Most important findings
-
We analyze four options for changing US corporate taxation: The Biden administration's full proposal increases corporate tax liabilities of US multinationals by $ 1.37 trillion over a decade; a partial version brings in $ 580 billion; GILTI's Pillar 2 compliance brings in $ 137 billion; and a revenue-neutral option to fix unintended issues with GILTI.
-
All four of these options would increase taxes on US multinationals' overseas activities.
-
The impact of the Biden administration's proposals would be most extreme and would add a 7.7 percent surcharge to the overseas profits of U.S. multinationals on top of an average overseas tax rate of 12.5 percent, resulting in a combined tax rate of 20.2 Percent leads. This would be significantly higher than the combined tax rates under applicable law (15.3 percent) and according to the other proposals (16.3-17.6 percent).
-
Both the full and partial versions of the Biden government's proposals would increase the profits shifted online from the US. If GILTI was similar to the second pillar, it would have less of a profit shifting impact.
-
Other countries raising their effective tax rates to 15 percent in response to the OECD's proposed minimum tax would reduce U.S. corporate tax revenues significantly after all of these proposals, but much less under current law.
introduction
US multinationals represent an important segment of business activity in the United States. Although only a small proportion of the total number of companies, these companies are disproportionately larger than purely domestic companies. According to data from the Bureau of Economic Analysis (BEA), US multinational corporations contributed 41.5 percent of private company capital spending in 2018 and employed 28.6 million people in the US percent of corporate research and development in the US, generating 76, 2 percent of all company income. From the 2016 IRS corporate tax data, they contributed 63.6 percent of all federal corporate tax revenue.
In contrast to domestic companies, multinational companies have more flexibility to move their profits to countries with lower taxes and to invest in jurisdictions with more favorable tax treatment. The Tax Cuts and Jobs Act (TCJA) of 2017 reformed the taxation of foreign profits of US multinational corporations with the aim of reducing or eliminating incentives to shift profits, with an emphasis on the provisions of the Global Intangible Low-Taxed Income (GILTI) . However, some aspects of the TCJA's international tax regulations have been criticized, including the unexpected interactions between the regulations. The following overview explains these issues in more detail.
In this context, it is important to understand how proposals to change international tax rules would affect US multinational corporations. The Biden government's Made in America Tax Plan would significantly increase taxes on foreign and domestic activities of US multinationals, and Congress might consider similar approaches to increasing revenue. (1) Meanwhile, the Organization for Economic Cooperation and Development (OECD.)) Has put forward its proposals to reshape the international tax system for large multinational companies (MNEs). (2) The OECD proposal consists of two parts: Pillar 1 would locate some taxable profits at the location of customers of a company, and Pillar 2 would impose a minimum tax of at least 15 percent per country. In the overview we go into the various proposals for the reorganization of international tax regulations.
In a previous paper, the Tax Foundation's new multinational tax model was used to examine the potential impact of the Made in America Tax Plan on US multinationals and domestic activities of US multinationals. While it would reduce offshoring incentives, as the administration has claimed, these incentives are low under current law and this effect would therefore be trivial. The plan would impose a significant surcharge on the overseas activities of US multinationals, putting them at a competitive disadvantage vis-à-vis foreign companies. However, it would further increase the tax rate on domestic activities, which would result in a net income increase from the US.
This paper uses an expanded and improved version of the multinational tax model. This new version includes separate calculations by industry and can model more tax proposals, as explained in the Methodology section below.
We use this expanded model to analyze various tax restructuring proposals for US multinational corporations. We focus on four main options:
- Biden Government Proposals
- A partial version of the administration's proposals
- A sales-neutral solution to problems with GILTI
- Make GILTI look like pillar 2
The analysis of the Biden government proposals provides an update to the previous analysis which includes modeling improvements and recent changes to the government proposals. The partial version reflects that the government's massive tax hikes may have difficulty getting Congress approved. (4) Accordingly, this proposal provides for more moderate changes to the corporate tax rate and international tax regulations. The third option addresses issues in the design of GILTI that result in higher tax burdens than expected. This option negotiates the reduction of these unintended tax burdens in exchange for a higher tax rate for GILTI in order to achieve a revenue-neutral policy over the 10-year budget window. Finally, our fourth option looks at the implications of aligning GILTI rules with the minimum taxes that would be levied under the OECD's proposal.
The Biden administration's proposals would increase US multinationals' corporate tax liabilities by $ 1.37 trillion over a decade, and the partial version would increase that tax liability by $ 580 billion over a decade. Aligning GILTI with Pillar 2 of the OECD proposal would raise $ 137 billion over a decade.
Unlike most other countries, the US levies substantial residual taxes on the overseas operations of its multinational corporations, which we estimate under current law to be 2.9 percent of those overseas profits, in addition to overseas taxes on those operations. All four of the options outlined above would increase this remaining US tax, but the Biden administration's proposal would be the most extreme and would increase it to 7.7 percent. This would be a significant disadvantage for US companies operating overseas; The same activities would be taxed 62 percent more under US ownership than under foreign ownership.
We are also examining possible behavioral responses to these tax changes. The full and partial versions of the Biden government's proposals would increase profit shifting from the US online. The revenue-neutral GILTI Fix would reduce profit shifting, and compliance with Pillar 2 would slightly increase profit shifting.
It is also important to consider the behavior of other countries. If low-tax countries raise their effective corporate tax rates to 15 percent – as envisaged in the OECD proposal – it has two potential effects on US tax revenues. Higher foreign tax rates reduce the incentive to shift profits out of the US, thereby increasing the domestic corporate tax base. However, because the US levies residual taxes on these overseas profits, higher overseas taxes on US multinationals are partially offset by a higher overseas tax credit (FTC). The second effect dominates the network; If foreign countries raise their tax rates to a minimum of 15 percent, the US will lose tax revenue. This effect is small under current law, but much greater with all four options.
Finally, we provide estimates of the impact of various options to change taxes on US multinational corporations on corporate tax liabilities as a tool for policy makers considering reforming these tax rules.
Overview of international tax policy
In the past, most countries' tax systems have generally relied on source-based taxation, where taxes are levied at the place where productive activity takes place. However, until 2017, the US system relied on a mixture of withholding and residency taxation. The old US system taxed both foreign and domestic companies on the profits made in the US, but also imposed residual taxes on the foreign profits of US multinationals. US multinational corporations' profits recorded in their controlled foreign corporations (CFCs) could be included in the parent company's taxable income in two ways: CFC passive profits – such as interest and royalty income – were automatically subscribed to taxable via subsection Income of their parent companies included F-rules, but active CFC profits were only taxable if they were returned to the parent company as dividends. By not paying dividends to their parent companies, CFCs could potentially defer these US taxes indefinitely.
In contrast to domestic corporations, multinational corporations have the flexibility to take advantage of cross-country tax differentials to reduce their tax burden, for example by performing profit shifting to move the location of reported profits from high-tax countries to low-tax countries. Profit shifting is often based on intangible assets such as patents. (5) The physical location of an idea is irrelevant to its use, but the legal location of the legal entity that owns the patent matters to the location of taxation of the profits generated by it. By owning patents and other intangible assets owned by affiliates in a tax haven or other low-tax country, a company could avoid taxes on much of its profits. In addition to reallocating profits through intangible assets, companies could often shift profits beyond the location of debt by focusing borrowing on affiliates in high-tax jurisdictions where the tax shield from debt is greater. (6)
In addition to shifting profits to low-tax countries, companies can also choose to locate discreet investments to reduce their tax burden. Devereux and Griffith found that this comprehensive location decision for profit margin relies on Effective Average Tax Rates (EATRs), a forward-looking measure of how much taxes will be levied on a profitable project over the life of that project. (7)
In addition to these general multinational responses to tax policy, the U.S. tax system faced two specific criticisms prior to 2017. First, the tax penalty for repatriating profits to the US parent company created a lockout effect, preventing US multinationals from accessing foreign profits “trapped” in their CFCs without incurring significant tax liabilities. (8) Also, the US corporate tax rate was changed of 35 percent plus state taxes, well above those in other developed countries; the average statutory corporate income tax rate of the OECD was 24 percent in 2017. (9)
This high tax rate exacerbated the incentives for profit shifting and offshoring, and the imposition of this high tax rate on some overseas operations by US multinationals put it at a competitive disadvantage. This competitive disadvantage led to a wave of inversions in which US multinationals took advantage of mergers with foreign firms to move their legal residence to countries with lower taxes. (10)
Current taxation of US multinationals
The TCJA attempted to address these issues with three major changes to the taxation of US multinationals. First, it lowered the corporate tax rate from 35 to 21 percent. This has significantly reduced the incentives to shift profits and the competitive disadvantage of the US tax system. Including state and local taxes, the combined U.S. corporate tax rate of 25.8 percent in 2020 was close to the OECD average of 23.5 percent. Second, the taxation of active CFC income only when returned to the parent company has been switched to taxing a portion of that income when it is earned, but at a lower rate, applying the new rules on Global Intangible Low Taxed Income (GILTI) became. Third, a lower tax rate has been established on intangible income, which in the US is reported through Foreign Intangible Income Deduction (FDII).
The GILTI and FDII regulations should significantly reduce or eliminate profit shifting incentives for US multinational corporations. GILTI should introduce a minimum tax on intangible income posted abroad and FDII provides a reduced tax rate on foreign source income posted in the U.S. To avoid complications in defining intangible income, both provisions instead exempted 10 percent of property, plant and equipment, an assumed normal return on these assets; This is known as the Qualified Business Asset Investment (QBAI) exception. GILTI aimed at a minimum tax on this income of between 10.5 and 13.125 percent (to 13.125 to 16.4 percent by 2026) with a 50 percent deduction (37.5 percent from 2026) and FDII at a reduced tax rate of 13.125 . to collect percent (increase to 16.4 percent in 2026) with a deduction of 37.5 percent (21.875 percent from 2026).
