The Tax Foundation appreciates the opportunity to respond to the public consultation on the reports on the Pillar 1 and Pillar 2 blueprints. The reports show how much technical work has been done since the “Unified Approach” was released, and they reveal how much work would need to be done to adapt tax systems to align with the new approach.
One key weakness in the documents is that they fail to provide a clear and consistent vision for the international tax system. However, there are some recurring assertions that underlay a basis for the Inclusive Framework’s work:
Assertion #1: The generation of residual profits can be related to market presence even in the absence of physical presence.
Assertion #2: Rules developed to address both preferential tax policies and the ability to artificially locate profits in low-tax jurisdictions could be improved.
The first assertion motivates Pillar 1, particularly Amount A. This assertion has led the OECD to develop a proposal that provides market jurisdictions with the ability to tax some share of residual profits. The second assertion motivates the global minimum tax envisioned under Pillar 2 which has led to proposed rules whereby a jurisdiction could tax an entity in a foreign location depending on whether it has been taxed below a minimum rate.
But assertions are not principles. In Pillar 1, the Inclusive Frameworks’ reliance on assertions rather than sound tax principles have led to seemingly arbitrary and often contradictory scoping decisions. In the case of Pillar 2, assertion-based solutions risk raising compliance costs and reducing investment. However, solutions based on economically sound principles can greatly reduce the policy’s impact on FDI and reduce compliance costs for firms.
The consultation document requests input on simplifications to both pillars. Simplifications are challenging to identify given the design of the proposals. A key complication is that the two pillars are based on somewhat contradictory approaches.
Following the assertion underlying Pillar 1 would suggest designing a broad destination-based tax system as a replacement for current source and residence rules. This could be in the form of a destination-based cash flow tax or global formulary apportionment by sales. However, the proposals in Pillar 1 do not go nearly that far and leave existing tax rules to stand alongside a new partial destination-apportionment system that would narrowly apply to certain industries and large companies. The design choices that limit the scope of Pillar 1 will create new distortions which will impact how much revenue changes hands between jurisdictions, the compliance costs for businesses, and the administration costs for governments.
Pillar 2 would bring about stronger rules for residence-based taxation in the face of various low-tax regimes including patent boxes, tax holidays, and other incentive schemes. The minimum tax would directly impact the tax costs for businesses that utilize these schemes and could increase investment costs across the globe. The Inclusive Framework could, however, design the tax to be neutral toward business investment decisions.
In addition to existing rules for multinationals that will not be changed by the proposals, taxpayers will have to consider the new tax rules that will apply in headquarter countries, countries where their intermediaries are located, and where their customers reside. While there may be principled justifications for a tax system that taxes business income at any one of those levels, a system that combines all three approaches with offsetting policies to avoid double taxation is likely to create implementation and compliance challenges. For simple, principled tax policy to prevail at the global level, the Inclusive Framework must work to design a system with fewer contradictions and costs for administration and compliance.
If the goal of the project is not fundamental reform, but rather about the taxation of digital goods and services in the location of the consumer, then the Inclusive Framework should note that more than 60 countries worldwide have implemented OECD standards for collection of value-added tax on digital goods and services.
Inclusive Framework countries should also recognize the potential for tax disputes under the blueprints, and that recent OECD data on mutual agreement procedure (MAP) cases shows that in 2019 transfer pricing cases took, on average, 30.5 months to complete and other cases took 22 months to complete on average. Both metrics have increased since 2016. Case loads have also remained at high levels.
Beyond these general comments, we offer a few narrow recommendations in the context of Pillar 1, Amount A, and on the design of Pillar 2. However, we recognize that unless a different approach is taken entirely, these recommendations will likely do little to simplify an overly complex approach to change international tax rules.
On Pillar 1, Amount A, the Inclusive Framework should minimize the distortionary effect of the various thresholds in the proposed design. This would mean applying the new rules to both automated digital services and consumer-facing businesses and setting the deemed residual profit threshold high enough so that Amount A does not become a new factor in business investment decisions. The global revenue threshold will likely create distortions in business behavior wherever it is set.
On Pillar 2, the goal of the Inclusive Framework should be to minimize the impact of a global minimum tax on business investment costs. This could be done by taking a full expensing approach to the tax base design.
Pillar 1, Amount A Scoping Decisions
This response focuses on three issues the Inclusive Framework should consider on Pillar 1, Amount A. First, the industry scoping decision should be aligned with the underlying assertion behind Pillar 1. Second, the deemed residual profits threshold could directly impact business investment costs if the threshold is set too low. Third, the global revenue threshold will result in the same challenges as any threshold in tax systems and could be ripe for abuse.
