countries (enumerated):
  - Profit shifting channel:
    - Raising the low-tax rate to the minimum reduces profits received from abroad (negative welfare impact).
    - Offsetting channels: higher rate raises revenue from immobile domestic base and from remaining profit received from abroad.
    - Reaction of high-tax countries is crucial: if high-tax countries raise their rates (by less than one-to-one), fall in profit shifting to low-tax jurisdictions is muted.
    - Hebous and Keen (2023) show a Pareto-improving efficient minimum rate exists; calibration suggests this Pareto-minimum rate can reasonably be in the territory of 17 percent (given an initial Nash equilibrium of 12.5 percent).
    - In this class of models, a zero-tax is not an equilibrium.
  - Investment channel:
    - If the low-tax jurisdiction is a capital exporter, higher tax abroad could reduce global demand for capital and returns earned on world markets.
    - SBIE can increase return of investments in low-tax jurisdictions: where profit is high and assets and payroll are low (below the SBIE), firms can increase after-tax profit by investing to max out SBIE.
  - Other effects:
    - Competition continues over out-of-scope multinationals and non-‘excess’ profits and SBIE.
    - Low-tax countries could reduce enforcement efforts under a minimum corporate tax.
  - Empirical behavioral response preserved:
    - IMF (2022) estimates that a 1-percentage-point change in the average foreign statutory tax rate leads the home rate to change between 0.25-0.4 percentage points in the same direction.

### Low-tax jurisdiction perspective on GloBE: adoption choices and policy window
- Adoption is optional, but jurisdictions must accept adoption by others; low-tax jurisdictions decide whether and how to design CIT reform in response to GloBE rules implemented by investment partners.
- Implementation timing and policy window:
  - Implementation progressing rapidly; entering into effect for some in 2024 in a few important capital exporting countries (notably EU and UK), while as at end of 2023 some delay emerged in adoption in several countries.
  - Delay allows many low-tax jurisdictions time to design and implement general CIT reforms, taking advantage of features of the minimum tax (treatment of certain immediate expensing and refundable tax credit regimes) before—or alongside—specific GloBE rules.

### Box 1 — Implementation timing and strategic considerations (selected jurisdictions)
- Key adoption timings and notes (as reported):
  - Singapore: IIR 2025; QDMTT 2025; UTPR 2025
  - Cyprus: IIR 2024; QDMTT 2025; UTPR 2025
  - Netherlands: IIR 2024; QDMTT 2024; UTPR 2025
  - EU: IIR 2024; QDMTT optional; UTPR 2025
    - Must transpose EU Directive end 2023 with IIR and UTPR (but can defer application to end 2029 if no more than 12 ultimate parent entities; elected by Estonia, Latvia, Lithuania, Malta, and Slovakia).
  - Switzerland: IIR Delayed; QDMTT 2024; UTPR Delayed
  - United Kingdom: IIR 2024; QDMTT 2024; UTPR 2025
  - United States: No legislative plans for IIR, QDMTT, or UTPR
    - US GILTI regime qualifies as a Blended CFC Tax Regime under GloBE rules.
    - A domestic corporate alternative minimum tax (at 15%) on the adjusted financial statement income of companies with profits > US$1 billion adopted.
- Selected low-tax jurisdictions without pre-existing CIT regimes (selected items preserved):
  - Barbados
    - No public IIR plans
    - QDMTT plans (2024, conditional on top-up tax payable elsewhere)
    - No public UTPR plans
    - General CIT: Proposal for a general CIT (at 9%) effective 2024, with QDMTT (at 15%) for multinational groups with revenues of at least €750 million that are subject to an IIR or UTPR elsewhere. A qualified refundable tax credit (QRTC) is also proposed (jobs credit based on payroll plus R&D credit).
  - Bermuda
    - No public IIR plans
    - No specific QDMTT plans
    - No public UTPR plans
    - General CIT: Adopted general CIT (at 15%) for multinational groups with revenues of €750 million or more (2025). Final law removed credit for U.S. GILTI but has temporary U.S. CFC income exclusion instead. QRTCs also proposed.
  - Guernsey: IIR 2025; QDMTT 2025; No public UTPR plans
  - Jersey: IIR 2025; QDMTT 2025; No public UTPR plans
  - Isle of Man: IIR 2025; QDMTT 2025; No public UTPR plans
  - UAE
    - No public IIR plans
    - QDMTT delayed (2025)
    - No public UTPR plans
    - General CIT: Adopted CIT (at 9%) on annual taxable profits above AED 375,000 (effective 2023), with delayed adoption of a QDMTT (at 15%).
  - Kuwait
    - No public IIR plans
    - No public QDMTT plans
    - No public UTPR plans
    - General CIT: Proposal to adopt CIT (at 15%) from 2025, without specific GloBE rules.
- Strategic considerations summarized:
  - Adopting the QDMTT is a dominant strategy on revenue grounds: ensures the country can collect top-up tax payable by in-scope multinationals with affiliates in the low-tax country.
  - QDMTT per se would not put a jurisdiction at an investment disadvantage because wholly owned in-scope companies will in any case be taxed at 15 percent effective tax if any parent in the chain implements Pillar Two.
  - No added compliance costs for the jurisdiction because in-scope multinationals must undertake the calculations irrespective of the jurisdiction’s policy.
  - Adopting the IIR is also likely: ensures revenue collection if other jurisdictions do not implement a QDMTT and renders CFC rules redundant for in-scope multinationals.
  - The UTPR case is currently weak for low-tax jurisdictions because it is a provision of last resort, raises treaty and administrative complications, and may create trade-retaliation risks.
  - On balance, low-tax jurisdictions may deprioritize UTPR implementation and focus on more pressing CIT issues.

