Tax Notes logo

The Baby and the Bathwater: Reflections on the TCJA's International Provisions

Posted on Feb. 1, 2021
Reuven S. Avi-Yonah
Reuven S. Avi-Yonah

Reuven S. Avi-Yonah is the Irwin I. Cohn Professor of Law at the University of Michigan. He thanks Kim Clausing, Steve Shay, and Bret Wells for their helpful comments.

In this article, the author considers how the international provisions of the Tax Cuts and Jobs Act could be improved.

Copyright 2021 Reuven S. Avi-Yonah. All rights reserved.

As the United States ushers in a new administration, it might be a good time to reflect on the Tax Cuts and Jobs Act. What should the Biden administration keep (the baby), and what should it discard (the bathwater)?

I believe the TCJA’s international provisions are its best1 and should generally be kept, with important modifications discussed below. The exception is the foreign-derived intangible income regime, a misguided export subsidy that should be discarded.


The global intangible low-taxed income regime is the best part of the TCJA. Finally, there is a real tax imposed on the accumulated offshore earnings of U.S.-based multinationals. It is imposed on a current basis, with only current tax credits, and those are limited to 80 percent. GILTI is a model for the OECD to follow as it develops pillar 2 as part of its continued efforts to find a way to tax the digital economy.

There are, however, four major problem areas.2

First, as many commentators have pointed out, the exemption for 10 percent of qualified business asset investments creates a perverse incentive to move tangible assets and jobs out of the United States, especially in conjunction with the participation exemption that means that this 10 percent return on assets will never be taxed anywhere. Both that exemption and the participation exemption should be abolished.3

Second, as has also been pointed out, the ability to cross-credit within the GILTI basket creates an incentive to invest in low-tax foreign jurisdictions and offset GILTI with higher foreign taxes elsewhere. That is likewise inappropriate when most U.S.-based multinationals are in an excess credit position because many foreign tax rates are higher than the U.S. corporate rate.4

Third, because the 50 percent deduction is allowed before allocable U.S. shareholder deductions, GILTI becomes a super subsidy if the allocable deductions are used against income taxed at 21 percent. That can and should be fixed by allocating U.S. shareholder deductions first and allowing the 50 percent deduction against net income only.

Finally, the low GILTI rate creates yet another incentive to shift profits offshore.5 As President Biden has proposed, the rate should be raised to 21 percent, which is still below the OECD average and creates no competitiveness concerns.


The base erosion and antiabuse tax is a good tax with terrible execution. Unlike GILTI, which raises more than $10 billion a year, the BEAT raises almost no revenue, and certainly far less than the Joint Committee on Taxation projected in 2017 (JCX-67-17).6

Again, there are four major problem areas.7

First, the exemption amount of $500 million is much too high. As one observer has suggested, it should be $25 million, which is the exemption from the interest limits under section 163(j).8

Second, while the absence of a foreign tax credit probably makes an exemption for cost of goods sold necessary to prevent double taxation, the U.S. Treasury should do much better in policing COGS to prevent the transformation of interest and royalties (subject to BEAT) into COGS (exempt).

Third, Treasury should revoke many of the Trump administration’s lax regulations, such as the broad definition of qualified derivatives (exempt) and the exemption for effectively connected income (which has no basis at all in the IRC).

Finally, if the domestic corporate rate is raised to 28 percent as Biden has recommended, the BEAT rate should be at least 14 percent.


Unlike GILTI and BEAT, FDII should simply be abolished. It confers a major benefit on U.S. multinationals that export goods or services with no evidence that it increases those exports. There is also no evidence of foreign-based multinationals moving export operations into the United States in response to FDII, or creating any additional jobs there. It is also inconsistent with the position the United States took in the OECD’s base erosion and profit-shifting initiative on limiting harmful tax competition via patent boxes (action 5).

Moreover, as I have stated elsewhere, FDII is a prohibited export subsidy under WTO law and likely to be challenged by our trading partners.9 There is no reasonable defense for it. One commentator has tried to defend FDII by arguing that it is not contingent on export performance because one can get it by only exporting services, which are not subject to the WTO Subsidies and Countervailing Measures Code.10 However, that argument was rejected by the WTO panels that struck down the foreign sales corporations and extraterritorial income regimes, which also applied to non-exports.

Another observer has argued that FDII is not a subsidy because it should be compared not with the 21 percent domestic rate but with the 10.5 percent GILTI rate.11 That argument also fails because most U.S. companies have already migrated their intangibles and therefore are unlikely to be able to shift more income offshore, while others that benefit from FDII cannot migrate the intangibles for regulatory reasons or because of the significant tax hurdles to intangibles migration. Thus, companies that benefit from FDII cannot transform it to GILTI without high transaction costs, so GILTI is not the proper baseline for measuring FDII as an export subsidy.


Of the TCJA’s three main international innovations, two (GILTI and BEAT) are important advances that have already inspired the OECD to try to follow suit with pillar 2 of BEPS 2.0. The third (FDII) is a violation of the WTO Subsidies and Countervailing Measures Code and BEPS action 5 and should be discarded.


1 The worst part is section 199A, which should be abolished. For discussion, see Daniel Shaviro, “Evaluating the New US Pass-Through Rules,” 2018(1) Brit. Tax Rev. 49 (2018). On the TCJA as a whole, see Reuven S. Avi-Yonah, “How Terrible Is the New Tax Law? Reflections on TRA17,” University of Michigan Public Law Research Paper No. 586 (Jan. 2018).

2 Other than the unjustified application of the subpart F high-tax exception to GILTI, which has no basis in the statute and leads to perverse results when combined with the expense allocation rules that create high taxes (from a U.S. perspective) in low-tax jurisdictions. It should be fixed by regulation.

3 Kimberly A. Clausing, “Fixing the Five Flaws of the Tax Cuts and Jobs Act,” 11(2) Colum. J. Tax L. 31 (2020).

4 Id.

5 Clausing, “Profit Shifting Before and After the Tax Cuts and Jobs Act,” 73(4) Nat’l Tax J. 1233 (2020).

6 See Thomas Horst, “Preliminary Estimates of the Likely Actual Revenue Effects of the TCJA’s Provisions,” Tax Notes Int’l, Sept. 16, 2019, p. 1153.

7 Bret Wells, “Get With the BEAT,” Tax Notes, Feb. 19, 2018, p. 1023.

8 Id.

9 Avi-Yonah and Martin Vallespinos, “The Elephant Always Forgets: US Tax Reform and the WTO,” University of Michigan Law and Economics Research Paper No. 18-006 (2018).

10 Grant D. Aldonas, “The WTO Consistency of the Deduction for FDII,” Tax Notes Int’l, Feb. 25, 2019, p. 815.

11 Chris William Sanchirico, “The New US Tax Preference for ‘Foreign-Derived Intangible Income,’” 71(4) Tax L. Rev. 625 (2018).


Copy RID