These two provisions significantly reduced profit shifting from the US. A previous analysis by the U.S. Tax Foundation found that the TCJA essentially eliminated the tax benefit of shifting profits overseas. (12) A recent analysis of the Penn Wharton Budget Model found that profit shifting behavior has also decreased as multinational corporations of Bringed at least $ 140 billion in profits to the US from 2018 to 2020. (13)
Although profit shifting has been significantly reduced, the design of the GILTI provisions has been criticized on several occasions. (14) The von Biden government has inadvertent incentives for offshoring property, plant and equipment created for them by the 10 percent exemptions in FDII and GILTI, Heavily criticized. (15) Essentially, by locating physical capital with a return of less than 10 percent abroad instead of in the USA, the higher real capital abroad reduces the CFC income subject to GILTI, while the lower real capital in the USA reduces the FDII – Eligible domestic income increases deduction. A previous analysis of tax incentives for U.S. multinational corporations found that forward-looking effective average tax rates on capital investments show this tax benefit for offshoring from the QBAI exemptions, but it is small and sensitive to assumptions about return on investment. (16) Additionally, this offshoring tax benefit is now much less than it was before the TCJA law.
The Biden government has also criticized cross-crediting under GILTI, where high taxes in high-tax countries can offset low taxes paid in low-tax countries. (17) However, the ability to pool foreign taxes across countries is a deliberate feature of GILTI so that it targets companies that earn low average tax rates on their overseas profits. By pooling foreign taxes when calculating the GILTI Foreign Tax Credit (FTC), the considerable compliance costs associated with a country-specific minimum tax are avoided.
Other criticisms of GILTI focused on its interaction with indirect cost allocation rules. Based on the rationale that some domestic spending supports the overseas operations of US multinationals, these regulations require that some domestic spending be reallocated from reducing domestic taxable income to reducing overseas taxable income. While this doesn't change total taxable income, it tightens the FTC restriction, causing a business to face GILTI obligations even when faced with a high overseas tax rate.
To illustrate this effect, consider the calculation in the following table. Suppose a multinational has $ 100 in domestic income and $ 100 in GILTI from its CFCs. The first column shows the company's tax calculations without GILTI, the next two columns show how GILTI affects this company when it has a low tax rate of 5 percent, and the last two columns when the company has a higher foreign tax rate of 18 Percent has. Let's see how tax liabilities change when $ 10 of domestic spending is allocated to GILTI income. For this calculation, we ignore the FDII deduction and all credits except the FTC and assume that all CFC income is active.
| No GILTI | GILTI for low-tax companies | GILTI for high-tax law firm | |||
|---|---|---|---|---|---|
| No assignment | With assignment | No assignment | With assignment | ||
|
Entrances |
|||||
| Domestic Income | 100 | 100 | 100 | 100 | 100 |
| GILTI admission | – | 100 | 100 | 100 | 100 |
| Foreign tax rate | – | 5% | 5% | 18% | 18% |
|
GILTI calculation |
|||||
| GILTI taxable income | – | 50 | 50 | 50 | 50 |
| On GILTI. taxes paid | – | 5 | 5 | 18th | 18th |
| Taxable Income | 100 | 150 | 150 | 150 | 150 |
|
FTC calculation |
|||||
| Reallocated effort | – | 0 | 10 | 0 | 10 |
| Income in the GILTI basket | – | 50 | 40 | 50 | 40 |
| Credit limit | – | 10.5 | 8.4 | 10.5 | 8.4 |
| Foreign taxes available for credit | – | 4th | 4th | 14.4 | 14.4 |
| Foreign tax credit | – | 4th | 4th | 10.5 | 8.4 |
|
Tax calculation |
|||||
| Tax before credits | 21 | 31.5 | 31.5 | 31.5 | 31.5 |
| US tax liability | 21 | 27.5 | 27.5 | 21 | 23.1 |
Without GILTI, the company's taxable income comes only from its $ 100 domestic income, resulting in a US tax liability of $ 21 at a tax rate of 21 percent. Under applicable law, a 100 percent inclusion of GILTI generates $ 50 of GILTI's taxable income and total taxable income of $ 150.
When calculating the FTC, the minimum of the foreign taxes available for crediting and the credit limit are selected, which corresponds to the taxable income in the GILTI basket multiplied by the statutory corporate tax rate of 21 percent. The low-tax company paid a 5 percent tax rate on $ 100 from GILTI, of which 80 percent is eligible for the FTC. Because the tax rate is low, the FTC restriction is non-binding, so the FTC is also $ 4. The combination of domestic income, GILTI taxable income of $ 50, and FTC of $ 4 results in a net US tax liability of $ 27.50. Because the FTC restriction is not binding, the indirect cost allocation has no effect.
On the other hand, the highly taxed company, which is theoretically faced with a foreign tax rate of 18 percent, should not have GILTI liability. If GILTI's taxable income is $ 50, the FTC limit, regardless of expense allocation, is GILTI's original tax of $ 10.50. Because foreign taxes of $ 14.40 (80 percent of the 18 percent tax on GILTI's $ 100) are eligible, the FTC restriction is binding. Although GILTI increased the company's initial pre-loan tax liability by $ 10.50, the FTC precisely offsets this, leaving a net tax liability of $ 21. However, if $ 10 of domestic spending is allocated to the GILTI basket, the income reported in the GILTI basket drops to $ 40, reducing the FTC limit to $ 8.40. This company's net US tax liability is $ 23.10, so it has $ 2.10 of US tax on its GILTI income despite being exposed to a high overseas tax rate.
Two methods are available to determine the amount of expenditure that should be reallocated to foreign sources. In the water edge approach, a multinational would distribute domestic spending in proportion to the foreign share of a given base. For example, to allocate the interest expense, a company calculates its total equity in affiliated foreign affiliated companies, divides this by the total assets of the company to get the foreign asset share, and multiplies this foreign share by the domestic interest expense. This calculation implicitly assumes that the company's total interest expenses are incurred in support of foreign activities in Germany, without taking into account interest expenses in the foreign subsidiaries.
Since firms generally borrow both domestically and from their overseas subsidiaries, an alternative approach – global cost allocation – would correct this. With global cost sharing, the company would instead split the total domestic and offshore costs proportionally to the offshore base portion, but the reallocated amount would be that proportional portion of the total cost that exceeds the actual overseas cost. The resulting reallocated expense is then considered in Subsection F, Foreign Branch Income, Foreign Dividend Income, and Intangible Income.
From 2021, companies should be able to choose whether to use the water-edge or the global approach for cost allocation. The American Rescue Plan Act of 2021, however, abolished the worldwide allocation of interest expense; Organizations can continue to use the global method for R&D and stewardship spending. (18)
While cost sharing imposes a residual tax on businesses based on their domestic activities, the lack of carry-forward for unused GILTI FTCs or losses may result in GILTI liability for a business facing a high foreign tax rate for life, but with temporary fluctuations in its overseas tax Profits and Tax Rates. Traditionally, companies are allowed to carry forward foreign taxes beyond the limit and to carry forward losses. A company's income or loss and its overseas taxes can fluctuate from year to year due to temporary timing effects. GILTI does not allow any forwarding of CFC losses or unused foreign taxes (beyond the border). A company with a foreign tax rate fluctuating between 10 and 20 percent owes GILTI taxes in low-tax years, but cannot use all foreign taxes for the FTC in high-tax years.
Proposals from the Biden administration
The Biden administration has argued that US multinationals are not paying their "fair share" of taxes. Accordingly, the government's proposals, both the original version of the Made in America Tax Plan and the proposal outlined in the Green Paper, would significantly increase taxes on the activities of US multinationals. (19) They would raise the corporate tax rate to 28 percent and Increase of the GILTI minimum rate to 21 percent (by increasing the GILTI inclusion rate to 75 percent). These proposals would also tighten GILTI by removing the QBAI exemption for property, plant and equipment and requiring that GILTI and the FTCs for GILTI and overseas branches be calculated on a country-by-country basis.
We believe the lifting of the QBAI exemption will also remove the interest expense, as in the Law to Eliminate Incentives for Outsourcing (pp. 20, 2021). (20) The proposals would also remove the FDII deduction, although the Greens Buch suggested replacing it with an unspecified R&D incentive. Since this replacement has not been described, we do not model it. The Green Paper also suggested applying the Section 265 deduction refusal to tax-exempt foreign income, and this would apply to the "excluded" portion of GILTI (25 percent in the Biden proposal).
All of these provisions would raise taxes for US multinationals. The higher domestic corporation tax rate and the repeal of the FDII would increase marginal tax rates on companies' domestically booked income, and the GILTI rate of 21 percent would increase marginal tax rates on their CFC income. The lifting of the QBAI exemption would broaden the base of GILTI and no longer apply to intangible income only, and along with the lifting of the FDII, it would essentially remove the offshoring incentive that the administration has focused on. The country-specific GILTI calculations would eliminate cross-crediting.
However, the government's proposals would exacerbate GILTI's timing problems and distortions in the distribution of expenditure. With full presentations, multinational corporations could essentially pool their income and losses, as well as their foreign taxes, across countries and over time. According to current law without presentation, they cannot be pooled over time, which leads to temporal distortions. According to the government's proposals, they could not be merged over time or across countries, further exacerbating the tax consequences of temporary shocks.