### Revenue and design mechanics — illustrative comparisons preserved
- Calibrating a domestic tax to collect the exact top-up tax is challenging; countries should not be fixated on doing so.
- Numerical illustrations preserved:
  - If net GloBE income is $100 and SBIE is $20:
    - General CIT at 15 percent on net GloBE income ($100) yields tax of $15.
    - The top-up tax collected elsewhere under either the IIR or the UTPR would be $12 (assuming net GloBE income of $100 less the SBIE of $20).
    - A general CIT at 15 percent on GloBE ‘excess profits’ ($12) after considering the SBIE would result in a rate below 15 percent (actually, 12 percent), and therefore would still require a top-up tax elsewhere.
    - A QDMTT applied to in-scope constituent entities on their entire GloBE ‘excess profits’ of $12 would apply irrespective of ownership; under the IIR the top-up tax payable elsewhere would be adjusted for ownership interests less than 100 percent. For an 80 percent owned entity, that implies $9.6.
  - Interaction with CFC rules example preserved: Company Y profit 100; Company V’s tax (at 21%, before credit) in A is $21; Company V can use credits under a CFC regime with global blending of $6 relating to another high tax jurisdiction. After credit, net tax is $15, which can be pushed down to Jurisdiction C, making C’s top-up tax zero, whereas a QDMTT in C would have yielded additional tax of $15 on top of the CFC tax in A. Company V saves $6 under global blending by paying $15 instead of $21 in A.
- Rationales for going beyond GloBE (three broad reasons preserved):
  1. Tax and economic policy: opportunity to diversify and raise revenue, including taxing economic rent of out-of-scope companies with appropriate tax design.
  2. Legal design: QDMTT and GloBE are add-ons requiring an underlying CIT; standalone QDMTT would need to be comprehensive and may raise treaty challenges.
  3. Avoiding STTR: STTR allows low-income countries to impose a top-up tax on specific cross-border payments if the foreign nominal CIT rate < 9 percent; to circumvent STTR a country needs some CIT in addition to the GloBE.
- Additional observations for zero-tax jurisdictions:
  - Lack of personal income taxes (PITs) complicates broad taxation of profits; neutrality across income sources and legal forms is difficult without PIT.
  - Choice of threshold for any profit tax matters (example VAT threshold noted at USD 100).
  - Niche responses targeting a few existing firms may be fragile; a proper underlying CIT is preferable.
  - Bermuda example: adopted CIT (rate 15%) with features interacting with U.S. tax rules illustrating design complexity.

### An efficient rent tax under GloBE: ACE versus cash-flow (R-based) taxation
- Preferred objective:
  - Economists prefer an efficient profit tax that targets only economic rent, implying a zero marginal effective tax rate (METR).
- Two broad designs:
  - Cash-flow taxation (R–based cash-flow): immediate expensing; no deduction for interest expense or returns to equity.
  - Allowance for Corporate Equity (ACE): notional deduction for a normal return while maintaining depreciation rules and interest deductions.
- GloBE treatment differences:
  - Immediate expensing is treated as a “temporary timing measure” and does not lower the GloBE effective tax rate.
  - ACE’s notional deduction lowers the covered tax rate under GloBE.
- Key implications (preserved technical statements and formulas):
  - Implication 1: The ACE is no longer a tax only on economic rent for sufficiently low statutory CIT rates (still possibly well above 15 percent). The ACE will imply a top-up tax on in-scope companies even if the investment is just earning the normal return.
  - Decomposition (assumptions: QDMTT in place, ACE as NQRTC, single period model):
    - (i) For investments that earn the normal return (or less), the ACE will result in a zero covered tax. Therefore, only the QDMTT matters. If the ACE is an NQRTC, the top-up rate is the full 15 percent, applied on profits minus SBIE.
    - (ii) For investments that earn economic rent, the ACE implies a top-up tax on in-scope companies if the statutory CIT rate is sufficiently low. The cutoff statutory CIT rate (휏휏) is given by:
      - 휏휏 = 15% × (1 − (푟푟.퐸퐸 / 휋휋))
    - Example: if the notional deduction is 8 percent and the equity-to-profit ratio is 5, a statutory tax rate of 25 percent (or above) is required for preventing the application of the QDMTT.
    - (iii) For projects with no economic rent, the average tax rate depends only on SBIE. If SBIE ≈ 휋휋, average tax rate ≈ zero. If SBIE → 0, average tax rate (even on normal return) approaches the minimum tax.
    - (iv) For projects with economic rent and no top-up tax (statutory rate high enough), average tax rate equals:
      - 퐴퐴퐴퐴퐴퐴 = (푇푇푇푇푇푇 푝푝푇푇푝푝푝푝) / 휋휋 = 휏휏 × (1 − (푟푟.퐸퐸 / 휋휋))
    - (v) For projects with economic rent and a top-up tax (statutory rate low enough), average tax rate is:
      - 퐴퐴퐴퐴퐴퐴 = (퐴퐴퐴퐴) / 휋휋 = 15% × (1 − (푆푆푆푆푆푆 퐸퐸 / 휋휋)) + 휏휏 × (1 − (푟푟.퐸퐸 / 휋휋)) × (푆푆푆푆푆푆 퐸퐸 / 휋휋)
    - Note: If economic rent is low (allowances of 8 or 10 percent), average tax rate is close to 15 percent.
  - Recap: ACE+QDMTT does not guarantee no-taxation of normal returns. The top-up rate for a project yielding the normal return is always 15 percent. ACE becomes a tax on economic rent only for sufficiently high statutory tax rates (single-period thresholds can be around 25 percent).
- Immediate expensing vs ACE under GloBE:
  - Immediate expensing treated as “temporary timing measure” (Article 4.4 of the Model Rules) and by itself does not trigger a top-up tax.
  - Domestic cash-flow tax with full expensing and same statutory rate (for example 15 percent) implies a zero METR if losses are refundable or carried forward with interest.
  - Implication 2: Given a relatively low statutory tax rate, under the GloBE, immediate expensing (R-based cash-flow tax) tends to result in a lower METR than the ACE, ceteris paribus.
  - Three cases (assumptions: QDMTT in place, ACE as NQRTC, period-by-period):
    - Case 1: No top-up tax under either ACE or immediate expensing — occurs only if statutory rate sufficiently high and investment yields economic rent.
    - Case 2: Top-up tax under ACE but not under cash-flow tax — occurs for statutory rates between 15 percent and the upper level given by 휏휏.
    - Case 3: Top-up taxes under both systems (휏휏 < 15%) — top-up tax always higher under ACE than under immediate expensing.
  - Note: In dynamic models refundability of tax losses matters; ACE can in some dynamic settings yield lower METR than cash-flow taxes for some investments.
- Why not other cash-flow taxes:
  - Banks under pure R-based cash-flow tax: interest income untaxed → banks subject to QDMTT.
  - R+F–based cash-flow: administratively difficult.
  - S–based cash-flow: Pillar Two treats new S-based systems unequally; eligible distribution systems must be in force on or before 1 July 2021 and meet conditions.
  - Financial Activity Tax approximates excess profit plus remuneration of financial institutions; if remuneration excluded and statutory rate low per 휏휏, becomes ACE-like and QDMTT binds.

