By Thomas Locher, research assistant with the Tax Foundation’s Center for Global Tax Policy, Washington, D.C.
With momentum building for an agreement on a global minimum tax and the G20’s broad backing of the OECD’s plan in their July 10 communique, questions of practicality and administrability mount. In particular, concerns arise with respect to the interaction between the income inclusion rule under the OECD’s Global Anti-Base Erosion (GloBE) proposal and the current controlled foreign corporation (CFC) rules in place across much of the OECD.
While the goals of both the income inclusion rule and CFC rules are similar (minimizing cross-border profit-shifting into jurisdictions with lower corporate tax rates), the means by which such goals are accomplished vary significantly between the two rules. These differences have raised questions about the efficacy of forging a new framework separate from the existing standards which, for many countries, were only recently put in place.
The issues raised by the overlay of an income inclusion rule on top of current CFC rules bear a striking resemblance to the disregard for the interactions between Subpart F rules and global intangible low-tax income (GILTI) shown by the US legislature in the 2017 reform effort.
There are three noteworthy parallels between the US’s Subpart F/GILTI collision and the impending CFC/income inclusion rule collision: differences in income-blending methods, differences in the type of income that is in-scope, and the creation of a separate tax rate for foreign income.
Differences in income blending could lead to challenges in allocating taxable income, foreign tax credits, and deductible losses. The differences in the type of income that is in-scope relates to the focus that most CFC rules place on passive income, while the income inclusion rule will capture both active and passive income. The income inclusion rule would bring an additional effective tax rate calculation into a company’s determination of tax liability on top of existing CFC rules that have effective tax rate tests.
When providing further guidance to countries as lawmakers prepare to implement the new agreement, the OECD should focus on ways to simplify the connections between the income inclusion rule and CFC rules. Countries should work to ensure that the two approaches to determining tax liability on foreign income do not leave companies in an uncertain position when making new cross-border investment decisions.
Blending
The mismatch between different regimes regarding the calculation of a multinational corporations’ group effective tax rate across entities or jurisdictions, or “blending,” can lead to significant administrative costs and new tax disputes as governments and companies work out how the income inclusion rule and CFC rules work (or don’t work) together. To illustrate this mismatch, we must first explore how each set of rules blends income to determine the appropriate effective tax rate.
ATAD and Subpart F
Under Subpart F rules, a parent corporation nets income at the entity level, meaning the taxable income from each CFC is attributed to the parent corporation, regardless of losses incurred by other CFCs. While losses by a CFC cannot be attributed to the parent organization or a separate CFC, they are able to be carried forward to offset future profits by the individual CFC. Additionally, the parent company may receive a foreign tax credit for taxes paid by a CFC in the jurisdiction in which it operates, and those foreign tax credits may be carried forward ten years, or back to the immediately preceding year.
The CFC rules across EU countries align to the European Commission’s Anti-Tax Avoidance Directive (ATAD). The ATAD allows for CFC rules to come in two distinct flavors. Both share important characteristics in the context of examining their interaction with the proposed global minimum tax. The first, Model A, broadly reflects the functionality of the US’s Subpart F rules in that it assigns passive income from CFCs to the parent company based on their share of ownership. Model B is more subjective; it assigns to the parent company all a CFC’s income from “non-genuine arrangements that have been put in place for the essential purpose of obtaining a tax advantage.”
Both ATAD models, like Subpart F, calculate taxable income at the CFC level, utilizing entity-level blending. To maintain consistency in assigning tax burdens across organizations, CFC income is calculated using the same accounting methods as that of the parent company. This prevents the use of certain tax-avoidance tools at the CFC level that would not normally be used by parent organizations, but it can also lead to inconsistencies in determining the effective tax rate levied against the CFC income if different accounting methods are used across jurisdictions.
Minimum taxes: GILTI & GloBE
Whereas Subpart F and other CFC rules around the world are built with a similar structure, GILTI (as currently implemented) and the proposed functionality of GloBE are not perfectly analogous. GILTI works at a global level, netting profits, losses, and taxes paid from all CFCs in one pool, while GloBE considers separately each jurisdiction in which the parent organization has affiliated operations. President Biden has expressed support for modifying GILTI to operate at a jurisdictional level.
Under GILTI, income is netted at the global level, using so-called “global blending.” This means that the income or losses from every CFC under the parent corporation’s control are attributed to the parent organization before the tax is calculated, allowing for losses from one CFC to offset profits from another. While this flexibility in blending profits and losses across entities typically makes global blending taxpayer-friendly, the GILTI rules do not allow for the carryforward or back of unused GILTI-specific foreign tax credits, severely limiting the intertemporal benefits of this system.