The Green Paper's proposal to refuse deduction is based on the complexity of the expense allocation rules. A multinational company should assign certain deductions – interest expenses, R&D expenses, and administrative expenses – to activities in its CFCs and overseas offices. The percentage allocated to CFCs is further divided into four categories: Subsection F Income, which is allocated to the passive foreign income basket; Included GILTI income that is assigned to the GILTI basket; excluded GILTI income; and dividends of 20 percent owned by foreign affiliates.
The portion attributable to the excluded GILTI income does not affect the tax liability of the parent company, nor does the portion attributable to these dividends. The Green Paper proposal would instead require that the allowance assigned to these exempt income groups be added back to US taxable income. While this might make sense for excluded GILTI income, for tax-exempt dividends, it runs counter to the way the US tax foreign income of its multinational corporations. The US system has conceptually switched from taxing active CFC income when repatriated as dividends to taxation so that dividends from CFCs should no longer affect their parents' tax liabilities. By denying deductions in relation to the exempted dividend income from CFC, the proposal would contradict the intended design of the tax system and restore part of the blocking effect on CFC profits that the TCJA removed.
An earlier analysis of the Made in America Tax Plan found that the proposal would increase net profit shifting contrary to what the government claims. (21) In essence, increasing the corporate tax rate by 7 percentage points would increase domestic taxes by more than their foreign tax increases Activities, and the repeal of the FDII would exacerbate this. The proposal would essentially remove the offshoring incentives for property, plant and equipment, although the tax increases on CFC activities would put US multinationals at a significant tax disadvantage vis-à-vis foreign owners of the same activities.
Other suggestions
Die Vorschläge der Biden-Regierung sind nicht die einzigen, die geprüft werden. Ein Vorschlag der Senatoren Ron Wyden (D-Oregon), Mark Warner (D-Virginia) und Sherrod Brown (D-Ohio) bietet einen alternativen, weniger extremen Vorschlag zur Steuererhöhung für US-Multis GILTI veranschlagt einen nicht spezifizierten Betrag und würde die QBAI-Ausnahmen für GILTI und FDII aufheben. Um Cross-Crediting zu reduzieren, schlugen sie vor, entweder auf eine länderspezifische Berechnung für GILTI umzustellen oder GILTI durch einen obligatorischen Hochsteuerausschluss nur auf Niedrigsteuerländer anzuwenden.
Über die US-Debatte hinaus hat das OECD Inclusive Framework von mehr als 130 Ländern Vorschläge für das Projekt Base Erosion and Profit Shifting (BEPS) vorgelegt.(23) Diese Vorschläge, die nur für die größten multinationalen Unternehmen gelten würden, bestehen aus zwei Säulen. Säule 1 sieht Verknüpfungsregeln vor, die einige hohe Gewinne multinationaler Unternehmen je nach Standort ihrer Verkäufe besteuern würden. Säule 2 würde eine länderspezifische Mindeststeuer auf die Auslandsaktivitäten multinationaler Konzerne mit Gewinnen über 750 Millionen Euro erheben, mit einer materiellen Befreiung von 7,5 Prozent der Sachanlagen und der Lohnsumme für fünf Jahre und mindestens 5 Prozent danach.
Es gibt zahlreiche Möglichkeiten, GILTI und andere internationale Steuervorschriften zu ändern, um mehr Einnahmen zu erzielen, die Steuerbelastung ausländischer Aktivitäten zu verringern oder unbeabsichtigte Wechselwirkungen mit anderen Bestimmungen zu beheben. Der folgende Abschnitt „Ergebnisse“ zeigt die Einnahmenauswirkungen vieler solcher Reformen und Änderungen in unserer Analyse.
Methodik
Das multinationale Steuermodell ist als eine Reihe repräsentativer MNUs in jeder der 40 Branchengruppen strukturiert, von denen jede eine Reihe von FCKW in 74 Branchen und 48 Regionen (42 Länder und sechs weitere Regionen) besitzt. Die CFCs erzielen Gewinne, zahlen ausländische Steuern und üben Aktivitäten aus, die sich auf die US-Steuerverbindlichkeiten ihrer Muttergesellschaften auswirken. Die MNUs der Muttergesellschaft erhalten Einkünfte aus FCKW sowie aus anderen inländischen und ausländischen Einkommensquellen und zahlen bundesstaatliche Körperschaftssteuern. Das Modell basiert auf Daten aus dem Jahr 2014, der jüngsten Benchmark-Umfrage der BEA zu den Aktivitäten von US-amerikanischen MNUs.
CFC-Berechnungen
Die Berechnungen für FCKW stützen sich auf die Tabellen des IRS zu FCKW, die Informationen wie Gewinne, ausländische Steuern, erhaltene und gezahlte Dividenden und Einkünfte des Unterabschnitts F liefern.(24) Wir ergänzen dies mit Daten der BEA zu realen Aktivitäten von Mehrheitseignern ausländische Tochtergesellschaften (MOFAs), um Sachanlagen, F&E und andere für Steuerberechnungen relevante Maßnahmen zu schätzen.(25) Durch die Kombination dieser Quellen verfügen wir über 3.456 repräsentative FCKW, eine für jedes Land und jede Branche.
Die grundlegenden Berechnungen für FCKW stammen größtenteils aus dem Recht vor dem TCJA. Wir verwenden den Gesamtgewinn und den durchschnittlichen ausländischen Steuersatz in einer gegebenen Länder-Branchen-Betrachtung, um ausländische Einkommensteuern und Gewinne nach Steuern zu berechnen. Wir berechnen auch die Einkünfte und Dividenden des Unterabschnitts F, die an die Muttergesellschaft gezahlt werden, basierend auf den effektiven Sätzen des Unterabschnitts F und den Dividendenrückführungsraten in den Daten.
Aus FCKW berechnen wir auch Artikel für die GILTI-Berechnungen. Wenn GILTI auf globaler Basis berechnet wird (wie nach geltendem Recht), müssen wir nur das getestete CFC-Einkommen für jede CFC-Beobachtung und den darauf gezahlten Steuersatz berechnen. Bei der Berechnung auf Länderbasis (wie in den Vorschlägen der Biden-Regierung) berechnen wir für jede CFC-Beobachtung als immaterielles Einkommen und darauf gezahlte Steuern, und wir berechnen die GILTI-Einbeziehung in das steuerpflichtige Einkommen und die GILTI-FTC, die sich aus jedem Land ergeben -Branchenbeobachtung. Weitere Einzelheiten zu den FTC-Berechnungen finden Sie in der folgenden Erläuterung.
Nach geltendem Recht werden die Berechnungen von GILTI anteilig auf der Grundlage des Eigentumsanteils der Muttergesellschaft an einem CFC berechnet. Dementsprechend skalieren wir bei der Zuordnung von CFC-Steuerposten zu den Eltern diese so herunter, dass die gesamte GILTI-Aufnahme in das Modell mit der in den neu veröffentlichten IRS-Daten für 2018 übereinstimmt.(26)
Berechnung des steuerpflichtigen Einkommens der Eltern
Die Daten für die Muttergesellschaften kombinieren die Ergebnisse der CFC-Berechnungen mit den IRS-Tabellen über Unternehmen, die die FTC beanspruchen.(27) Diese ermöglichen es uns, relevante Variablen nach Branchen zu extrahieren, wie Vermögenswerte, Einnahmen, Gewinne vor Steuern, verschiedene Einkünfte aus ausländischen Quellen (zB ausländische Zinserträge und ausländische Zweigniederlassungen), Gesamtbruttoeinkommen, Abzüge und Anpassungen. Wir ergänzen dies mit Daten der BEA zu den realen Aktivitäten der Muttergesellschaften von US-amerikanischen MNUs, um Sachanlagen, F&E, gezahlte Zinsen und andere Maßnahmen abzuschätzen.(28)
Für jede Muttergesellschaft extrahieren wir zunächst relevante Variablen aus ihren konstituierenden CFCs, wie z. Wenn die GILTI FTC nach Ländern berechnet wird, aggregieren wir die Einnahmen und die FTC von GILTI für jedes Land. Wenn nicht, aggregieren wir das getestete CFC-Einkommen, die Sachanlagen, die gezahlten Zinsen und die Steuern, die auf das getestete CFC-Einkommen gezahlt wurden. Wir verwenden diese, um DII und ausländische Steuern zu berechnen, die auf DII gezahlt werden. If the GILTI rules impose a mandatory high-tax exemption, we restrict the aggregation to come from country-industry observations with foreign tax rates below the specified rate for the high-tax exemption.
We then compute initial taxable income in each of five categories: passive foreign income, which consists of foreign interest income, royalty and license fee income, non-CFC dividends and gross-up, gross-up on subpart F income, and (under pre-TCJA law) CFC dividends and the accompanying gross-up; foreign branch income; active foreign income, which includes foreign service income and other foreign income; taxable income from GILTI; and domestic taxable income.
Using initial taxable income in each basket, we then compute the FDII deduction. We compute deduction eligible income by excluding subpart F income, GILTI income, and foreign branch income. Foreign-derived deduction eligible income also excludes the portion of initial domestic taxable income attributable to domestic sales, the share of which we estimate from the share of sales to domestic and foreign parties by U.S. parent companies. We compute FDII as deduction eligible income less 10 percent of tangible assets of the parent, multiplied by the foreign-derived share of deduction eligible income, if this quantity is positive. We then multiply this by the FDII deduction rate and split this deduction among the passive, other, and domestic taxable income baskets. Both the GILTI and FDII calculations exclude foreign oil and gas extraction income under current law.