### Conclusions and policy implications (concise bullets)
- Policy design objectives:
  - Protection from top-up tax payable elsewhere should be an explicit tax design objective.
  - Adopting the QDMTT (and possibly the IIR) is part of that protection; jurisdictions should adopt the GloBE as a backstop on top of a well-designed profit tax.
- Legal implementation suggestions:
  - Foundational legal infrastructure: core definitions, collection and enforcement provisions.
  - Include customary international tax and anti-avoidance provisions: functional transfer pricing rules, economic substance rules for foreign source income exemptions.
  - Implement GloBE top-up tax rules as a backstop closely following the Model Rules and possibly enacted through a separate legal instrument leveraging foundational profit tax infrastructure.
- Key implications preserved:
  - The normal return can be taxed even under an efficient rent tax design, especially at relatively low statutory tax rates (possibly above 15 percent), unless SBIE ≥ normal return for all marginal investors each year.
  - GloBE breaks the equivalence among economic rent tax designs, tilting toward the R-based cash-flow tax because immediate expensing does not lower the covered tax and thus preserves a zero METR.
  - Under ACE, an in-scope company in any year may face a non-zero top-up rate unless statutory tax rate or SBIE is sufficiently high (single-period formula suggests thresholds can be around 25 percent).
  - Under an R-based cash-flow tax, banks will end up under QDMTT in a pure R-based design.
  - New S-based cash-flow regimes are constrained by Pillar Two eligibility rules.
  - Policy takeaway: Zero- or low-tax jurisdictions can design profit taxes to ensure zero tax on investment at the margin; generally this requires a rate of at least 15 percent. Effective tax on economic rent remains moderate and well below 15 percent. Even with efficient rent taxes around 15 percent, jurisdictions may continue to receive some foreign profits as they remain in the lower range of international profit taxation.

*Source: Deciphering the GloBE in a Low-Tax Jurisdiction, IMF Working Paper No. WP/2024/064 (wpiea2024064-print-pdf).*

### 1. Introduction

### 1. Introduction

### Overview
- The ongoing widespread adoption of a minimum effective rate of corporate tax under the Inclusive Framework agreement is changing the rules of tax competition, to some degree limiting it with important ramifications for zero-tax or low-tax jurisdictions.
- Key questions:
  - Can the minimum corporate tax rate be set so that all countries, including low-tax jurisdictions, are better off?
  - What responses (to a binding minimum tax) can zero- or low-tax jurisdictions consider given the specificities of the Pillar Two rules?
- Focus of the paper: implications of (and possible reactions to) Pillar Two from the standpoint of a zero- or low-tax jurisdiction.
- Central message: The QDMTT is not the solitary reaction; ideally it is a backstop/top-up tax rather than the general tax system. A general tax system should target only economic rent, but the GloBE rules have key implications for choices of tax rate and base.

### Key observations for jurisdictions contemplating introducing a CIT from scratch
- The QDMTT (qualified domestic minimum top-up tax) gives the source country the right to collect any minimum tax, but legal design and policy flexibility issues make adoption decisions non-trivial—especially where initially there is no corporate income tax (CIT).
- At this juncture there is an opportunity to endorse a broader tax reform that can be beneficial for low-tax jurisdictions, subject to design constraints implied by the Pillar Two rules.

### Illustrative footnote observations (preserved)
- Whether to introduce a profit tax is relevant for zero-tax jurisdictions, while the design of the tax is revenant for both zero- and low-tax jurisdictions. Note, though, even in a jurisdiction without a corporate income tax, there can be some fees or other light taxes. Such jurisdictions that attract profits from abroad typically tend to be small (Kanbur and Keen, 1993) and exhibit a stable and relatively high institutional quality (Dharmapala and Hines, 2009), among other characteristics (Dharmapala 2023).

---

### 2. Where We Are Right Now

### The problem: profit shifting and tax competition
- Existing CIT arrangements rely on:
  - Separate accounting for affiliates and the arm's length principle.
  - Tax treaties allocating taxing rights; broadly, source countries tax income from “production” (permanent establishment) and residence countries tax passive income.
- Challenges:
  - The arm's length principle has become increasingly complex and difficult to enforce given global firms, hard-to-value intangibles, and intergroup services (Hebous, 2021).
  - Tax treaties can enable treaty shopping via reduced withholding rates (Erokhin and Weichenrieder, 2023).
- Outcomes of these arrangements:
  - (i) Prevalence of shifting of profits by multinationals to low-tax jurisdictions.
  - (ii) Tax competition between countries over profits and investment.
- Empirical estimate preserved:
  - Tørsløv et al. (2023) estimate that 36 percent of multinational profits (around $600 billion) are shifted to low-tax jurisdictions in 2015.
  - The benefits from ‘paper profits’ are challenging to quantify or directly observe but manifest in large observed aggregate FDI and specialized sectors in many low-tax jurisdictions.