The Biden administration has proposed shifting to a jurisdictional system, while continuing to disallow the carryforward of excess foreign tax credits. This system would mean that GILTI does not account for differences in profitability across different stages of a business’s lifecycle, which would inflate GILTI liability over time.
The proposed GloBE income inclusion rule operates on a jurisdictional level. This means that a multinational firm is subject to an effective tax rate test in all the countries in which it operates. This requires proportionally assigning profits, losses, and taxes paid from all shareholder-held CFCs to specific jurisdictions and calculating the effective tax rate in each. Where the parent organization controls a large number of CFCs operating in a number of different jurisdictions, assigning income and tax credits could take on a level of complexity and subjectivity that undermines the goal of limiting profit-shifting and goes against the principles of sound tax policy.
Types of income
Another notable difference between the global CFC rules and the proposed GloBE income inclusion rule is the scope of the taxable income under each. CFC rules typically specify the type of income to which they apply, though the rules are operated with a level of subjectivity that allows them to more specifically target certain types of tax avoidance without putting an undue burden on CFCs or shareholders carrying out legitimate business operations.
In some instances, only passive income is included; in other cases, all income from non-genuine arrangements is included. Some countries complicate matters by deeming different proportions of income as either passive or active based on set thresholds. Additionally, CFC income is often subject to an effective tax rate test, prior to any income being assigned back to the parent organization. This initial effective rate test is usually in reference to the parent’s domestic tax rate, and, if satisfied, this test exempts CFC income from the parent’s taxable base, at least until the GloBE rules are applied.
In contrast, the income inclusion rule, like the tax on GILTI, applies to all types of income, with a formulaic substantive carve-out for the amount of income deemed attributable to certain substantive indicators. This likely means that in countries with a CFC regime that reflects ATAD’s Model A or the US’s Subpart F rules, the parent organization’s income would have to undergo a number of examinations, each with stark technical differences, to reach a potentially uncertain final answer for the effective tax rates in all the jurisdictions in which it operates. Then it would need to calculate any additional tax liability under the income inclusion rule.
Separate tax rates
Underlying nearly all these considerations is the separate rates at which CFC income is taxed (or not) in the home jurisdiction and how that would impact income inclusion rule tax liability. For CFCs that satisfy the effective rate test under the CFC regime, the full income is included in the GloBE base at the global minimum rate with foreign tax credits limited to the taxes paid in the CFC jurisdiction to determine additional income inclusion rule liability.
Conversely, if the CFC does not satisfy the CFC effective rate test, the entirety of their passive income could be taxed at the full domestic rate. Both the corporate income tax paid in the CFC jurisdiction and the tax on the passive income in the parent jurisdiction would then be credited back to the CFC’s jurisdiction for the purpose of calculating the effective tax rate on all income from all CFCs operating in that jurisdiction for the income inclusion rule. The parent organization is then responsible for the difference between the effective tax paid and the minimum tax under GloBE. In this case, income from a single CFC is subject to tax at three separate stages and three separate rates.
Conclusion
Operating two separate rules with similar policy goals and a potentially significant overlap in the taxable base is an inherently risky approach to tax policy. This is doubly true when the taxable income is being calculated at two different organizational levels (both the CFC and the jurisdiction). The global approach to blending from GILTI is easily applied in concert with Subpart F as the tested income is assigned back to the final shareholder. In that case, any exclusions or deductions for taxes paid on Subpart F income are considered in only one jurisdiction, and by one entity. The arithmetic is not so simple when done across a number of (often competing) jurisdictions.
The process will likely be a laborious one. The parent entity would first determine whether the CFC’s passive income is includable under current CFC rules, i.e., whether it exceeds the de minimis threshold and any active-to-passive ratio test. The parent entity would then calculate the effective tax rate on the total CFC income. If the effective tax is below the domestic threshold, the parent entity would pay the full domestic corporate tax rate on the passive income. The organization would then divvy up CFC profits, losses, and tax credits by jurisdiction. Then, for each jurisdiction in which it operates, it would calculate the income across all CFCs, determine the effective tax rate paid in each jurisdiction, and pay a top-up tax to its home jurisdiction for all jurisdictions that fall under the 15 percent GloBE rate.
For a company with established operations, assets, and employees across the globe, this process might be a complex, but mechanical, calculation. For companies expanding or making investment decisions to compete in the global marketplace, the complexities and potential subjectivity baked into the process will pose a confounding hurdle to foreseeing the international tax implications of establishing themselves in foreign markets.
Where there are policies already employed to combat profit-shifting, the added calculation does more to increase compliance costs and discourage investment than it may do to provide additional tax base protection. Countries may be better served standardizing and building off of the CFC rules already in place than operating two contrasting rules in parallel. A uniform definition of taxable CFC income and a standardized effective tax rate test for that income would limit the compliance costs of creating additional rules that operate separately from the framework already in place.
It must be read every authorization