The industry scoping decision will determine which types of businesses will have to pay under Pillar 1, Amount A. Current discussions revolve around automated digital services and consumer-facing businesses. A principled approach to destination-based taxation would not draw distinctions between industries. If the only choice is whether to include both automated digital services and consumer-facing businesses or only automated digital services, then the Inclusive Framework should include both. Arbitrary distinctions between the two would likely complicate the enforcement of Amount A unless both are included.
Deemed Residual Profits Threshold
Business investment decisions can be distorted by taxes in a variety of ways. One way is through the taxation of the normal return on investment. The normal return on investment can vary by industry or individual business and is separate from profits that are beyond the normal required return, economic profit. Economic profit is not sensitive to taxation since it is, by definition, beyond what a normal investor might expect as a rate of return.
The assertion underlying Pillar 1, Amount A suggests that a portion of excess or residual profits should be taxed in market jurisdictions. However, if the Inclusive Framework sets the deemed residual profit threshold too low, then both normal return and economic profits would be subject to Amount A. This means that Amount A could be influential on business investment decisions and distort investment behavior. A low deemed residual profits threshold paired with a high share of reallocated profits would exacerbate this distortion.
Because individual industries and businesses can differ significantly, the Inclusive Framework has a challenge of defining a threshold for deemed residual profits that will have divergent impacts. To avoid Amount A becoming an influence on business hiring and investment decisions, a high threshold for deemed residual profits should be chosen.
Global Revenue Threshold
The global revenue threshold will be a distorting factor at almost any level. The potential complexity of calculating, paying, and getting double tax relief under Amount A could cause businesses to plan around the global revenue threshold. This could be particularly true for businesses near the threshold.
Recent research has provided evidence of bunching below the €750 million threshold for country-by-country reporting. This means that the threshold has directly influenced business behavior and introduced new avoidance behavior. It would be reasonable to expect that adding requirements related to that threshold could contribute to a more significant response by businesses.
Arbitrary thresholds result in arbitrary behavior, and the global revenue threshold is simply a number with little justification but significant impact for business compliance and government enforcement costs.
Pillar 2 Design
Pillar 2 is based on the assertion that multinational companies are still engaging in substantial amounts of profit shifting despite the Base Erosion and Profit Shifting (BEPS) rules developed to prevent such behavior. Following that assertion would lead the Inclusive Framework to address the remaining gaps with new policies.
The rules which were developed as part of the original BEPS project have only been implemented in recent years, and their full effects are unclear. The Inclusive Framework should understand the effectiveness of those rules before designing complicated “improvements.”
Unfortunately, data follows policy implementations with a lag, and even the most recent studies do not take the new U.S. tax rules and the European Union’s Anti-Tax Avoidance Directive into account. However, several studies have pointed to the impact of specific policies related to the BEPS project. The economic research shows a clear trade-off between protecting tax bases and increasing investment and compliance costs. In other words, tougher base erosion rules lead to higher compliance costs and less investment.
A 2015 study by economists Peter Egger and Georg Wamser found the design of Germany’s controlled foreign corporation legislation led German multinationals to reduce investment in fixed foreign assets by an average estimated €7 million. Thin capitalization rules studied by economist Thiess Buettner and his coauthors in 2014 increase the cost of capital and have negative effects on employment and investment, particularly foreign direct investment. IMF economists Ruud de Mooij and Li Liu found in a 2018 study that the impact of tighter transfer pricing regulations is similar to the effect of increasing the corporate tax rate by one-quarter. A study (most recently revised in 2019) by economists Michael Overesch and Hubertus Wolff found that transparency measures like country-by-country reporting increase compliance costs and effective tax rates.
Even more recently, a study by IMF economists Alexander Klemm and Li Liu points out that limiting profit shifting increases the costs of capital and can thus have direct effects on investment decisions and tax competition. Klemm and Liu back up their argument by pointing to research (including some studies cited earlier) showing the connection between profit shifting and investment effects.
The Inclusive Framework should work to design a minimum tax that is neutral to business investment decisions to avoid costly impacts on global investment. An effective minimum tax in this vein could tax profits mainly from intangible assets that are shifted to low-tax jurisdictions and leave profits from tangible, substantive activities mostly untouched and left to jurisdictions to tax outside of Pillar 2 rules.
A Minimum Tax Base that is Neutral to Investment Decisions
What exactly does an investment-neutral tax entail? Put simply, such a tax allows businesses to immediately deduct the cost of new investments when calculating their taxable profits, an approach called “full expensing.” Under a full expensing approach, businesses pay taxes when profits exceed investment costs, so businesses that are planning large expansions and building new facilities need not account for the tax impact on those decisions except when profits exceed their investment costs. In economic terms, such profits are generally referred to as “excess profits” or “economic rents,” meaning they are profits above what investors might demand as a normal return on investment.
