Pillar 2 and the United States: What’s Next

January 1 marked the official effective date of the 15 percent global corporate minimum tax imposed by pillar 2 as part of the G-20/OECD/inclusive framework base erosion and profit-shifting 2.0 project. Pillar 2 went into effect in Australia, Canada, the EU, Japan, Norway, South Korea, and the United Kingdom, with more countries expected to adopt it soon, including low-tax countries like Barbados, Ireland, Luxembourg, the Netherlands, and Switzerland. Critics have argued that pillar 2 violates tax treaties, customary international law, or bilateral investment treaties.¹ But it seems unlikely that legal challenges against it will succeed. Most of the countries that have adopted pillar 2 can override tax treaties by domestic legislation (such as Australia, Canada, and the United Kingdom) directive (the EU), or referendum (Switzerland).² The existence of customary international tax law is disputed, and it would be a brave national court that relied on it to invalidate a reform endorsed by over 140 countries (after all, if enough countries adopt a rule, they can effect a change in customary international law).³ A bilateral investment treaty arbitration challenge by an investor is more plausible, but if a losing country refunds the tax it collected, this could trigger additional tax liability in another country the multinational operates in, so it’s unclear what the multinational would gain.⁴

It may also be possible to mount a challenge under EU law, although the Court of Justice of the European Union usually does not override directives that were adopted unanimously by all member states. But even if the directive is struck down, given that enough other countries have adopted, or will soon adopt, pillar 2, the practical effect would simply be to shift tax revenue from EU to non-EU countries.

It would therefore seem that pillar 2 is a fait accompli (unlike pillar 1, which is doomed to failure because the United States will not ratify the multilateral tax convention implementing it). What would be the effect on U.S. multinationals if, as seems likely, the United States does not implement pillar 2 in the near future?

There are three reasons to believe the effect would be much smaller than some of the critics of pillar 2 have suggested.

First, the effective date of the application of the UTPR (formerly known as the undertaxed payments rule) has been delayed to 2026 for countries with a statutory corporate tax rate above 20 percent. Because the U.S. statutory rate is 21 percent, this transitional safe harbor should shield U.S. multinationals from the UTPR through the end of 2025. But in 2026, under current law both the global intangible low-taxed income rate and the foreign-derived intangible income rate increase from 13.125 percent (with foreign tax credits) to 16.406 percent, and therefore both will be above the pillar 2 minimum corporate tax rate of 15 percent. Although GILTI would still not be a qualifying income inclusion rule because it is not applied on a country-by-country basis, this change makes it less likely that U.S. multinationals will be subject to a top-up tax under either a qualified domestic minimum top-up tax (QDMTT) or a UTPR because in both cases the higher GILTI tax would be taken into account in calculating the effective tax rate. In addition, the increase in the FDII tax rate makes it less likely that the UTPR would apply to nullify the benefit of the FDII.

Second, the corporate alternative minimum tax adopted by the United States in 2022 applies a 15 percent minimum tax rate to the book income of U.S. multinationals on a global basis. Although there are differences between how the corporate AMT and the pillar 2 tax are calculated, the corporate AMT comes close to being a combination of a QDMTT on U.S.-source income and an IIR on foreign-source income and will make it less likely that U.S. multinationals will be subject to a top-up tax under the UTPR.⁵

Third, recent OECD administrative guidance has made it clear that both refundable and transferrable tax credits such as those enacted by the United States in the Creating Helpful Incentives to Produce Semiconductors and Science Act (the CHIPS Act) and the Inflation Reduction Act are qualified refundable tax credits (QRTCs).⁶ QRTCs have a much smaller impact on the ETR calculation than non-QRTCs and therefore make it less likely that a U.S. multinational will be subject to the UTPR on its domestic operations after 2025. In addition, tax equity investment arrangements commonly used in the United States for renewable energy credits and for low-income housing credits may qualify as qualified flow-through tax benefits (QFTBs). The special rule for QFTBs is beneficial to U.S. tax credits because many of them can be used through tax equity investment arrangements, and when they do, the tax credits will neither reduce the ETR nor be included in income, so that the benefit is completely unaffected by the UTPR.

U.S. critics point out that the UTPR might nullify the effect of other U.S. tax expenditures, like the expensing of research and experimentation (assuming Congress reinstates it, as seems likely because there is bipartisan support for it).⁷ But this is not plausible, for three reasons.