### Agreement on a minimum corporate tax (Pillar Two)
- Adoption and scope:
  - Discontent with profit shifting and tax competition triggered unilateral minimum taxes and stricter anti-tax avoidance rules, culminating in an agreement on the minimum corporate tax (under Pillar Two), with 140 subscribing jurisdictions as of November 2023.
  - Pillar Two minimum tax applies to multinational groups with global revenues above €750 million, with some exceptions (OECD, 2023).
- GloBE effective rate and top-up tax mechanics:
  - The GloBE effective rate is computed as the ratio of ‘covered taxes’ to GloBE income.
  - GloBE income is the accounting net income (or loss), adjusted for specific items (for example excluding intra-group dividends).
  - If the GloBE effective rate is below 15 percent, a top-up tax applies.
  - In a zero-CIT country, the GloBE effective rate for any in-scope entity is zero.
  - The resulting 15 percent top-up tax rate is multiplied by the ‘excess profit’ defined as GloBE income minus a substance-based income exclusion (SBIE) in the jurisdiction (set at 5 percent of each of tangible assets and payroll, after a transition period).
  - Numerical illustration preserved:
    - Suppose the covered tax is zero, the GloBE income is 100, and the SBIE is 20. The resulting top-up tax is 12 (that is, the effective rate on excess profit is 15 percent but the average paid tax is 12 percent).
- Allocation of the top-up tax under GloBE rules:
  - If the source country adopts a QDMTT, the top-up tax will be collected by the source country.
  - Otherwise, the headquarters country collects the top-up tax through an income inclusion rule (IIR).
  - If neither rule is implemented or fully captures the top-up tax, other countries along the chain can collect via the undertaxed profits rule (UTPR).
- Revenue impact estimates preserved:
  - Earlier estimates put the global impact at 5.7 percent (IMF, 2022).
  - OECD (2024) estimates it to be between 6.5 percent and 8.1 percent of the global CIT revenues.
  - The distribution of global revenue gain depends on country responses (adopting a minimum tax or not and possible changes to the tax base and rates) and on multinationals’ responses (profit shifting and real investment).

---

### 3. Can Low-Tax Countries Be Better Off Under a Minimum Corporate Tax? What Theory Predicts

### Conceptual setup
- Consider a binding minimum corporate tax with a rate set between initial unconstrained equilibrium rates of high-tax and low-tax countries.
- For high-tax countries:
  - Higher tax abroad protects their tax base, increasing revenues.
  - But higher tax abroad also denies profit-shifting opportunities, lowering shareholder incomes and consumption (Johannesen, 2022).
  - Net effect positive if the societal value of the revenue effect dominates; subsidy competition could offset gains (Janeba and Schjelderup, 2023).
- Presumption: reduction in profit shifting will dominate and high-tax countries will be better off under the minimum tax.

### Effects on low-tax countries (enumerated)
- Profit shifting:
  - Forcing the low-tax country to raise its rate to the minimum reduces profits received from abroad (negative welfare impact).
  - Offsetting channels:
    - Higher rate raises revenue from remaining base, which includes: (i) immobile domestic tax base; and (ii) profit received from abroad (low-tax jurisdiction remains recipient of foreign profits because of remaining tax rate differential).
  - The reaction of the high-tax country is crucial:
    - If the high-tax country raises its rate (by less than one-to-one), the fall in profit shifting to the low-tax jurisdiction is muted.
    - Hebous and Keen (2023) show a Pareto-improving efficient minimum rate exists that can make both high- and low-tax countries better off: less profit leaves the high-tax country while the low-tax country collects higher revenues from its immobile base and remaining incoming profit.
    - Calibration suggests this Pareto-minimum rate can reasonably be in the territory of 17 percent (given an initial Nash equilibrium of 12.5 percent).
    - In this class of models, a zero-tax is not an equilibrium: explicit societal benefits must exist for the zero-tax jurisdiction from attracting profits from abroad.
- Investment:
  - If the low-tax jurisdiction is a capital exporter, a higher tax rate abroad could reduce global demand for capital and the return it earns on world markets (Keen and Konrad, 2013).
  - The SBIE can increase the return of investments in the low-tax jurisdictions (Schjelderup and Stähler, 2023): where profit is high and assets and payroll are low (below the SBIE), firms can increase after-tax profit by investing to max out SBIE.
- Other theoretical effects:
  - Competition continues over out-of-scope multinationals and over non-‘excess’ profits and SBIE; expanding coverage of in-scope multinationals is superior to raising the minimum rate (Haufler and Kato, forthcoming).
  - Tax enforcement incentives may change; low-tax countries could reduce enforcement efforts under a minimum corporate tax (Hindriks and Nishimura, 2022).
- Empirical behavioral response preserved:
  - IMF (2022) estimates that a 1-percentage-point change in the average foreign statutory tax rate leads the home rate to change between 0.25-0.4 percentage points in the same direction.

---

### 4. A Low-Tax Jurisdiction Perspective on the GloBE

### Should low-tax jurisdictions adopt the GloBE?
- Adoption is optional, but jurisdictions must accept adoption by others; low-tax jurisdictions decide whether and how to design CIT reform in response to GloBE rules implemented by investment partners.
- Implementation timing:
  - Implementation of the minimum tax is progressing rapidly, entering into effect for some in 2024 in a few important capital exporting countries—notably in the EU and UK—while as at end of 2023 some delay has emerged in adoption in several countries.
- Policy window:
  - This delay allows many low-tax jurisdictions time to design and implement general CIT reforms, taking advantage of features of the minimum tax such as treatment of certain immediate expensing and refundable tax credit regimes, before—or alongside—specific GloBE rules.

---

*Source: IMF Working Paper — "1. Introduction" (wpiea2024064-print-pdf).*

### Box 1. Implementation of GloBE Rules in Selected Jurisdictions

### Box 1. Implementation of GloBE Rules in Selected Jurisdictions

### Overview and timing of adoption
- GloBE adoption plans are ongoing; outside Europe there is an emerging trend to delay adoption until at least 2025.
- Many low tax countries and investment hubs announced implementing a general CIT before—or alongside—specific GloBE rules.