We then compute taxable income in each basket as initial taxable income in the basket less its allocated FDII deduction. Totaling the taxable income in each basket gives taxable income before the application of the section 163(j) interest limitation, which constrains the interest paid deduction to interest received plus 30 percent of adjusted taxable income, which we define as taxable income adding back interest paid and subtracting interest received (plus depreciation for 2018-2021). We add back the disallowed portion of interest paid to get total taxable income.
Indirect Expense Allocation Calculations
We model indirect expense allocation for interest expense and for R&D expense, for which we have data from the BEA. Our calculations omit the allocation of stewardship expenses.
To implement reallocation under the water’s edge method, we compute the domestic and foreign basis for reallocation. For interest expense, foreign basis is equity in related foreign affiliates, and domestic basis is total parent assets less this related party equity. For R&D expense, we use gross foreign income of MOFAs for foreign basis and gross income of the U.S. parent company as domestic basis. The share of the expense to reallocate is the total expense multiplied by the foreign basis and divided by total basis.
For elective worldwide expense allocation, we use the same basis for R&D expense, but we use total CFC assets as the foreign basis for interest expense. Instead of allocating only domestic expense, we apportion the combined domestic and foreign expenses in proportion to the foreign share of total basis. The amount to reallocate is the excess (if positive) of this proportionally apportioned expense over the actual foreign expense incurred. Because the worldwide reallocation is elective, the parent chooses whichever method reallocates the smaller amount.(29)
Once we have determined the amount of expenses to reallocate, we split this into four categories in proportion to gross income in each category. The portion allocated to subpart F income is subtracted from the passive basket for the calculation of the FTC limitation, and the portion allocated to foreign branch income is similarly subtracted from the foreign branch basket. Part is allocated to tested CFC income; this amount is then multiplied by the GILTI inclusion rate and subtracted from the GILTI basket, and the remaining amount (net of the GILTI inclusion rate) is allocated to exempt foreign income and does not affect the FTC calculations. The final portion is allocated to CFC dividends and gross-up, which is also considered exempt foreign income and does not affect FTC calculations.
Under the Biden administration’s proposal for the application of section 265 to exempt foreign income, the portions allocated to exempt GILTI income and exempt CFC dividends would be added back to taxable income, and so would increase tax liabilities.
Foreign Tax Credit Calculations
The FTC calculations are applied for each of the passive, branch, other, and GILTI baskets. If the GILTI FTC is calculated on a per-country basis, then a separate FTC calculation is implemented for each CFC.
Multiplying taxable income in each of these baskets (after indirect expense allocation) by the statutory tax rate gives the FTC limitation for the basket. We compute taxes paid on the income in each basket, and adding foreign taxes carried forward from previous years gives total taxes available for credit. For GILTI calculations, the taxes in this basket are 80 percent of the foreign taxes paid on deemed intangible income. While no foreign taxes may be carried forward for GILTI under current law, we consider proposals allowing it.
Using a representative firm to calculate the FTC is problematic. In the actual calculations, a firm claims the lesser of the limitation or the taxes available for credit. With heterogeneous firms, some will be constrained by the limitation, and others by the taxes available for credit; when aggregating across firms, this produces a total FTC smaller than both the aggregated FTC limitation and the aggregated taxes available for credit. Thus, a representative firm systematically overestimates the value of the FTC relative to aggregating over many heterogeneous firms.
To overcome this problem, we can express the discrepancy between the FTC limitation and the total taxes available for credit as a lognormal-distributed random variable with spread parameter . Under this assumption, we can express the expected value of the FTC, aggregated over many firms, as the following.
In this equation, L is the FTC limitation, T is total taxes available for credit, and Φ is the cumulative distribution function for the standard normal distribution. The spread parameter is calibrated to satisfy this equation in the IRS FTC tables for each industry. When the limitation is much larger than the total taxes available for credit, more weight gets put on the taxes available for credit, and vice versa. The log-standard deviation of causes the FTC to fall below both the aggregated limitation and the aggregated total taxes available for credit.
We apply this formula to each foreign income and tax basket to get the FTC for each basket. Combining these gives the total FTC.
Other Calculations
Multiplying total taxable income by the statutory tax rate gives taxes before credits. Subtracting off the FTC and other tax credits gives total federal corporate income tax liabilities of U.S. MNEs.
In addition to calculating tax liabilities, we also allocate parent tax liabilities arising from CFC activities by computing U.S. taxable income from CFC activities (via subpart F, GILTI as of 2018, and dividends received until 2017) and split the associated FTCs.
Profit-Shifting Response
In theory, profit shifting responds to the marginal tax shield from relocating profits across jurisdictions. For example, if a foreign tax rate is 10 percent and the U.S. tax rate is 21 percent, shifting profits to the foreign jurisdiction provides a tax shield of 11 percentage points.
There are two methods to calculate tax shields: statutory tax rates or average tax rates. Statutory tax rates may be more likely to apply at the margin for some profits, and they do not depend on CFC activities like investment that affect the actual tax liabilities. However, statutory tax calculations miss important complications in international taxes. For example, statutory tax rate calculations are not useful for estimating the incentive effects of cross-crediting under GILTI. Moreover, the definition of the tax base can be more important than the tax rate. For example, Ireland’s 12.5 percent statutory corporate tax rate produces an effective rate of approximately 3 percent on the profits of U.S. MNEs booked in Ireland. Many countries also have special provisions, including tax treaties and favorable patent box provisions, that produce lower effective tax rates. Because of these limitations, we use average foreign tax rates when calculating profit-shifting incentives.
For the average tax rate calculations, we need to allocate the parent’s tax liabilities arising from CFC activities to each CFC it owns. If GILTI is calculated on a per-country basis, it is straightforward to compute the CFC’s contribution to its parent’s tax liabilities. If GILTI calculations are pooled across countries, we split GILTI income in proportion to the CFC’s contribution to the parent’s tested CFC income, and we split the parent’s GILTI FTC in proportion to foreign taxes paid on its per-country deemed intangible income. The model also computes inclusions via subpart F, dividends paid, and their gross-up, and it allocates the parent’s FTC based on the contributions of these taxes paid to the parent’s total foreign taxes paid.
For the U.S. domestic tax rate, we use the statutory tax rate net of the FDII deduction, and we estimate the probability of FDII eligibility by dividing the actual FDII deduction by the FDII deduction rate and deduction eligible income in each industry. The domestic tax rate less the average foreign tax rate gives the tax shield. We estimate profit shifting using a semi-elasticity of 0.8 with respect to this tax shield, based on the main estimate from Heckemeyer and Overesch.(30)
While the model can also compute investment responses to tax changes, we do not include those in the estimates in this paper.
Results
Our main results focus on four proposals for changing the taxation of U.S. multinationals. The first is the Biden administration’s proposals as described in the Green Book.(31) This proposal raises the corporate tax rate to 28 percent, raises the GILTI minimum rate to 21 percent by raising the GILTI inclusion rate to 75 percent, repeals the QBAI exemption for GILTI, requires that the GILTI FTC be calculated on a per country basis, repeals the FDII deduction, and applies the section 265 denial of deductions to excluded GILTI income and exempt foreign dividends.
The second proposal represents a partial version of the Biden proposals, reflecting potential difficulties getting the full proposal through Congress. Instead of a 28 percent corporate tax rate, this proposal would use a 25 percent rate. Instead of raising the GILTI inclusion rate to 75 percent and repealing FDII, this proposal accelerates the increase in the GILTI inclusion rate and reduction of the FDII deduction rate from 2026 to occur in 2022. At the 25 percent corporate tax rate, these establish a 15.625 percent minimum rate on GILTI and a 19.53 percent FDII rate (also equal to the upper bound on the GILTI rate ignoring expense allocation). It maintains the repeal of the QBAI exemption. However, instead of switching to a per country GILTI calculation, this proposal would use the mandatory high-tax exemption in the proposal from Senators Wyden, Warner, and Brown. Under this approach, tested CFC income and taxes paid on it in countries with effective tax rates on that income above a certain cutoff would be excluded from taxable income and the FTC. The Senators’ proposal leaves the cutoff rate ambiguous; a lower rate reduces cross-crediting (thus reducing the FTC and raising revenue) but also reduces the revenue raised by indirect expense allocation. We use a 22.5 percent cutoff, consistent with the current optional high-tax exemption set at 90 percent of the statutory corporate tax rate.
The third proposal, the revenue-neutral GILTI fix, would alleviate two issues with the design of GILTI that make it a surtax instead of a minimum tax. It would restore worldwide interest expense allocation, and it would allow full GILTI FTC carryforwards. Because each of these provisions loses revenue, it would compensate for these by raising the minimum GILTI rate to 14.1225 percent (instead of 10.5 percent before 2026 and 13.125 percent thereafter).
The final proposal modeled here makes GILTI resemble the OECD’s Pillar 2 proposal. This would raise the GILTI minimum rate to 15 percent, but it would also require substantial changes to the GILTI FTC: repealing the 20 percent haircut on foreign taxes eligible for the GILTI haircut, allowing GILTI FTC carryforwards, and calculating GILTI and its FTCs on a country-by-country basis. It would also switch from a 10 percent GILTI exemption for tangible assets to using exemptions on tangible assets and on payroll, at 7.5 percent rates on each for five years and 5 percent on each thereafter. Although this captures the essence of the OECD’s proposal, there are three potential discrepancies. First, the OECD’s proposal would rely on financial statement income rather than tax income, with unspecified adjustments to reflect timing differences. Second, the OECD’s proposal does not clarify whether the substance carveout for tangible assets and payroll is net of interest paid, as is the case under GILTI, or not. For modeling this, we assume it is net of interest. Third, the OECD proposal would allow losses to be carried forward. We do not have the data to model this change, so the true effect of the proposal may be smaller than our estimated results.