First, given the OECD’s reluctance to challenge the United States directly, it seems likely that it could issue further administrative guidance that exempts a revived U.S. R&E expensing from the UTPR even after the safe harbor expires. According to EY, Lily Batchelder, Treasury assistant secretary for tax policy, said at a recent conference that the United States continues to work with the inclusive framework partners on additional administrative guidance, and that an important U.S. priority is the treatment of the U.S research and development tax credit.⁸ Batchelder added that the United States continues to raise the issue in multilateral negotiations and that she is hopeful there will be progress.⁹

Second, Congress can convert R&E expensing into a refundable credit. Such a credit would be a QRTC and therefore would affect the ETR much less than a non-QRTC. A refundable credit for R&E is more effective than expensing because it is available to genuinely innovative start-ups that have losses rather than being limited to hugely profitable corporations, where expensing can eliminate U.S. tax liability without producing anything genuinely new. Congress already approved refundable (“direct pay”) credits on a bipartisan basis in the CHIPS Act, and can do so again, especially because the cost will be mitigated by the fact that under the OECD rules, the refunds can occur over a four-year period.

Third, even if Congress adopts a nonrefundable R&E credit (like the current credit under section 41), the effect of the UTPR on these credits is small. In a recent column, Tax Notes contributing editor Martin A. Sullivan showed that although the substance-based income exclusion (SBIE) has only a small impact on the ETR, its impact is magnified when it is combined with non-QRTCs.¹⁰ He first explains the impact of the SBIE on the incentive to shift tangible investments:

The incentive clearly exists, but what is its quantitative effect? The short answer is, not much. . . . Here’s a simple example: If a multinational increases capital spending by $1,000 in the first year that pillar 2 is in effect in a country with a tax rate of 8 percent, excess profits would be reduced by $80 (8 percent of $1,000). Those profits are taxed at a top-up tax rate of 7 percent (the 15 percent pillar 2 rate minus the jurisdiction’s 8 percent rate). So the $1,000 increase in investment would reduce pillar 2 tax by $5.60 (7 percent of $80).

The incentive provided by SBIE in this case is equivalent to an investment tax credit of 0.56 percent ($5.60 divided by $1,000). When the transition for the SBIE is fully phased in after a decade, the amount of SBIE attributable to tangible assets will be reduced from 8 percent to 5 percent, so the equivalent [income tax credit] would be a mere 0.35 percent (five-eighths of 0.56 percent).¹¹

But, as Sullivan points out, the impact of the SBIE would be much more significant if it is combined with a non-QRTC like the R&E credit:

What isn’t so obvious, and what is relevant to our investigation of incentive effects, is that SBIE reduces the dampening effect of pillar 2 tax on tax credits — both non-QRTCs and QRTCs. And because that dampening effect is much larger for non-QTRCs, the impact of SBIE on the incentive effect of non-QRTCs can be significant.

On the right-hand side of Table [1], SBIE is assumed to be $40 (40 percent of financial statement income). Pillar 2 taxable excess profit is $60. In the non-QRTC column, the $10 of credit reduces covered taxes, which reduces jurisdictional ETR to zero. The 10-percentage-point reduction in the tax rate that applies to the $60 of excess profit means the presence of the $10 credit raises pillar 2 tax by $6. The non-QRTC’s net effect on tax is no longer small. In other words, non-QRTCs, such as the U.S. research credit, may still provide a substantial — albeit diminished — incentive effect under pillar 2. In this example, the incentive is 40 percent of its pre-pillar-2 level. . . .

Using calculations similar to those in Table [1], Table [2] summarizes the value of a $10 tax credit under a range of assumptions. The takeaway: If SBIE is large, it restores much of the value of non-QRTCs that pillar 2 otherwise would have eliminated. It is even possible, as has been pointed out in an excellent paper by Michael P. Devereux and John Vella, that a non-QRTC can be more valuable than a QRTC.¹²

Table 1. Pillar 2 Effects on QRTCs and Non-QRTCs and SBIE Effects

	Substance Carveout = $0			Substance Carveout = $40
	No Tax Credit	Non-QRTC = $10	QRTC = $10	No Tax Credit	Non-QRTC = $10	QRTC = $10
Financial statement income	$100	$100	$100	$100	$100	$100
QRTC	$0	$0	$10	$0	$0	$10
Globe income	$100	$100	$110	$100	$100	$110
SBIE	$0	$0	$0	$40	$40	$40
Excess profit	$100	$100	$110	$60	$60	$70
Minimum tax rate is 15%	15%	15%	15%	15%	15%	15%
Corporate tax rate	10%	10%	10%	10%	10%	10%
Taxable income (assumed = financial statement income)	$100	$100	$100	$100	$100	$100
Pre-credit corporate tax	$10	$10	$10	$10	$10	$10
Non-QRTC	$0	$10	$0	$0	$10	$0
Covered tax (numerator of globe ETR)	$10		$10	$10		$10
Globe income (denominator)	$100	$100	$110	$100	$100	$110
Globe ETR	10%	0%	9.09%	10%	0%	9.1%
Top-up tax rate	5%	15%	5.91%	5%	15%	5.9%
Top-up tax	$5	$15	$6.50	$3	$9	$4.14
Pillar 2 tax increase		$10	$1.50		$6	$1.14
Pillar 2 tax increase as percentage of credit (offsetting the benefit of the credit)		100%	15%		60%	11.4%
Net benefit ($ amount) of credit		$0	$8.50		$4.00	$8.86
Source: Reprinted from Martin A. Sullivan, “The Not-So-Obvious Effects of Pillar 2 on Tangible Capital Investment,” Tax Notes Int’l, Jan. 1, 2024, p. 23. Sullivan created the table using his own calculations, which draw upon examples from Michael P. Devereux and John Vella, “The Impact of the Global Minimum Tax on Tax Competition,” 15 World Tax J. 323 (2023).