### Selected countries: IIR, QDMTT, and UTPR implementation timing (as reported)
- Singapore: IIR 2025; QDMTT 2025; UTPR 2025
- Cyprus: IIR 2024; QDMTT 2025; UTPR 2025
- Netherlands: IIR 2024; QDMTT 2024; UTPR 2025
- EU: IIR 2024; QDMTT optional; UTPR 2025
  - Must transpose EU Directive end 2023 with IIR and UTPR (but can defer application to end 2029 if no more than 12 ultimate parent entities; elected by Estonia, Latvia, Lithuania, Malta, and Slovakia).
- Switzerland: IIR Delayed; QDMTT 2024; UTPR Delayed
- United Kingdom: IIR 2024; QDMTT 2024; UTPR 2025
- United States: No legislative plans for IIR, QDMTT, or UTPR
  - US GILTI regime qualifies as a Blended CFC Tax Regime under GloBE rules.
  - A domestic corporate alternative minimum tax (at 15%) on the adjusted financial statement income of companies with profits > US$1 billion adopted.

### Selected low tax jurisdictions (without pre-existing CIT regimes)
- Barbados
  - No public IIR plans
  - QDMTT plans (2024, conditional on top-up tax payable elsewhere)
  - No public UTPR plans
  - General CIT: Proposal for a general CIT (at 9%) effective 2024, with QDMTT (at 15%) for multinational groups with revenues of at least €750 million that are subject to an IIR or UTPR elsewhere. A qualified refundable tax credit (QRTC) is also proposed (jobs credit based on payroll plus R&D credit).
- Bermuda
  - No public IIR plans
  - No specific QDMTT plans
  - No public UTPR plans
  - General CIT: Adopted general CIT (at 15%) for multinational groups with revenues of €750 million or more (2025). Final law removed credit for U.S. GILTI but has temporary U.S. CFC income exclusion instead. QRTCs also proposed.
- Guernsey: IIR 2025; QDMTT 2025; No public UTPR plans
- Jersey: IIR 2025; QDMTT 2025; No public UTPR plans
- Isle of Man: IIR 2025; QDMTT 2025; No public UTPR plans
- UAE
  - No public IIR plans
  - QDMTT delayed (2025)
  - No public UTPR plans
  - General CIT: Adopted CIT (at 9%) on annual taxable profits above AED 375,000 (effective 2023), with delayed adoption of a QDMTT (at 15%).
- Kuwait
  - No public IIR plans
  - No public QDMTT plans
  - No public UTPR plans
  - General CIT: Proposal to adopt CIT (at 15%) from 2025, without specific GloBE rules.

### Strategic considerations for low-tax jurisdictions
- Adopting the QDMTT is a dominant strategy on revenue grounds:
  - It ensures the country can collect top-up tax payable by in-scope multinationals with affiliates in the low-tax country rather than another jurisdiction.
  - QDMTT per se would not put a jurisdiction at an investment disadvantage because wholly owned in-scope companies will in any case be taxed at 15 percent effective tax (if any parent in the chain is in a jurisdiction that implements Pillar Two).
  - Where existing profit exceeds the SBIE, companies may find it beneficial to increase employment and assets in the jurisdiction to benefit from the SBIE exemption.
  - No added compliance costs for the jurisdiction because in-scope multinationals must undertake the calculations irrespective of the jurisdiction’s policy.
- Adopting the IIR is also likely:
  - Ensures revenue collection if other jurisdictions do not implement a QDMTT.
  - Renders CFC rules redundant for in-scope multinationals.
  - Not adopting the IIR can be a strategy to attract headquarters, but this requires in-scope multinationals to find country pairs where neither QDMTT nor IIR has been adopted and relocation is economically worthwhile.
- The case for implementing the UTPR is currently weak for low-tax jurisdictions, because:
  - (i) The UTPR is a provision of last resort; transitional arrangements can further weaken its practical importance initially.
  - (ii) Implementation risks include questions about consistency with tax treaties and other potential conflicts with international law, since offshore undertaxed profits can become subject to tax in UTPR jurisdictions even when those profits have no connection with them.
  - (iii) The UTPR requires a higher level of administrative co-operation, increasing administrative and compliance complexities and risks of cross-border tax disputes.
  - (iv) The UTPR raises some international trade risks (in the form of possible retaliation) that are yet to be clarified.
- On balance, low-tax jurisdictions may deprioritize UTPR implementation and focus on more pressing CIT issues; this is consistent with the approach signaled by several countries.

### Revenue and design considerations (illustrative mechanics and examples)
- Calibrating a domestic tax to collect the exact top-up tax is challenging; countries should not be fixated on doing so.
- Example comparisons (as described):
  - If net GloBE income is $100 and SBIE is $20, then:
    - General CIT at 15 percent on net GloBE income ($100) yields tax of $15.
    - The top-up tax collected elsewhere under either the IIR or the UTPR would be $12 (assuming net GloBE income of $100 less the SBIE of $20).
  - A general CIT at 15 percent on GloBE ‘excess profits’ ($12) after considering the SBIE would result in a rate below 15 percent (actually, 12 percent), and therefore would still require a top-up tax elsewhere.
  - A QDMTT applied to in-scope constituent entities on their entire GloBE ‘excess profits’ of $12 would apply irrespective of ownership; under the IIR the top-up tax payable elsewhere would be adjusted for ownership interests less than 100 percent. For an 80 percent owned entity, that implies $9.6.
  - Footnote: This top-up tax still becomes payable elsewhere even if a QDMTT was limited to wholly-owned in-scope entities.
- Interaction with CFC rules can open tax-planning avenues:
  - GloBE recognizes general CIT first (a covered tax), followed by the top-up tax in order: QDMTT, then IIR, then UTPR.
  - Blended CFC regimes (e.g., U.S. GILTI) qualify for more favorable CFC allocation rules, which can result in tax positions that push low-tax outcomes to certain jurisdictions.
  - Numerical illustration: assume Company Y profit 100; Company V’s tax (at 21%, before credit) in A is $21; Company V can use credits under a CFC regime with global blending of $6 relating to another high tax jurisdiction. After credit, net tax is $15, which can be pushed down to Jurisdiction C, making C’s top-up tax zero, whereas a QDMTT in C would have yielded additional tax of $15 on top of the CFC tax in A. Company V saves $6 under global blending by paying $15 instead of $21 in A.
- Broader rationales for going beyond GloBE (three broad reasons):
  1. Tax and economic policy perspective: opportunity to diversify and raise revenue, including taxing economic rent of out-of-scope companies with appropriate tax design; design an underlying rent tax system better suited for the country and more robust to future GloBE changes.
  2. Legal design perspective: QDMTT and GloBE rules are designed as an add-on requiring an underlying CIT. A standalone QDMTT would need to be comprehensive, self-executing, and self-administrable and may raise challenges in relation to existing tax treaties and other international initiatives. Countries may need to preserve tax sovereignty over depreciation rules and design of tax credits or other cost-based incentives.
  3. Avoiding application of the subject-to-tax rule (STTR): STTR allows low-income countries to impose a top-up tax on specific cross-border payments if the foreign nominal CIT rate is below 9 percent; to circumvent STTR a country needs some CIT in addition to the GloBE.
- Additional observations for zero-tax jurisdictions:
  - Lack of personal income taxes (PITs) complicates broad taxation of profits; neutrality across income sources and legal forms can be difficult without PIT.
  - Choice of threshold for any profit tax matters (example VAT threshold noted at USD 100).
  - Niche responses targeting a few existing firms may be fragile and can lose appeal quickly as other countries change rules; a proper underlying CIT is preferable to leaving revenue for other jurisdictions to collect.
  - Example: Bermuda adopted a CIT (rate 15%) with features interacting with U.S. tax rules (crediting and temporary income exclusions) illustrating the complexity and changing nature of design choices.

