Categories
Uncategorized

Expense Allocation: A Hidden Tax on Domestic Activities and Foreign Profits

The U.S. corporate tax code is a complicated behemoth, loaded with numerous arcane provisions—some of these providing special tax breaks, others imposing special tax penalties. Among the latter group, indirect expense allocation rules penalize domestic activities and impose a hidden surtax on foreign profits. While arcane, expense allocation rules are relevant to current debates because they result in a heavier tax burden for U.S. companies under current law than the recently negotiated global minimum tax proposal.

The U.S. approach to taxing foreign profits is to include a portion of those profits in U.S. taxable income and provide a credit for taxes paid on it. This foreign tax credit (FTC) is subject to a limitation of foreign taxable income multiplied by the U.S. corporate tax rate. Under this approach, the U.S. is only supposed to tax foreign profits if those profits faced a foreign tax rate lower than the U.S. rate; if they faced a higher tax rate, this approach is not intended to subject those profits to any residual U.S. taxes.

Expense allocation rules thwart this intended effect. These rules require that certain domestic expenses—interest expense, research and development (R&D) expenses, and stewardship expenses (expenses related to overseeing foreign investments)—be partially reclassified as foreign expenses. This in turn reclassifies some of foreign taxable income as domestic taxable income. This doesn’t change total taxable income; the only effect is to reduce the foreign tax credit limitation. If the tax rate on foreign profits is low enough, the limitation does not constrain the FTC, so expense allocation has no effect; but when the foreign tax rate is higher, the foreign tax credit limitation binds, and expense reallocation reduces the FTC. Thus, for firms facing high foreign tax rates relative to the U.S. tax rate, expense allocation rules impose double taxation on part of those foreign profits.

Prior to 2018, expense allocation rules were not particularly important. The U.S. imposed a high corporate tax rate of 35 percent on a narrow slice of foreign profits when repatriated as dividends; at such a high tax rate, the foreign tax credit limitation had little effect. The Tax Cuts and Jobs Act (TCJA) of 2017 reformed our international tax rules, in part by switching from this high tax on a narrow base to a low tax of 10.5 percent on a larger income base, with a tax credit for 80 percent of foreign taxes on that income. This new tax, the Global Intangible Low-Taxed Income (GILTI) provision, was intended to impose a minimum tax only affecting firms with low effective foreign tax rates. For firms with foreign tax rates above 13.125 percent, the foreign tax credit limitation binds, so these firms were not supposed to be affected by GILTI. But this planned design overlooked how GILTI interacts with expense allocation. Expense allocation imposes double taxation on the foreign profits of firms facing already relatively high foreign tax rates. This transforms GILTI from a minimum tax to a surtax on foreign profits. Moreover, because the foreign tax credit limitation typically binds under GILTI, expense allocation has become much more distortionary.

So why do we have expense allocation rules? They are based on the principle that multinationals can use some domestic activities to support their foreign activities, such as borrowing in the U.S. to finance investment abroad. Because of the high U.S. tax rate before the TCJA, multinationals had strong incentives to locate a disproportionate share of their deductions in the U.S. as part of shifting profits to lower-tax countries.

This justification for expense allocation may be reasonable for stewardship expenses, which generally include management and oversight costs related to foreign investments, but not for research and development expenses. As President Reagan said, “If you want more of something, subsidize it; if you want less of something, tax it.” We generally try to encourage more research and development in the U.S.; that’s why we subsidize it through the R&D credit. It doesn’t make sense to also penalize spending on R&D. In fact, a new tax reform proposal from Senators Ron Wyden (D-OR), Mark Warner (D-VA), and Sherrod Brown (D-OH) would repeal expense allocation for R&D and stewardship.