For example, if a business earns $1,000 in revenue and has $900 in labor costs and costs of materials, that business will turn a $100 profit. If $90 of that profit is reinvested in the business with a purchase of new machinery or technology, then taxable profit is reduced to $10. The share of tax paid on that taxable profit would result in the effective tax rate and determine whether a business is paying above or below the minimum tax rate.
Some countries have corporate tax systems that include full expensing. Most countries, however, require businesses to deduct those costs over time according to depreciation schedules. This can mean a business may have to deduct its investment costs in small amounts over several years. The time value of money means that those costs will not be fully deducted, though. Depreciation schedules eat away at deductible costs and inflate taxable profits so that taxes fall on both economic rents and the normal return on investment and cause corporate income taxes to distort investment decisions.
Full expensing avoids those distortions and some countries have adopted full expensing to promote investment. That includes policies that are available for some investments in the U.S. and Canada while other countries including Estonia, Latvia, and Georgia have designed their entire corporate tax approach around the idea and simply tax business cash flow.
These points about investment neutrality are particularly important for the Inclusive Framework’s effort on the global minimum tax.
Because there is not a common, worldwide definition of taxable corporate income, the blueprints show a policy that relies on public financial statements as a starting point for defining corporate income for the purposes of the minimum tax. However, because financial profits can differ significantly from taxable profits, partially because of different tax treatments of investment costs, it is necessary to define a tax base that accounts for those differences.
Using a full expensing approach would take the concept of the minimum tax in the direction of not only a minimum rate but also a minimum base. This approach has multiple benefits.
First, it would neutralize the effect the minimum tax has on new investment decisions. Investment is a key driver of global growth, and a minimum tax with a full expensing tax base could avoid creating new barriers to growth and investment.
Second, the minimum tax would become a tax on excess profits or economic rents. A business in scope of the policy would only pay the minimum tax if its profits (in excess of investment costs) are taxed below the minimum rate.
Third, full expensing for a minimum tax could create the right incentives for countries that may want to change their tax policies in response to the adoption of a minimum tax. If, for instance, a minimum tax of 10 percent is adopted with full expensing, a country that wants to align its policies with the minimum tax would have a template for a tax policy that minimizes investment distortions.
Fourth, a minimum tax with full expensing would be neutral in a way that special patent box regimes, research and development (R&D) tax subsidies, and lengthy tax holidays are not. While a minimum tax with full expensing would be neutral toward new investments, it would erode the tax windfalls from patent boxes and tax holidays. Countries would still be able to offer these incentives, but if those incentives result in a low effective tax rate (after deducting investment costs), the global minimum tax would apply as a top-up tax.
Finally, a minimum tax that incorporates full expensing into its design would have an element of simplicity relative to the overall complexity of the proposal. Because countries differ on their definitions of taxable income and their treatment of investment, businesses must keep track of the multiple sets of rules and their implications for tax deductions. Full expensing would mean that businesses would not have to engage in multiyear tracking of their investment costs, but instead they would be able to deduct the costs in the year they are incurred.
The blueprints on Pillar 1 and Pillar 2 lay out a set of rules that will require significant resources from businesses and governments to both comply with and administer. The complexity of the system will likely mean increased tax uncertainty for taxpayers and new incentives for both governments and businesses to respond to.
Responses to those incentives are incredibly challenging to measure ahead of time, although the OECD’s Impact Assessment includes significant work in evaluating possible outcomes.
However, one does come away with a sense that the two pillars represent a sort of Rube Goldberg machine for determining how much and where a business should pay taxes. A business that pays Amount A in a market country will have to consider how and to what extent withholding taxes on royalties or corporate taxes paid to that same jurisdiction will offset the Amount A liability or whether double tax relief will come from a separate jurisdiction.
Layering new rules on top of the current system is likely to create additional uncertainties and compliance challenges. The recent experience of the U.S. international tax reform shows how a new system of tax rules (specifically GILTI) layered onto existing rules for taxing foreign income (like Subpart F) can create contradictions that make it much more difficult to administer and for taxpayers to comply with.
The complexities are compounded by issues discussed earlier regarding scope and thresholds. Businesses may see the new rules as sufficiently complex as to change incentives for expansion to foreign markets and heighten further economies of scale (that could lead to increasing consolidation of large multinationals) when it comes to compliance. Additionally, businesses that are out of scope due to their current industry would face incentives to avoid adapting and changing their business lines and being caught by Pillar 1, Amount A.