We also consider a set of other options for changing the level of GILTI calculations, changing the structure of the GILTI exemption, and changing indirect expense allocation rules.
Results for Main Options
We first focus on the change in federal corporate income tax liabilities of U.S. multinationals. This is not the same as an actual revenue score, which also includes revenues from non-multinationals and should also reflect offsets from reduced capital gains and dividends for the shareholders of multinationals. A full score should also include the full set of firm responses to tax policies; the estimates here only include the profit-shifting response to tax rate differentials.
Table 2 presents the results for the full Biden administration proposal. The higher corporate tax rate would raise multinationals’ tax liabilities by $753 billion. Raising the GILTI rate to 21 percent would raise $192 billion. At a 21 percent GILTI rate, repealing the QBAI exemption would raise $146 billion, and implementing country-by-country GILTI calculations would raise a further $102 billion. Repealing the FDII deduction would raise their federal tax liabilities by $89 billion. The section 265 denial of deductions against exempt income would raise a further $90 billion. In total, this proposal would raise taxes on U.S. multinationals by $1.37 trillion over a decade.
| Change in Federal Corporate Income Tax Liabilities of U.S. MNEs from Biden Administration Proposal, Billions of Dollars | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Provision | 2022 | 2023 | 2024 | 2025 | 2026 | 2027 | 2028 | 2029 | 2030 | 2031 | Total |
| 28% CIT rate | 55.8 | 60.1 | 64.9 | 69.2 | 77.8 | 80.4 | 82.8 | 85.0 | 87.2 | 89.6 | 752.9 |
| + 21% GILTI rate | 22.3 | 23.8 | 25.6 | 27.2 | 14.5 | 14.9 | 15.4 | 15.8 | 16.2 | 16.6 | 192.2 |
| + No QBAI exemption | 11.5 | 12.3 | 13.2 | 14.1 | 14.8 | 15.2 | 15.7 | 16.1 | 16.5 | 16.9 | 146.4 |
| + Country-by-country GILTI | 7.9 | 8.3 | 8.9 | 9.4 | 10.5 | 10.9 | 11.2 | 11.5 | 11.7 | 12.1 | 102.3 |
| + Repeal FDII | 8.9 | 9.9 | 10.9 | 11.8 | 7.4 | 7.5 | 7.8 | 8.0 | 8.2 | 8.4 | 88.8 |
| + Sec. 265 application | 7.1 | 7.6 | 8.2 | 8.7 | 9.1 | 9.3 | 9.6 | 9.9 | 10.1 | 10.4 | 90.0 |
| Total | 113.5 | 122.1 | 131.8 | 140.5 | 134.0 | 138.3 | 142.4 | 146.1 | 149.9 | 154.0 | 1372.5 |
| Note: 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: Tax Foundation’s Multinational Tax Model. | |||||||||||
Table 3 presents the results for the partial version of the administration’s proposals. Raising the corporate income tax rate to 25 percent raises multinationals’ corporate income tax liabilities by $432 billion over a decade. Accelerating the GILTI and FDII rate changes from 2026 to 2022 raises $56 billion, much smaller than FDII repeal and the 21 percent GILTI rate in the full proposal. Imposing a mandatory high-tax exclusion raises $74 billion, approximately three-fourths of the revenue from switching to a country-by-country GILTI FTC.
Stacked on top of these changes, repealing the QBAI exemption only raises $18 billion. This effect is much smaller than as part of the full Biden administration’s proposal for two reasons. First, the GILTI rate is lower, so changing the GILTI base calculation has a smaller effect. Second, foreign tangible assets of U.S. MNEs are disproportionately located in higher-tax countries relative to the location of profits. Because income from those countries is already being excluded under the mandatory high-tax exclusion, repealing the QBAI exemption has a much smaller effect.
In total, this proposal would raise U.S. multinationals’ federal corporate tax liabilities by $580 billion over a decade, a substantial tax hike but much less extreme than the full Biden administration proposal.
| Change in Federal Corporate Income Tax Liabilities of U.S. MNEs from Partial Proposal, Billions of Dollars | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Provision | 2022 | 2023 | 2024 | 2025 | 2026 | 2027 | 2028 | 2029 | 2030 | 2031 | Total |
| 25% CIT rate | 32.0 | 34.5 | 37.3 | 39.7 | 44.6 | 46.1 | 47.5 | 48.7 | 50.0 | 51.3 | 431.7 |
| + Accelerate GILTI/FDII rate changes | 12.2 | 13.2 | 14.3 | 15.2 | 0.3 | 0.2 | 0.2 | 0.2 | 0.1 | 0.1 | 56.0 |
| + Mandatory high-tax exemption at 22.5% | 5.8 | 6.2 | 6.7 | 7.1 | 7.4 | 7.6 | 7.9 | 8.1 | 8.3 | 8.5 | 73.6 |
| + No GILTI QBAI exemption | 1.4 | 1.5 | 1.6 | 1.8 | 1.8 | 1.9 | 1.9 | 2.0 | 2.0 | 2.1 | 18.1 |
| Total | 51.5 | 55.5 | 59.9 | 63.8 | 54.2 | 55.8 | 57.4 | 58.9 | 60.4 | 62.0 | 579.5 |
| Note: 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: Tax Foundation’s Multinational Tax Model. | |||||||||||
Table 4 presents the results for the revenue-neutral GILTI fix. Restoring worldwide interest expense allocation reduces U.S. multinationals’ federal corporate tax liabilities by $23 billion over a decade, and allowing GILTI FTC carryforwards loses a further $47 billion. To pay for these, we set the GILTI minimum rate to 14.1645 percent, which achieves budget neutrality within the 10-year window. However, such a proposal could violate the Byrd rule, as it raises more revenue in the initial years and loses revenue in later years of the budget window and beyond. This timing pattern comes from the higher GILTI rate being a larger increase relative to the current 10.5 percent than to the 13.125 percent beginning in 2026. Additionally, FTC carryforwards accumulate over the first few years of allowing them before stabilizing.
| Change in Federal Corporate Income Tax Liabilities of U.S. MNEs from Revenue-Neutral GILTI Fix, Billions of Dollars | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Provision | 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 |
| + Allow GILTI FTC carryforwards | 0.0 | -2.0 | -3.1 | -3.9 | -6.2 | -6.1 | -6.2 | -6.4 | -6.6 | -6.9 | -47.3 |
| + 14.1645% GILTI rate | 10.0 | 10.5 | 11.2 | 11.9 | 5.0 | 4.6 | 4.4 | 4.3 | 4.3 | 4.4 | 70.5 |
| Total | 8.2 | 6.6 | 6.0 | 5.8 | -3.5 | -3.9 | -4.3 | -4.7 | -5.0 | -5.2 | 0.0 |
| Note: 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: Tax Foundation’s Multinational Tax Model. | |||||||||||
Finally, Table 5 presents the results for making GILTI resemble the OECD’s Pillar 2 proposal. Raising the minimum GILTI rate to 15 percent would raise $99 billion, but removing the 20 percent haircut on foreign taxes paid on GILTI loses $36 billion. On net, a 15 percent minimum tax instead of the GILTI bands under current law (10.5 to 13.125 percent currently, and 13.125 to 16.4 percent beginning in 2026) raises U.S. multinationals’ federal corporate income tax liabilities by $63 billion over a decade. Calculating GILTI on a country-by-country basis would raise $125 billion, more than 40 percent of which would be lost to allowing GILTI FTC carryforwards. Switching to the OECD’s tangible asset and payroll substance carve-outs instead of the current 10 percent exemption for tangible assets has a small net effect, losing a little revenue during the five years of 7.5 percent exemptions and raising a little revenue under the subsequent 5 percent exemptions. On net, bringing GILTI into compliance with Pillar 2 would raise $137 billion over a decade. Note that this modeling omits loss carryforwards, which would reduce revenue from Pillar 2 compliance.
| Change in Federal Corporate Income Tax Liabilities of U.S. MNEs from Pillar 2 Compliance, Billions of Dollars | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 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 | 9.7 | 10.3 | 11.1 | 11.8 | 12.8 | 13.2 | 13.6 | 13.9 | 14.3 | 14.7 | 125.4 |
| + Allow GILTI FTC carryforwards | 0.0 | -2.6 | -4.0 | -5.0 | -5.8 | -6.5 | -7.0 | -7.4 | -7.8 | -8.2 | -54.3 |
| + Use OECD QBAI/payroll exemptions | -0.4 | -0.2 | -0.2 | -0.2 | -0.1 | 0.9 | 0.8 | 0.7 | 0.7 | 0.7 | 2.7 |
| Total | 19.0 | 17.7 | 17.9 | 18.3 | 10.0 | 10.9 | 10.8 | 10.7 | 10.8 | 10.9 | 136.9 |
| Note: 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: Tax Foundation’s Multinational Tax Model. | |||||||||||
Industry-Level Analysis
In addition to estimating the changes in firms’ corporate income tax liabilities, we can allocate taxes from subpart F and GILTI to the controlled foreign corporations of multinationals, insofar as they give rise to subpart F and GILTI inclusions and to FTCs. Table 6 presents the average tax rates on CFC activities by industry in 2022, calculated by combining foreign taxes, initial tax liability from subpart F, and GILTI inclusions, net of the GILTI FTC and the subpart F share of the passive FTC. The total foreign tax and the net residual U.S. tax are divided by CFC profits to obtain the average tax rates.