Table 2. Value of a $10 Tax Credit Under Pillar 2

SBIE	Non-QRTC	QRTC	QRTC Advantage
$0	$0.00	$8.50	$8.50
$25	$2.50	$8.73	$6.23
$50	$5.00	$8.95	$3.95
$75	$7.50	$9.18	$1.68
$100	$10.00	$9.41	-$0.59
Source: Reprinted from Martin A. Sullivan, “The Not-So-Obvious Effects of Pillar 2 on Tangible Capital Investment,” Tax Notes Int’l, Jan. 1, 2024, p. 23. Sullivan created the table using his own calculations.

Thus, assuming Congress enacts a nonrefundable R&E credit instead of expensing, any company doing real activity of a substantial kind in the United States (as opposed to simply booking income from intangibles there), is more likely to benefit from these SBIE effects in a way that helps the R&E credit be effective, and that is the sort of real activity the R&E credit is supposed to encourage. As Sullivan explains:

The substance-based income exclusion (SBIE) provides little relief from pillar 2 tax on profits that U.S. multinationals have packed into tax havens, but it’s a different story for their low-taxed profits in the United States. Our best estimate of the average foreign SBIE as a percentage of foreign profit is 17 percent. For the United States that tax-reducing number is 39 percent, more than double the average foreign figure.¹³

So, because of all the profit shifting out of the United States and because of the relatively large amounts of substance (such as high salaries) in the United States, U.S. multinationals already have on average some significant protection from pillar 2 effects on non-QRTCs because of relatively high SBIEs.

One of the repeated critiques of pillar 2 is that (1) because former tax havens like Switzerland can adopt a QDMTT, any additional revenue will accrue to them rather than to large countries that adopt the IIR or the UTPR, and (2) these tax havens can still engage in tax competition by refunding the added revenue from the QDMTT to multinationals via QRTCs.¹⁴ But these critiques miss the point, which is not to shift revenues from tax havens to large countries or to eliminate competition, but rather to level the playing field.

The corporate tax has two goals: to ensure that rich shareholders like Elon Musk, Jeff Bezos, and Mark Zuckerberg pay some current tax on the unrealized appreciation in their shares, and to enable countries to regulate multinationals by providing tax incentives and disincentives. The first goal is achieved by collecting a corporate tax on large multinationals, and from a distributive perspective it does not matter which country collects the tax. Assuming, as the recent economics literature shows, that the corporate tax is borne by wealthy shareholders and top management, then Musk et al. will bear the burden of the tax regardless of whether it is collected by the United States or by Switzerland.¹⁵

The second goal is achieved by leveling the playing field. A U.S. multinational can pay between 0 and 10.5 percent on foreign profits, depending on its level of tangible investment offshore. U.S. profits, however, are subject to tax between 13.125 percent (the FDII rate) and 21 percent. This creates an incentive to move jobs and tangible investment and to shift profits offshore, and it nullifies the effect of U.S. tax expenditures, because if the ETR on foreign profits is zero then no nonrefundable tax expenditure targeting domestic activities like R&E can be effective.

If, however, both U.S. and foreign profits are subject to the same ETR of 15 percent and if both can be reduced below that rate by the SBIE, QRTCs, or non-QRTCs combined with SBIE, then U.S. tax expenditures are effective. For example, if they reduce the U.S. ETR on domestic income from 21 percent to 15 percent, they achieve the intended result of creating an incentive for the desired activity in the United States without triggering UTPR and without creating an incentive to shift profits or activities offshore. The SBIE prevents a reduction of the ETR in both the United States and in foreign jurisdictions, so it creates no shifting effect out of the United States either. And even if a foreign jurisdiction adopts a QDMTT and then refunds the revenue to a U.S. multinational via QRTCs, the multinational can achieve the same result in the United States by using non-QRTCs, because a high SBIE in the United States will largely protect the multinational from the UTPR.

One can certainly criticize the OECD for adopting the SBIE rule and for allowing QRTCs, because both can reduce the ETR below 15 percent and thereby undermine the first goal of the corporate tax (to tax the rich).¹⁶ This is why I have argued that the OECD should in the future eliminate both the SBIE and QRTCs.¹⁷ But the regulatory goal of the corporate tax can be achieved even if the SBIE and QRTCs stay, and neither QRTCs nor the UTPR are likely to harm U.S. multinationals or the United States as an investment destination even if Congress neither enacts pillar 2 nor adopts QRTCs.