*Source: Authors’ compilation.*

### 5. An Efficient Rent Tax: ACE versus Cash-Flow under the GloBE Rules

### 5. An Efficient Rent Tax: ACE versus Cash-Flow under the GloBE Rules

### Overview
- Economists prefer an efficient profit tax that targets only economic rent, implying a zero marginal effective tax rate (METR).
- Two broad designs:
  - Cash-flow taxation (R–based cash-flow): entire investment immediately expensed; no deduction for interest expense or returns to equity.
  - Allowance for Corporate Equity (ACE): notional deduction for a normal return while maintaining depreciation rules and interest deductions.
- Under the GloBE rules, immediate expensing is treated as a “temporary timing measure” and does not lower the GloBE effective tax rate; by contrast, the ACE’s notional deduction lowers the covered tax rate under GloBE.

### The ACE and the QDMTT — main implications and mechanics
- The ACE is appealing because it covers financial and non-financial companies and uses familiar accounting concepts (depreciation and interest deductions).
- Under GloBE, the ACE lowers the GloBE effective rate more than full expensing and can lead to a top-up tax even when investments earn only the normal return, especially at relatively low statutory CIT rates (still possibly well above 15 percent).
- The effect depends on whether the ACE is treated as a qualified refundable tax credit (QRTC) increasing covered income or as a non-qualified refundable tax credit (NQRTC) lowering covered taxes; magnitude differs though qualitative effect is same.

- Implication 1: The ACE is no longer a tax only on economic rent for sufficiently low statutory CIT rates (still well above the 15 percent). The ACE will imply a top-up tax on in-scope companies even if the investment is just earning the normal return.

- Decomposition of Implication 1 (assuming a QDMTT is in place, ACE as NQRTC, and a single period investment model):
  - (i) For investments that earn the normal return (or less), the ACE will result in a zero covered tax. Therefore, only the QDMTT matters. If the ACE is an NQRTC, the top-up rate is the full 15 percent, to be applied on profits minus SBIE.
  - (ii) For investments that earn economic rent, the ACE implies a top-up tax on in-scope companies if the statutory CIT rate is sufficiently low (certainly 15 percent or lower). The cutoff statutory CIT rate (휏휏) is given by:
    - 휏휏 = 15% × (1 − (푟푟.퐸퐸 / 휋휋))
  - Example: if the notional deduction is 8 percent and the equity-to-profit ratio is 5, a statutory tax rate of 25 percent (or above) is required for preventing the application of the QDMTT.
  - For projects that earn very high economic rent, the covered tax rate approaches the statutory rate; the top-up tax rate approaches 15% − 휏휏 (if positive, otherwise zero).
  - (iii) For projects with no economic rent, the average tax rate depends only on SBIE. If SBIE ≈ 휋휋, average tax rate ≈ zero. If SBIE → 0, average tax rate (even on normal return) approaches the minimum tax.
  - (iv) For projects with economic rent and no top-up tax (statutory rate high enough), the average tax rate equals:
    - 퐴퐴퐴퐴퐴퐴 = (푇푇푇푇푇푇 푝푝푇푇푝푝푝푝) / 휋휋 = 휏휏 × (1 − (푟푟.퐸퐸 / 휋휋))
  - (v) For projects with economic rent and a top-up tax (statutory rate low enough), the average tax rate is:
    - 퐴퐴퐴퐴퐴퐴 = (퐴퐴퐴퐴) / 휋휋 = 15% × (1 − (푆푆푆푆푆푆 퐸퐸 / 휋휋)) + 휏휏 × (1 − (푟푟.퐸퐸 / 휋휋)) × (푆푆푆푆푆푆 퐸퐸 / 휋휋)
  - Example note: if economic rent is low (allowances of 8 or 10 percent), average tax rate is close to 15 percent.
- Recap: In the period-by-period example, ACE+QDMTT does not guarantee no-taxation of normal returns. The top-up rate for a project yielding the normal return is always 15 percent. Top-up amount is positive if SBIE < normal return, zero if SBIE ≥ normal return. ACE becomes a tax on economic rent only for sufficiently high statutory tax rates, well above 15 percent (see Equation (1)).

### Immediate Expensing (vs ACE) and the QDMTT
- Immediate expensing is treated as a “temporary timing measure” and gives rise to an upward adjustment to covered taxes to reflect temporary differences (Article 4.4 of the Model Rules). Thus immediate expensing does not impact the GloBE effective tax rate and by itself does not trigger a top-up tax.
- Domestic design implication: a cash-flow tax with full expensing and the same statutory rate (for example 15 percent) implies a zero METR if losses are refundable or carried forward with interest.
- Theoretically, absent a minimum tax, ACE and cash-flow tax are equivalent. Under GloBE they diverge because ACE lowers the GloBE effective rate while immediate expensing does not.