The first column presents the average foreign tax rates on these profits, 12.5 percent on average. All other columns present the combined foreign and residual U.S. tax rates. Under current law, the U.S. imposes a 2.9 percent residual tax of CFC profits. Under the Biden plan this residual U.S. tax would rise to 7.7 percent, resulting in a combined tax rate of 20.2 percent. The partial version of the Biden administration proposal, the GILTI fix, and Pillar 2 compliance would raise this combined tax rate similarly in 2022, to 17.6, 16.3, and 16.9 percent respectively. Note that residual tax from the GILTI fix is much smaller in later years; in 2022, the effect is just the higher GILTI rate and worldwide expense allocation, as firms would begin with no GILTI foreign taxes carried forward.
These average foreign tax rates on CFC activities demonstrate a significant tax disadvantage to U.S. ownership of activities abroad. Under current law, these combined tax rates suggest that the same activities owned by foreign companies facing no residual U.S. tax would have significantly lower tax burdens and thus be more valuable. Under current law, U.S. ownership of foreign profits incurs taxes 23 percent higher than foreign ownership. Under the Biden administration’s proposal, this would almost triple, with U.S. ownership facing 62 percent higher taxes.
| Industry | Foreign | Current law | Biden | Partial | GILTI fix | Pillar 2 |
|---|---|---|---|---|---|---|
|
Mining |
||||||
| Oil and gas extraction, coal mining | 34.6 | 34.4 | 36.6 | 34.8 | 34.4 | 34.5 |
| All other mining | 32.7 | 35.1 | 38.7 | 36.8 | 35.1 | 35.9 |
|
Construction |
30.0 | 31.9 | 34.5 | 33.0 | 32.1 | 32.6 |
|
Manufacturing |
||||||
| Food manufacturing | 23.8 | 25.4 | 29.7 | 27.3 | 25.6 | 26.5 |
| Beverage and tobacco products | 13.3 | 16.1 | 24.0 | 20.1 | 17.6 | 19.2 |
| Paper manufacturing | 25.3 | 26.0 | 30.4 | 28.1 | 26.0 | 27.0 |
| Petroleum and coal products manufacturing | 25.2 | 25.4 | 29.9 | 27.5 | 25.3 | 26.4 |
| Pharmaceutical and medicine manufacturing | 7.4 | 11.4 | 14.6 | 13.2 | 12.1 | 11.6 |
| Other chemical manufacturing | 20.3 | 22.6 | 27.3 | 24.6 | 22.4 | 23.6 |
| Plastics and rubber products manufacturing | 17.7 | 18.8 | 22.7 | 20.6 | 19.1 | 19.9 |
| Nonmetallic mineral product manufacturing | 20.6 | 21.8 | 31.7 | 27.5 | 21.8 | 25.2 |
| Primary metal manufacturing | 31.6 | 32.4 | 39.8 | 35.9 | 32.3 | 34.9 |
| Fabricated metal products | 22.7 | 24.4 | 27.6 | 25.9 | 24.4 | 24.8 |
| Machinery manufacturing | 29.0 | 30.7 | 34.4 | 32.1 | 31.0 | 31.5 |
| Computer and electronic product manufacturing | 4.4 | 9.6 | 18.0 | 14.2 | 11.4 | 12.9 |
| Electrical equipment, appliance and component manufacturing | 11.0 | 15.3 | 26.0 | 21.0 | 17.8 | 19.1 |
| Motor vehicles and related manufacturing | 23.5 | 24.5 | 29.7 | 27.8 | 24.5 | 26.5 |
| Other transportation equipment manufacturing | 22.5 | 25.7 | 31.0 | 28.1 | 25.3 | 26.3 |
| Other manufacturing | 18.4 | 20.7 | 24.9 | 22.6 | 21.1 | 21.9 |
|
Wholesale trade |
||||||
| Machinery, equipment, and supplies | 19.1 | 23.2 | 28.0 | 24.9 | 24.4 | 24.5 |
| Other miscellaneous durable goods | 18.8 | 22.7 | 27.6 | 24.1 | 24.2 | 24.2 |
| Drugs, chemicals, and allied products | 5.8 | 9.4 | 14.1 | 11.6 | 10.9 | 11.1 |
| Groceries and related products | 20.5 | 22.9 | 27.0 | 24.7 | 22.6 | 23.7 |
| Other miscellaneous nondurable goods | 19.5 | 21.3 | 26.8 | 24.0 | 22.3 | 23.3 |
|
Retail trade |
23.5 | 24.7 | 31.1 | 27.6 | 25.2 | 27.1 |
|
Transportation and warehousing |
22.9 | 23.5 | 30.0 | 27.0 | 23.5 | 25.7 |
|
Information |
||||||
| Publishing (except internet) | 9.6 | 13.2 | 16.4 | 14.9 | 14.3 | 13.4 |
| Motion picture and sound recording | 29.0 | 29.9 | 31.9 | 30.6 | 30.1 | 30.2 |
| Other information | 10.6 | 13.3 | 21.0 | 17.5 | 14.3 | 16.4 |
|
Finance and insurance |
||||||
|
Nondepository credit intermediation |
21.3 | 23.7 | 28.4 | 25.9 | 24.4 | 25.2 |
|
Securities, commodity contracts, and other financial investments |
15.5 | 17.9 | 23.1 | 20.3 | 19.1 | 19.7 |
|
All other finance industries |
5.6 | 11.8 | 18.9 | 15.1 | 13.9 | 14.6 |
|
Insurance and related activities |
22.0 | 23.6 | 26.5 | 24.2 | 24.6 | 24.7 |
|
Real estate and rental and leasing |
13.5 | 14.3 | 22.4 | 18.9 | 14.4 | 18.0 |
|
Professional, scientific, and technical services |
16.7 | 21.0 | 26.4 | 23.3 | 22.0 | 22.4 |
|
Management of holding companies |
7.5 | 10.1 | 13.5 | 11.7 | 11.0 | 11.2 |
|
Administrative and support and waste management and remediation |
12.8 | 14.1 | 16.9 | 15.3 | 14.7 | 15.0 |
|
Arts, entertainment, and recreation |
28.0 | 28.1 | 37.0 | 32.2 | 28.1 | 31.5 |
|
Accommodation and food services |
20.0 | 20.3 | 26.7 | 23.8 | 20.3 | 22.8 |
|
Other industries |
20.2 | 20.7 | 27.6 | 24.4 | 20.7 | 23.5 |
|
All industries |
12.5 | 15.3 | 20.2 | 17.6 | 16.3 | 16.9 |
| Note: This table presents the average tax rates on CFC profits. The first column only includes foreign taxes as a share of CFC profits. All other columns include foreign taxes and residual U.S. taxes from subpart F and GILTI inclusions net of the GILTI FTC and the share of the passive FTC attributable to subpart F income. Major industries are bolded. All estimates include no profit-shifting responses, to avoid conflating that effect on average tax rates. | ||||||
| Source: Tax Foundation’s Multinational Tax Model. | ||||||
Not all industries are equally affected; high-tech industries are disproportionately among those facing the highest tax hikes. The most heavily affected industry—electrical equipment, appliance and component manufacturing—would face a 10.7 percentage point tax hike on its CFC profits under the full Biden administration proposal. The computer and electronic product manufacturing industry would face an 8.4 percentage point tax hike, and information services other than publishing and recording would face a 7.7 percentage point tax hike.
However, industries not often associated with profit shifting are also among the most affected. Nonmetallic mineral product manufacturing is the second most-heavily affected industry under the Biden plan, facing a 9.9 percentage point tax hike. Real estate, rental and leasing is also heavily affected, facing an 8 percentage point tax hike. Both of these industries rely relatively heavily on tangible assets and are thus more exposed to repeal of the QBAI exemption.
Profit-Shifting Responses to International Tax Changes
The main estimates include a moderate profit-shifting response to higher U.S. taxes on foreign activity. The Biden administration has claimed that their proposal would eliminate profit-shifting incentives, but this is not the case. From Table 6, the full Biden administration proposal would raise the average tax rate on CFC activity by 4.9 percentage points, but a 28 percent corporate tax rate and repeal of the FDII deduction raise the average tax rate on U.S. activity by more than 7 percentage points. On net, this would increase profit shifting out of the U.S.
To consider the magnitude of this effect, Table 7 presents the 10-year change in federal corporate income tax liabilities of U.S. multinationals for each proposal under different profit-shifting assumptions. The first column uses purely static results, with no behavioral responses. The second column uses a semi-elasticity of 0.8 with respect to the average tax rate differential between the U.S. parent and each CFC. This is the main approach we use; the numbers in the second columns match the totals in Tables 2, 3, 4, and 5. The third column uses a larger profit-shifting response estimated by Dowd, Landefeld, and Moore.(32) Using confidential microdata on multinationals and their CFCs, they estimate separate semi-elasticities for CFCs in tax havens and in non-haven countries. For average tax rates, they estimate a semi-elasticity of 0.684 in non-haven countries and a much larger semi-elasticity of 4.16 in tax havens.(33) The last two columns present the revenue loss to profit shifting under each of these assumptions, relative to the estimates with no profit-shifting responses.
| Change in Federal Corporate Income Tax Liabilities of US MNEs, under Different Profit Shifting Assumptions, Billions of Dollars | |||||||
|---|---|---|---|---|---|---|---|
| Proposal | Change in CIT Liabilities | Loss to Profit Shifting | |||||
| Static | SE = 0.8 | SE from DLM | SE = 0.8 | SE from DLM | |||
| Biden proposal | 1451.1 | 1372.5 | 1214.1 | -78.5 | -237.0 | ||
| Partial Biden | 615.6 | 579.5 | 522.3 | -36.1 | -93.3 | ||
| GILTI fix | -2.5 | 0.0 | 11.4 | 2.5 | 13.9 | ||
| Pillar 2 | 145.0 | 136.9 | 130.7 | -8.1 | -14.3 | ||
| Note: This table presents the 10-year change in federal corporate income tax liabilities of U.S. multinationals in billions of dollars for each proposal. The first three columns present the change in these liabilities from each proposal under different profit-shifting assumptions. The static estimates use no profit-shifting response. The second column uses the moderate profit-shifting response, with a semi-elasticity of 0.8 with respect to the tax rate differential between the U.S. and each CFC. The third column uses the profit-shifting response from Dowd, Landefeld, and Moore (2018), with a semi-elasticity of 4.16 for tax havens and a semi-elasticity of 0.684 for other countries. Tax havens are identified by the Congressional Research Service. The last two columns present the revenue leakage from profit shifting, calculated by subtracting the static results. | |||||||
| Source: Tax Foundation’s Multinational Tax Model. | |||||||
The full and partial Biden proposals both increase profit shifting. This greater profit shifting reduces U.S. tax revenues, as profits shifted abroad only claim a residual U.S. tax instead of the higher tax if booked domestically. Comparing our main results with the static scenario, profit shifting reduces the domestic tax liabilities of U.S. multinationals under the Biden plan by $79 billion over a decade under the Biden administration proposal, and the U.S. loses $36 billion under the partial version of it. Using the greater tax haven responsiveness from Dowd, Landefeld, and Moore, the full Biden administration proposal loses $237 billion to profit shifting.