On the other hand, reducing the tax advantage for interest expense could be good policy, as multinationals can shift profits out of high-tax countries by locating a larger share of their debt in those countries. A potentially reasonable approach to countering this would require that multinationals pool their global interest expenses and allocate these in proportion to their assets in the U.S. and abroad. This method—known as worldwide expense allocation—was supposed to take effect in 2021, but the American Rescue Plan canceled it. Instead, multinationals must use the “water’s edge” method, which effectively assumes that multinationals have no foreign interest expense.

But more importantly, interest expense allocation is an inferior tool to address overborrowing. The TCJA created a thin-capitalization rule that restricts net interest to 30 percent of earnings before interest, taxes, depreciation, and amortization (EBITDA), with this tightening to 30 percent of earnings before interest and taxes (EBIT) in 2022. Unlike expense allocation, thin capitalization rules are common in many countries and consistent with the recommendations of the OECD’s Base Erosion and Profit Shifting project. Given that we are already addressing overborrowing using this thin-capitalization rule, additional penalties from interest expense allocation may be overkill.

Table 1 below—from a recent Tax Foundation analysis of options for international tax reform—presents estimates of how much revenue is raised by expense allocation. Restoring worldwide interest expense allocation reduces the federal corporate income tax liabilities of U.S. multinationals by $23.3 billion over a decade. Repealing interest expense allocation entirely reduces their tax liabilities by $66.8 billion over a decade, and repealing R&D expense allocation reduces these liabilities by $78 billion. Due to a lack of data, we cannot model stewardship expenses. These are not trivial amounts; expense allocation rules impose substantial tax liabilities.

Table 1. Effects of Expense Allocation on Federal Corporate Income Tax Liabilities of U.S. MNEs
Change in federal corporate income tax liabilities of U.S. multinationals
Policy 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 Total
Restore worldwide interest expense allocation -1.8 -1.9 -2.1 -2.2 -2.4 -2.4 -2.5 -2.6 -2.6 -2.7 -23.3
Eliminate interest expense allocation -5.4 -5.7 -6.0 -6.4 -6.7 -6.9 -7.1 -7.3 -7.5 -7.7 -66.8
Eliminate R&D expense allocation -6.4 -6.7 -7.1 -7.5 -7.8 -8.1 -8.3 -8.5 -8.7 -8.9 -78.0

Notes: This table presents the change in federal corporate income tax liabilities of U.S. multinationals in billions of dollars for each provision. All estimates include profit shifting in response to the average tax rate differential with a semi-elasticity of 0.8.

Source: Cody Kallen, “Options for Reforming the Taxation of U.S. Multinationals,” Tax Foundation, Aug. 12, 2021, https://taxfoundation.org/us-multinational-tax-reform-options-gilti/, Table 9.

We can also consider how large these tax burdens are on a relative scale. Because expense allocation only matters for firms with allocable expenses and foreign profits, we can think of the higher tax liabilities from expense allocation as a surtax on foreign profits or as a surtax on domestic expenses. Table 2 presents the 2022 tax wedges from interest and R&D expense allocation as shares of these expenses and as shares of the total foreign profits of U.S. multinationals.

Table 2. Burdens of Expense Allocation
Tax burden of R&D expense allocation, as a percent of…
US R&D Foreign profits
1.13 0.23
Tax burden of interest expense allocation, as a percent of…
Interest paid Foreign profits
1.19 0.38

Notes: This table presents the change in revenue from repealing each type of expense allocation as shares of the relevant domestic expense, and as shares of total foreign profits, which include CFC profits, direct passive foreign income, branch income, and other foreign income of U.S. multinationals.

As a share of domestic R&D by U.S. multinationals, expense allocation imposes a 1.13 percent surtax on this R&D spending. By comparison, according to IRS data from 2014, the total research tax credit for corporations was only 5.5 percent of total qualified research expenses. In other words, R&D expense allocation negates approximately one-fifth of the tax value of the R&D credit for U.S. multinationals. Expense allocation imposes a similar burden as a share of gross interest paid, at 1.19 percent. As shares of total foreign profits, these tax burdens are much smaller.