- Implication 2: Given a relatively low statutory tax rate, under the GloBE, immediate expensing (R-based cash-flow tax) tends to result in a lower METR than the ACE, ceteris paribus.

- Implication 2 organized into three cases (assuming QDMTT in place, ACE as NQRTC, period-by-period):
  - (i) Case 1: No top-up tax under either ACE or immediate expensing. Occurs only if the statutory rate is sufficiently high (per Equation (1)) and the investment yields economic rent. In this case both designs are equivalent (no distortions).
  - (ii) Case 2: Top-up tax under ACE but not under cash-flow tax. Occurs for statutory tax rates between 15 percent and the upper level given by Equation (1). ACE loses its efficiency features; cash-flow tax maintains efficiency (except banks subject to QDMTT).
  - (iii) Case 3: Top-up taxes under both systems (휏휏 < 15%). Top-up tax is always higher under ACE than under immediate expensing since 휏휏 / (1 − (푟푟∙퐸퐸 / 휋휋)) < 휏휏. Thus average tax rate is always higher under ACE than under immediate expensing.

- Note: In dynamic models the refundability of tax losses is important; the ACE can in some dynamic settings result in a lower METR than cash-flow taxes for some investments.

### Why Not Other Forms of Cash-Flow Taxes?
- Banks require special treatment: under an R-based cash-flow tax, banks’ interest income is untaxed and they end up subject to the QDMTT.
- Other cash-flow forms:
  - R+F–based cash-flow: includes net financial transactions; administratively difficult.
  - S–based cash-flow (distribution-based): taxes net distributions (dividends + buybacks − new equity). Pillar Two guidance does not treat new S-based cash-flow taxes equally to temporary timing measures or pre-existing distribution systems; eligibility for special treatment requires being in force on or before 1 July 2021 and meeting rate and scope conditions.
- Financial Activity Tax: base approximates excess profit plus remuneration of financial institutions; functions like a VAT on financial intermediation. If remuneration excluded and statutory rate is low (per Equation (1)), the financial activity tax becomes a form of ACE and QDMTT binds. Including remuneration raises the covered tax but requires included remuneration and profit portions to be relatively high to avoid QDMTT.
- METR can be negative (a subsidy) without triggering a top-up tax if credits are provided as QRTC (for example on top of interest deductions or immediate expensing). Such policies typically address positive externalities (e.g., R&D).

### Conclusions and policy implications
- Protection from top-up tax payable elsewhere should be a tax design objective. Adopting the QDMTT (and possibly the IIR) is part of that protection, but jurisdictions should adopt the GloBE as a backstop on top of a well-designed profit tax.
- Legal implementation suggestions:
  - Foundational legal infrastructure: core definitions, collection and enforcement provisions.
  - Customary international tax and anti-avoidance provisions: functional transfer pricing rules, economic substance rules for foreign source income exemptions.
  - GloBE top-up tax rules could be implemented as a backstop closely following the Model Rules and possibly enacted through a separate legal instrument leveraging foundational profit tax infrastructure.

- Key implications:
  - The normal return can be taxed even under an efficient rent tax design, especially at relatively low statutory tax rates (possibly above 15 percent), unless SBIE ≥ normal return for all marginal investors each year.
  - GloBE breaks the equivalence among economic rent tax designs, tilting toward the R-based cash-flow tax because immediate expensing does not lower the covered tax and thus preserves a zero METR.
  - Under ACE, an in-scope company in any year may face a non-zero top-up rate (effectively under QDMTT) unless statutory tax rate or SBIE is sufficiently high (single-period formula suggests thresholds can be around 25 percent).
  - Under an R-based cash-flow tax, banks will end up under QDMTT in a pure R-based design.
  - New S-based cash-flow regimes are intolerable under GloBE because they would lead to a lower covered tax.
- Policy takeaway: Zero- or low-tax jurisdictions can design profit taxes to ensure zero tax on investment at the margin; generally this requires a rate of at least 15 percent. Effective tax on economic rent remains moderate and well below 15 percent. Even with efficient rent taxes around 15 percent, jurisdictions may continue to receive some foreign profits as they remain in the lower range of international profit taxation.

### Box 3 — Summary of Design Options for Efficient Taxation of Economic Rent
- R–based cash flow tax:
  - Tax base: net real transactions (exclude financial flows such as interest payments, net debt issuance, net dividends).
  - Features: immediate expensing, normal return untaxed, no debt bias, losses refunded or carried forward at appropriate interest.
  - Examples of full expensing implementations: Hungary (small businesses only), United Kingdom, United States (but existing systems often keep some interest deductibility).
- (R+F)–based cash flow tax:
  - Tax base: net real transactions plus net financial transactions (received borrowing and interest less interest paid and debt repayment).
  - Equivalent to R-based for non-financial firms but administratively difficult (Mexico’s 2008–2014 experience).
- S–based cash flow tax:
  - Tax base: net distributions (dividends + share buybacks − new equity issued); equivalent in principle to (R+F).
  - Examples: Estonia and Latvia distribution-based CITs.
  - Pillar Two treatment: new S-based cash-flow taxes are not treated equally; eligible distribution tax system must meet specific conditions (tax payable on distribution, rate ≥ minimum, in force on/before 1 July 2021).
- Allowance for Normal Return (ACE):
  - Provides a notional deduction r × equity; can eliminate investment distortion and debt bias if notional return aligns with interest rate.
  - Past adopters include Belgium and Italy; allowance often linked to long-term government bond yields.
  - Risks: unilateral ACE can enable international tax planning.
- Financial Activity Tax:
  - Base: excess profit + remuneration of financial institutions; approximates tax on economic rents in finance.
  - If low statutory rate and remuneration excluded, becomes ACE-like and QDMTT binds. Including remuneration raises covered tax but requires high included portions to avoid QDMTT.