On he other hand, the GILTI fix reduces profit shifting. Switching to worldwide expense allocation and allowing FTC carryforwards only cut taxes for firms with binding FTC limitations, which comes from facing higher foreign tax rates. These firms facing higher foreign taxes generally engage in less profit shifting. Meanwhile, the higher GILTI rate raises taxes mostly on firms with low foreign tax rates, which are also those more likely to engage in profit shifting. On net, this proposal raises foreign taxes on firms engaging in more profit shifting and cuts taxes for those that do so less, resulting in a net decrease in profit shifting.
The effect of profit shifting on the scores for Pillar 2 compliance also show net revenue losses due to greater profit shifting, but the revenue losses are much smaller than for the full or partial versions of the Biden administration’s proposals.
Foreign Country Responses to Pillar 2
MNEs are not the only entities that respond to tax changes. Other countries could potentially respond as well by raising taxes. Because the U.S. only imposes a residual tax on CFC profits net of the FTC, an increase in foreign taxes is partially offset by a larger FTC. In response to higher foreign tax rates, U.S. multinationals would also be incentivized to reduce profits shifted to those countries.
Moreover, the OECD’s Pillar 2 proposal imposing a 15 percent country-by-country minimum tax could incentivize low-tax countries to raise their effective tax rates to collect that corporate tax revenue themselves instead of allowing a multinational’s home country to collect it. We model the exposure of the U.S. tax system to this effect by supposing that the average tax rate in each CFC observation were raised to a minimum 15 percent. Table 8 presents this effect on the federal corporate income tax liabilities of U.S. multinationals.
| Change in Federal Corporate Income Tax Liabilities of US MNEs, under Different Foreign Tax Responses, Billions of Dollars | |||
|---|---|---|---|
| Proposal | Current rates | 15% minimum | Difference |
| Current law | 0.0 | -20.4 | -20.4 |
| Biden proposal | 1372.5 | 1214.8 | -157.8 |
| Partial Biden | 579.5 | 502.3 | -77.2 |
| GILTI fix | 0.0 | -62.5 | -62.5 |
| Pillar 2 | 136.9 | -43.5 | -180.4 |
| Note: This table presents the 10-year change in federal corporate income tax liabilities of U.S. multinationals in billions of dollars for each proposal. The first column uses current foreign tax rates. The second column supposes that each country-industry CFC observation faced an average tax rate of at least 15 percent. The third column presents the reduction in federal corporate tax liabilities from this effect. | |||
| Source: Tax Foundation’s Multinational Tax Model. | |||
Conceivably, the higher foreign taxes could increase or decrease U.S. tax revenues. The reduction in profit shifting increases profits booked domestically and taxes on those profits, but the higher foreign taxes assessed result in higher FTCs, reducing tax revenue. In all the scenarios above, the latter effect dominates, and the higher foreign taxes result in net decreases in U.S. tax revenue.
The exposure of U.S. tax revenue to foreign tax hikes depends on how much foreign income is included in U.S. taxable income. As more foreign income is included—such as by repealing the QBAI exemption for GILTI—a larger share of total foreign taxes is included in the FTC calculations. It also depends on how much the FTC limitation constrains the FTC. At a higher corporate income tax rate or higher GILTI rate, the limitation matters less, and so the marginal revenue offset from the FTC is larger. Because indirect expense allocation and the lack of carryforwards for GILTI foreign taxes make the limitation more important, restoring worldwide expense allocation and allowing carryforwards also increase the exposure of the U.S. to foreign tax hikes.
Interestingly, cross-crediting reduces U.S. exposure to tax hikes in low-tax countries, so proposals to reduce cross-crediting—such as country-by-country GILTI calculations and the mandatory high-tax exemption—increase U.S. exposure to the foreign tax hikes. For example, consider a U.S. firm with half its CFC income in Ireland facing a 3 percent tax rate and half facing a 27 percent tax rate in Germany. With GILTI calculated on a global basis, its average foreign tax rate is 15 percent (higher than 13.125 percent), so the FTC limitation binds; if Ireland raises its effective tax rate to 15 percent, this does not change the FTC. On the other hand, if GILTI is calculated separately by country, the credit limitation binds in Germany but not in Ireland; if Ireland raises its tax rate, the FTC also increases.
Under current law, if other countries raise their effective corporate tax rates, this would reduce U.S. corporate tax liabilities by $20 billion over a decade. Because the Biden administration’s proposals—both the actual and partial versions—capture larger shares of foreign income and tax them at higher rates, these are more heavily exposed to foreign countries raising their tax rates, losing $158 billion to this under the full Biden plan and $77 billion under the partial version. The GILTI fix and Pillar 2 compliance also significantly increase U.S. exposure to foreign tax hikes. If all other countries raise their effective tax rates to at least 15 percent, under current law the only revenue raised by GILTI comes from temporary timing effects and indirect expense allocation. Allowing FTC carryforwards essentially eliminates the tax on timing effects. If other countries raise their effective foreign tax rates to at least 15 percent in response to Pillar 2, the U.S. will lose revenue by implementing the GILTI fix or by making GILTI comply with Pillar 2.
Results for Other Options
Finally, we present a set of other options for modifying the international tax rules facing U.S. multinationals in Table 9. We group these into three categories: changes to the level of GILTI calculations and the FTC; changes to the substance exemption for GILTI; and changes to indirect expense allocation rules.
| Change in Federal Corporate Income Tax Liabilities of U.S. MNEs from Various Options, $billions, Billions of Dollars | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Policy | 2022 | 2023 | 2024 | 2025 | 2026 | 2027 | 2028 | 2029 | 2030 | 2031 | Total |
|
GILTI calculation and FTC options |
|||||||||||
| Allow GILTI FTC carryforwards | 0.0 | -2.0 | -3.1 | -3.8 | -6.1 | -6.0 | -6.2 | -6.3 | -6.6 | -6.8 | -46.9 |
| Country-by-country | 7.6 | 8.1 | 8.6 | 9.2 | 12.1 | 12.5 | 12.8 | 13.2 | 13.5 | 13.9 | 111.4 |
| Country-by-country, with carryforwards | 7.6 | 6.1 | 5.6 | 5.3 | 5.9 | 6.3 | 6.5 | 6.7 | 6.8 | 6.9 | 63.6 |
| Mandatory high-tax exemption at 18.9% | 3.6 | 3.8 | 4.1 | 4.3 | 5.7 | 5.8 | 6.0 | 6.2 | 6.3 | 6.5 | 52.2 |
| Mandatory high-tax exemption at 18.9%, with carryforwards | 3.6 | 2.2 | 1.7 | 1.3 | 1.2 | 1.7 | 1.9 | 1.9 | 2.0 | 2.0 | 19.4 |
|
GILTI exemption options |
|||||||||||
| Remove QBAI exemption | 3.7 | 4.0 | 4.3 | 4.5 | 6.6 | 6.8 | 7.0 | 7.2 | 7.4 | 7.6 | 59.0 |
| 7.5% QBAI & payroll exemptions | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 1.0 |
| 5% QBAI & payroll exemptions | 1.6 | 1.7 | 1.8 | 1.9 | 2.8 | 2.9 | 3.0 | 3.1 | 3.1 | 3.2 | 25.2 |
|
Expense allocation options |
|||||||||||
| 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 |
| Switch to water’s edge R&D expense allocation | 1.8 | 1.9 | 2.1 | 2.2 | 2.3 | 2.4 | 2.5 | 2.5 | 2.6 | 2.7 | 23.1 |
| 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 |
| Apply section 265 denial of deductions to excluded foreign income | 8.4 | 9.0 | 9.7 | 10.3 | 9.0 | 9.3 | 9.6 | 9.8 | 10.1 | 10.4 | 95.7 |
| Note: 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: Tax Foundation’s Multinational Tax Model. | |||||||||||
Under current tax rates, allowing GILTI foreign tax carryforwards reduces federal corporate income tax liabilities by $47 over a decade. Implementing country-by-country GILTI calculations raises them by $111 billion, and allowing FTC carryforwards reduces this to $64 billion. Implementing the mandatory high-tax exemption at an 18.9 percent cutoff from the proposal by Senators Wyden, Warner, and Brown accomplishes almost half of switching to a country-by-country calculation, raising liabilities by $52 billion, although this falls to $19 billion when allowing carryforwards.