Finally, we consider how expense allocation would interact with the OECD’s Pillar 2 proposal for a minimum tax. The proposal calls for a 15 percent minimum tax rate, calculated on a per-country basis, with carryforwards and timing adjustments. It would also provide substance carveouts of 7.5 percent of tangible assets and payroll for the first five years, and 5 five percent of these thereafter.

Formally, the OECD’s proposal describes Pillar 2 as a top-up tax, meaning a residual tax on foreign profits to bring the effective tax rate on those profits up to 15 percent. The proposal itself does not mention expense allocation, as such rules are not common in other countries. One potential method of implementing a top-up tax is to include foreign income (in excess of the substance carveouts) and provide a foreign tax credit, as in the standard U.S. approach to taxing foreign profits. In the absence of expense allocation rules, this would achieve the intended top-up tax. However, with expense allocation rules, U.S. multinationals would still owe residual tax on profits from countries with effective tax rates above 15 percent. In these countries, the foreign tax credit limitation would bind, and so expense allocation would make such an implementation of Pillar 2 a surtax, rather than a minimum tax.

A recent Tax Foundation analysis considered modifying GILTI to resemble Pillar 2 by setting the GILTI inclusion rate to 5/7 (to achieve a 15 percent rate), repealing the 20 percent haircut on the GILTI foreign tax credit, calculating this on a per-country basis, allowing GILTI foreign tax credit carryforwards, and switching to the OECD’s exemptions for tangible assets and payroll. Combined, these raise the federal corporate income tax liabilities of U.S. multinationals by $106 billion over a decade. However, if we also repeal expense allocation, this instead would lose $41 billion.

Table 3. Effects of Pillar 2 Compliance on Federal CIT Liabilities of U.S. MNEs
Change in federal corporate income tax liabilities of U.S. MNEs from realistic proposal, $billions
Provision 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 Total
15% GILTI rate 12.4 13.2 14.2 15.1 6.8 7.0 7.2 7.4 7.6 7.8 98.8
 + Remove GILTI FTC haircut -2.8 -3.0 -3.2 -3.4 -3.6 -3.7 -3.8 -3.9 -4.0 -4.1 -35.7
 + Country-by-country GILTI FTC 7.2 7.7 8.3 8.9 9.2 9.5 9.8 10.1 10.3 10.6 91.7
 + Allow GILTI FTC carryforwards 0.0 -2.4 -3.7 -4.7 -5.4 -6.0 -6.5 -6.9 -7.3 -7.7 -50.8
 + Use OECD QBAI/payroll exemptions -0.4 -0.3 -0.3 -0.3 -0.2 0.8 0.7 0.7 0.7 0.7 2.1
 + Repeal expense allocation -11.8 -12.6 -13.5 -14.3 -14.7 -15.2 -15.7 -16.1 -16.5 -16.9 -147.2
Total with expense allocation 16.4 15.2 15.3 15.6 6.7 7.5 7.4 7.3 7.2 7.3 106.0
Total without expense allocation 4.6 2.7 1.9 1.4 -8.0 -7.7 -8.3 -8.8 -9.2 -9.6 -41.2

Notes: This table presents the change in federal corporate income tax liabilities of U.S. multinationals in billions of dollars for each provision. All estimates include profit shifting in response to the average tax rate differential with a semi-elasticity of 0.8.

Source: Cody Kallen, “Options for Reforming the Taxation of U.S. Multinationals,” Tax Foundation, Aug. 12, 2021, https://taxfoundation.org/us-multinational-tax-reform-options-gilti/, Table 5; Tax Foundation’s Multinational Tax Model.

Expense allocation would make this U.S. version of Pillar 2—if implemented using our usual approach—more stringent than Pillar 2 as implemented by countries without expense allocation. The current-law combination of GILTI and expense allocation is tougher than Pillar 2 without expense allocation. 

Source