*Source: IMF Working Paper — 5. An Efficient Rent Tax: ACE versus Cash-Flow under the GloBE Rules*

### References

### References

### Academic articles and book chapters
- Adam, Stuart, Helen Miller, 2023, Full Expensing and the Corporation Tax Base, IFS Green Budget - Chapter 10.
- Devereux, Michael, and John Vella, 2014, Are We Heading towards a Corporate Tax System Fit for the 21st Century?, Fiscal Studies 35(4), 449–75.
- Devereux, Michael, and Rachel Griffith, 2003, Evaluating Tax Policy for Location Decisions, International Tax and Public Finance 10: 107–126.
- Dharmapala, Dhammika, 2023, The Institutional and Historical Characteristics of Tax Havens, Mimeo.
- Dharmapala, Dhammika, and Hines, James R, 2009, Which Countries Become Tax Havens?, Journal of Public Economics 93(9-10), 1058–1068.
- Hebous, Shafik, and Michael Keen, 2023, Pareto-Improving Minimum Corporate Taxation, Journal of Public Economics 225, 104952.
- Hebous, Shafik, and Alexandre Klemm, 2020, Destination-Based Allowance for Corporate Equity, International Tax and Public Finance 27, 753–777.
- Hebous, Shafik, and Martin Ruf, 2017, Evaluating the Effects of ACE Systems on Multinational Debt Financing and Investment, Journal of Public Economics 156, 131–149.
- Janeba, E. and Schjelderup, G., 2023, The Global Minimum Tax Raises More Revenues Than You Think, or Much Less, Journal of International Economics 145, 103837.
- Johannesen, Niels, 2022, The Global Minimum Tax, Journal of Public Economics 212, 104709.
- Kanbur, Ravi, and Keen, Michael 1993, Jeux Sans Frontières: Tax Competition and Tax Coordination when Countries Differ in Size, American Economic Review 83(4) (1993), 877–892.
- Keen, M, and King, J, 2002, The Croatian Profit Tax: an ACE in Practice, Fiscal Studies 23(3), 401–418.
- Keen, Michael, and Konrad, Kai, 2013, The theory of international tax competition and coordination Auerbach Alan J., Chetty Raj, Feldstein Martin, Saez Emmanuel (Eds.), Handbook of Public Economics, Vol. 5, Elsevier, Amsterdam (2013), 257–328.
- Overesch, Michael, Dirk Schindler, and Georg Wamser, 2024. The Impact of CFC Rules and GILTY on Tax Havens.
- Schjelderup, Guttorm and Stähler, Frank, 2023, The Economics of the Global Minimum Tax, International Tax and Public Finance, forthcoming.
- Tørsløv, Thomas, Ludvig Wier, Gabriel Zucman,2023, The Missing Profits of Nations, Review of Economic Studies 90(3, 1499–1534.
- Wilson, John D., 1986, A Theory of Interregional Tax Competition, Journal of Urban Economics 19 (3), 296–315.
- Zodrow, George, Mieszkowski, Peter, 1986, Pigou, Tiebout, Property Taxation, and the Underprovision of Local Public Goods, Journal of Urban Economics 19 (3), 356–370.

### Policy papers, working papers, and reports
- De Mooij, Ruud, Dinar Prihardini, Antje Pflugbeil, and Emil Stavrev, 2020, International Taxation and Luxembourg’s Economy, IMF Working Paper No. 2020/264.
- Dharmapala, Dhammika, 2023, The Institutional and Historical Characteristics of Tax Havens, Mimeo.
- Erokhin, Dmitry and Alfons Weichenrieder, 2023, Conduit Countries and Treaty Shopping, Mimeo.
- G20, 2010, Financial Sector Taxation. The IMF's Report to the G-20 and Background Material.
- Haufler, Andreas and Hayato Kato, Forthcoming, A Global Minimum Tax for Large Firms Only: Implications for Tax Competition.
- Hebous, Shafik, 2021, Global Firms, National Corporate Taxes: An Evolution of Incompatibility, In Corporate Income Taxes under Pressure. Chapter 4. IMF.
- Hebous, Shafik, and Andualem Mengistu, 2024, Efficient Economic Rent Taxation under a Global Minimum Corporate Tax, IMF Working Paper.
- Hebous, Shafik, Dinar Prihardini, and Nate Vernon, 2022, Excess Profit Taxes: Historical Perspective and Contemporary Relevance, IMF WP 2022/187.
- Hindriks, J. and Yukihiro, N., 2022, The Compliance Dilemma of the Global Minimum Tax. Mimeo.
- IMF, 2023, International Corporate Tax Reform, Policy Paper.
- IMF, 2022, Coordinating Taxation Across Borders, Fiscal Monitor, Chapter 2.
- IMF, 2016, Tax Policy, Leverage and Macroeconomic Stability, IMF Policy Paper.
- IMF, 2014, Spillovers in International Corporate Taxation, Policy Paper.
- IFS, 2011, Tax by Design: The Mirrlees Review, Institute for Fiscal Studies.
- IFS, 1991, A Report of the IFS Capital Taxes Group. Institute for Fiscal Studies.
- OECD (2024), Economic Impact Assessment of Global Minimum Tax: Summary, January 2024.
- OECD (2023), Tax Challenges Arising from the Digitalisation of the Economy – Administrative Guidance on the Global Anti-Base Erosion Model Rules (Pillar Two), OECD/G20 Inclusive Framework on BEPS, OECD, Paris.

### Theoretical and empirical contributions on tax havens and tax competition
- Dharmapala, Dhammika, and Hines, James R, 2009, Which Countries Become Tax Havens?, Journal of Public Economics 93(9-10), 1058–1068.
- Erokhin, Dmitry and Alfons Weichenrieder, 2023, Conduit Countries and Treaty Shopping, Mimeo.
- Janeba, E. and Schjelderup, G., 2023, The Global Minimum Tax Raises More Revenues Than You Think, or Much Less, Journal of International Economics 145, 103837.
- Overesch, Michael, Dirk Schindler, and Georg Wamser, 2024. The Impact of CFC Rules and GILTY on Tax Havens.
- Tørsløv, Thomas, Ludvig Wier, Gabriel Zucman,2023, The Missing Profits of Nations, Review of Economic Studies 90(3, 1499–1534.

*Deciphering the GloBE in a Low-Tax Jurisdiction Working Paper No. WP/2024/064*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024064-print-pdf.pdf_