For the substance exemption, repealing it entirely would raise $59 billion under the current system. Switching to an exemption for 7.5 percent of tangible assets and payroll (as under the first five years of Pillar 2) raises $1 billion on net, and switching to 5 percent raises $25 billion.
Indirect expense allocation raises significant revenue. Under current law, interest expense is allocated by the water’s edge method discussed in the Overview section, but firms can elect to use the worldwide method for R&D expense allocation. Restoring worldwide interest expense allocation reduces firms’ federal corporate tax liabilities by $23 billion over a decade, and restricting firms to using the water’s edge approach for R&D expense would raise $23 billion. Eliminating expense allocation rules entirely for interest expense would reduce federal tax liabilities by $67 billion, and eliminating expense allocation rules for R&D expense would reduce tax liabilities by $78 billion. On the other hand, applying the administration’s proposed denial of deductions for excluded CFC dividends and GILTI not included in taxable income would raise corporate tax liabilities by $96 billion over a decade.
graduation
In this paper, we considered the effects on federal corporate income tax liabilities of U.S. multinationals under four main proposals. The Biden administration’s full proposal would raise these liabilities by $1.37 trillion over a decade. A partial version of this to reflect potential congressional compromises would raise $580 billion. A revenue-neutral GILTI fix would trade restoring worldwide interest expense allocation and GILTI FTC carryforwards for a higher GILTI minimum tax rate, and bringing GILTI into compliance with the OECD’s Pillar 2 minimum tax would raise $137 billion.
All these proposals would raise the U.S. tax on foreign activities of U.S. multinationals, although the Biden administration’s proposal would have the largest effect, raising the residual U.S. taxes on these activities from 2.9 percent of profits under current law in 2022 to 7.7 percent. This would put U.S. ownership of foreign activity at a substantial competitive disadvantage relative to foreign ownership, which would not face these residual U.S. taxes. However, the Biden administration’s proposals would increase profit shifting on net, contrary to their claims. While the proposal increases the residual U.S. tax on foreign activity by 4.9 percentage points on average, it increases the tax on profits booked in the U.S. by at least 7 percentage points. Under our main estimates, this increased profit shifting would reduce the revenue raised by the administration’s proposal by $79 billion over a decade.
We also identify a serious exposure to potential foreign tax hikes. If low-tax foreign countries and tax havens raise their tax rates to at least 15 percent—an intentional and incentivized response to the OECD’s Pillar 2 minimum tax—increased FTCs in the U.S. would reduce U.S. tax revenue by more than the reduced profit shifting. Under current law, this exposure is small. However, expanding the GILTI base, increasing GILTI tax rates, allowing FTC carryforwards, and reducing cross-crediting all increase U.S. exposure to these foreign tax hikes. Although convincing other countries to raise their tax rates would reduce the competitive disadvantage from residual U.S. taxes on foreign activity, the loss of U.S. tax revenue merits reconsideration of whether such a proposal is in the interest of the U.S.
(1) U.S. Department of the Treasury, “The Made in America Tax Plan,” April 2021, https://home.treasury.gov/system/files/136/MadeInAmericaTaxPlan_Report.pdf.
(2) OECD/G20 Base Erosion and Profit Shifting Project, “Statement on a Two-Pillar Solution to Address the Tax Challenges Arising from the Digitalisation of the Economy,” July 2021, https://www.oecd.org/tax/beps/statement-on-a-two-pillar-solution-to-address-the-tax-challenges-arising-from-the-digitalisation-of-the-economy-july-2021.pdf.
(3) Cody Kallen, “Effects of Proposed International Tax Changes on U.S. Multinationals,” Tax Foundation, Apr. 28, 2021, https://www.taxfoundation.org/biden-international-tax-proposals/.
(4) Erik Wasson and Steve T. Dennis, “Manchin Balks at Biden’s Corporate Tax Increase, Favors 25% Rate,” Bloomberg, Apr. 5, 2021, https://www.bloomberg.com/news/articles/2021-04-05/manchin-balks-at-biden-s-corporate-tax-increase-favors-25-rate.
(5) Harry Grubert, “Intangible Income, Intercompany Transactions, Income Shifting, and the Choice of Location,” National Tax Journal 56:1 (2003), https://econpapers.repec.org/article/ntjjournl/v_3a56_3ay_3a2003_3ai_3a1_3ap_3a221-42.htm.
(6) Mihir A. Desai, C. Fritz Foley, and James R. Hines Jr., “A Multinational Perspective on Capital Structure Choice and Internal Capital Markets,” The Journal of Finance 59:6 (December 2004), https://onlinelibrary.wiley.com/doi/10.1111/j.1540-6261.2004.00706.x.
(7) Michael P. Devereux and Rachel Griffith, “Evaluating Tax Policy for Location Decisions,” International Tax and Public Finance 10 (March 2003), https://link.springer.com/article/10.1023/A:1023364421914.
(8) John R. Graham, Michelle Hanlon, and Terry Shevlin, “Barriers to Mobility: The Lockout Effect of U.S. Taxation of Worldwide Corporate Profits,” National Tax Journal 63:4.2 (December 2010), http://dx.doi.org/10.17310/ntj.2010.4S.12.
(9) Kari Jahnsen and Kyle Pomerleau, “Corporate Income Tax Rates around the World, 2017,” Tax Foundation, Sept. 7, 2017, https://www.taxfoundation.org/corporate-income-tax-rates-around-the-world-2017/.
(10) Bloomberg, “Tracking Tax Runaways,” Mar. 1, 2017, https://www.bloomberg.com/graphics/tax-inversion-tracker/.
(11) Elke Asen, “Corporate Income Tax Rates around the World, 2020,” Tax Foundation, Dec. 9, 2020, https://www.taxfoundation.org/publications/corporate-tax-rates-around-the-world/.
(12) Cody Kallen, “Effects of Proposed International Tax Changes on U.S. Multinationals.”
(13) Alexander Arnon, Zheli He, and Xiaoyue Sun, “Profit Shifting and the Global Minimum Tax,” Penn Wharton Budget Model, July 21, 2021, https://budgetmodel.wharton.upenn.edu/issues/2021/7/21/profit-shifting-and-the-global-minimum-tax.
(14) Daniel Bunn, “U.S. Cross-border Tax Reform and the Cautionary Tale of GILTI,” Tax Foundation, Feb.17, 2021, https://www.taxfoundation.org/gilti-us-cross-border-tax-reform/.
(15) U.S. Department of the Treasury, “The Made in America Tax Plan.”.
(16) Cody Kallen, “Effects of Proposed International Tax Changes on U.S. Multinationals.”
(17) U.S. Department of the Treasury, “The Made in America Tax Plan.”
(18) American Rescue Plan Act of 2021, P.L. 117-2, 117th Congress, 2021.
(19) U.S. Department of the Treasury, “The Made in America Tax Plan”; U.S. Department of the Treasury, “General Explanations of the Administration’s Fiscal Year 2022 Revenue Proposals,” May 2021, https://home.treasury.gov/system/files/131/General-Explanations-FY2022.pdf.
(20) Removing Incentives for Outsourcing Act, S. 20, 117th Congress, 2021.
(21) Cody Kallen, “Effects of Proposed International Tax Changes on U.S. Multinationals.”
(22) Sens. Ron Wyden, Sherrod Brown, and Mark Warner, “Overhauling International Taxation: A framework to invest in the American people by ensuring multinational corporations pay their fair share,” Senate Finance Committee, April 2021, https://www.finance.senate.gov/download/overhauling-international-taxation.
(23) OECD/G20 Base Erosion and Profit Shifting Project, “Statement on a Two-Pillar Solution to Address the Tax Challenges Arising from the Digitalisation of the Economy.”
(24) Internal Revenue Service Statistics of Income Tax Stats, “Controlled Foreign Corporations,” https://www.irs.gov/statistics/soi-tax-stats-controlled-foreign-corporations.
(25) Bureau of Economic Analysis, “U.S. Direct Investment Abroad (USDIA): Activities of U.S. Multinational Enterprises,” https://www.bea.gov/international/di1usdop.
(26) Internal Revenue Service Statistics of Income Tax Stats, “International Tax Studies Based Upon Provisions Introduced by the Tax Cuts and Jobs Act (TCJA) of 2017,” https://www.irs.gov/statistics/soi-tax-stats-international-tcja-studies.
(27) Internal Revenue Service Statistics of Income Tax Stats, “Corporate Foreign Tax Credit Statistics,” https://www.irs.gov/statistics/soi-tax-stats-corporate-foreign-tax-credit-statistics.
(28) Bureau of Economic Analysis, “U.S. Direct Investment Abroad: Activities of U.S. Multinational Enterprises.”
(29) This ignores that a firm electing to use worldwide expense allocation must do so for 10 years; modeling this is not feasible with our data.
(30) Jost. H. Heckemeyer and Michael Overesch, “Multinationals’ profit response to tax differentials: Effect size and shifting channels,” Canadian Journal of Economics 50:4 (November 2017), https://onlinelibrary.wiley.com/doi/full/10.1111/caje.12283.
(31) U.S. Department of the Treasury, “General Explanations of the Administration’s Fiscal Year 2022 Revenue Proposals.”
(32) Tim Dowd, Paul Landefeld, and Anne Moore, “Profit shifting of U.S. multinationals,” Journal of Public Economics 148 (April 2017), https://www.sciencedirect.com/science/article/pii/S004727271730018X.
(33) Tax havens follow the Congressional Research Service identification: the Bahamas, Bermuda, Switzerland, Cayman Islands, Hong Kong, Ireland, Luxembourg, the Netherlands, Singapore, and the British Virgin Islands.