The unforeseen and unintended consequences multiply due to the complexity of the project as it is designed. The OECD has rightfully requested input on simplifications of the various pieces of the project, but the structure itself is perhaps what creates the largest barriers to true simplification. The simplification expected in the form of removing problematic unilateral measures like digital services taxes could be replaced by a much more challenging system.
 OECD, Consumption Tax Trends 2020: VAT/GST and Excise Rates, Trends and Policy Issues, December 2020, https://doi.org/10.1787/152def2d-en.
 OECD, “Mutual Agreement Procedure Statistics for 2019,” November 2020, https://www.oecd.org/tax/dispute/mutual-agreement-procedure-statistics.htm.
 The terms here (normal return, economic profit, routine return) are compared in Box 1 of Tibor Hanappi and Ana Cinta González Cabral, “The Impact of the Pillar One and Pillar Two Proposals on MNE’s Investment Costs: An Analysis Using Forward-Looking Effective Tax Rates,” OECD Taxation Working Papers No. 50, Oct. 12, 2020, https://www.oecd-ilibrary.org/taxation/the-impact-of-the-pillar-one-and-pillar-two-proposals-on-mne-s-investment-costs_b0876dcf-en.
 A variant of this point is made in footnote 19 of Hanappi and Cabral, “The Impact of the Pillar One and Pillar Two Proposals on MNE’s Investment Costs: An Analysis Using Forward-Looking Effective Tax Rates.”
 The OECD finds a minimal increase in investment costs at a 10 percent deemed residual profits threshold, in Hanappi and Cabral, “The Impact of the Pillar One and Pillar Two Proposals on MNE’s Investment Costs: An Analysis Using Forward-Looking Effective Tax Rates.”
 For more information on this bunching behavior, see Section 5 of Felix Hugger, “The Impact of Country-by Country Reporting on Corporate Tax Avoidance,” ifo Working Papers No. 304, May 2020, https://www.ifo.de/DocDL/wp-2019-304-hugger-corporate-tax-avoidance.pdf.
 Much of this section is adapted from Daniel Bunn, “Profit Shifting: Evaluating the Evidence and Policies to Address It,” Tax Foundation, Jan. 31, 2020, https://taxfoundation.org/profit-shifting-evaluating-evidence-policies/.
 For reviews of studies of economic effects of anti-BEPS measures, see Daniel Bunn, “Ripple Effects from Controlled Foreign Corporation Rules,” Tax Foundation, June 13, 2019, https://taxfoundation.org/controlled-foreign-corporation-rules-effects/; Elke Asen, “The Economics Behind Thin-Cap Rules,” Tax Foundation, June 27, 2019, https://taxfoundation.org/thin-cap-rules-economics/; Daniel Bunn, “The Impacts of Tightening up on Transfer Pricing,” July 11, 2019, https://taxfoundation.org/impacts-of-tightening-up-on-transfer-pricing/; and Daniel Bunn, “The Trade-offs of Tax Transparency Measures,” Tax Foundation, July 25, 2019, https://taxfoundation.org/trade-offs-tax-transparency-measures/.
 Peter H. Egger and Georg Wamser, “The Impact of Controlled Foreign Company Legislation on Real Investments Abroad. A Multi-Dimensional Regression Discontinuity Design,” Journal of Public Economics 129 (September 2015): 77–91, https://doi.org/10.1016/j.jpubeco.2015.07.006.
 Thiess Buettner, Michael Overesch, and Georg Wamser, “Anti Profit-Shifting Rules and Foreign Direct Investment,” International Tax and Public Finance 25 (June 2018): 553–80, https://doi.org/10.1007/s10797-017-9457-0.
 Ruud de Mooij and Li Liu, “At a Cost: The Real Effects of Transfer Pricing Regulations,” IMF Economic Review 68 (March 1, 2020): 268–306, https://doi.org/10.1057/s41308-019-00105-0.
 Michael Overesch and Hubertus Wolff, “Financial Transparency to the Rescue: Effects of Country-by-Country Reporting in the EU Banking Sector on Tax Avoidance,” SSRN Scholarly Paper, Feb. 8, 2019, https://doi.org/10.2139/ssrn.3075784.
 Alexander D. Klemm and Li Liu, “The Impact of Profit Shifting on Economic Activity and Tax Competition,” IMF Working Paper No. 19/287, Dec. 20, 2019, https://www.imf.org/en/Publications/WP/Issues/2019/12/20/The-Impact-of-Profit-Shifting-on-Economic-Activity-and-Tax-Competition-48741.
 Most of this section has been adapted from Daniel Bunn, “Designing a Global Minimum Tax with Full Expensing,” Tax Foundation, Sept. 23, 2020, https://taxfoundation.org/designing-a-global-minimum-tax-with-full-expensing/.