The Revenue Consequences of Pillar 2: Five Key Considerations

Recently, an analysis from the Joint Committee on Taxation¹ was released by two key Republican taxwriting committee lawmakers, Senate Finance Committee ranking member Mike Crapo of Idaho, and House Ways and Means Committee Chair Jason Smith of Missouri. These members have concluded that the United States stands to lose revenue under the OECD tax deal from pillar 2,² but this conclusion ignores important features of the JCT’s analysis that work in the other direction (detailed below). Importantly, pillar 2 will reduce the tax elasticity of the corporate tax base, allowing for more robust corporate tax revenue going forward.

The JCT begins its analysis by generating a modified baseline that acknowledges the adoption of pillar 2 by compliant jurisdictions, including Canada, the European Union, Japan, Liechtenstein, South Korea, Switzerland, and the United Kingdom. The modified baseline assumes that the adoption of the pillar 2 agreement by these countries will substantially reduce profit shifting, though the JCT is agnostic about the counterfactual regarding where that profit will ultimately be reported. It notes that U.S. revenue effects over 2023-2033 could be either positive (up $224 billion) or negative (down $174 billion), depending on the ultimate location of the shifted profits. The JCT says that it assumes something in between.³

Building on this modified baseline, the JCT then estimates several scenarios, noting that relative to that baseline, the United States will lose $122 billion in revenue over the next 10 years if the rest of the world (beyond those jurisdictions that are compliant) adopts major provisions of pillar 2 and the United States does not — but the United States will lose only $56 billion if it adopts alongside the rest of the world. Assuming the United States chooses the bare minimum adoption of the global agreement, this would make the overall 10-year revenue effect close to a wash, if one takes a simple average of the upper and lower bound baselines; the overall effect could be positive with a greater weight on the upper bound. If the United States joins the compliant countries in adoption, but no more countries adopt beyond those, the United States would gain $236 billion over 2023-2033, beyond whatever revenue is generated in the new baseline.⁴

From this analysis, it is clear that the United States will net more tax revenue if it adopts the agreement rather than standing aside, regardless of what we assume about which other countries adopt. It is also important to note that the global agreement provides the policy space for the United States to go beyond bare minimum adoption and apply a stronger minimum tax and a higher corporate rate, as the Biden administration has proposed.⁵ Below, I provide some context for the uncertainty around the JCT calculations, the importance of the baseline, and the policy hypotheticals that are most relevant for guiding U.S. tax policy.

Also, I remind readers of the basic logic of the international tax agreement. What gets lost in the details of these mind-numbingly complex calculations that likely mystify most readers (and even many of the highly qualified experts who work on them) is that some of the basic realities are simply not that complex. The international tax agreement is good for any country that wants to raise revenue from the corporate tax. We may not be that country, but if we are, it is also good for us.

I. Agreements Exist for a Reason

Countries do international agreements for a reason; they are intended to overcome global collective action problems. The status quo resulted in too little corporate tax revenue and too much low- or no-taxed multinational income. The JCT calculated that U.S. multinationals paid an effective tax rate of about 8 percent on their income in the wake of the Tax Cuts and Jobs Act.⁶ At the same time, U.S. corporate tax revenue was lower than those in many peer countries (as a share of GDP), even though the U.S. corporate community generates very high profits.⁷

This problem was exceedingly difficult for countries to tackle on their own because each country had an incentive to defect from any cooperative equilibrium by lowering its own tax rate and thus luring mobile multinational income away from higher-tax-rate countries. Consequently, corporate tax rates have declined steadily for several decades; OECD corporate tax rates averaged about 40 percent in the late 1980s but are about half that level in recent years.

These forces are what motivated the G-20/OECD inclusive framework base erosion and profit-shifting rounds. The first round (2013-2015) resulted in some incremental improvements but didn’t really address the root problems or the architecture of the international tax system. The most recent round, culminating in the 2021 agreement on two-pillar tax reform, went further, assuring that multinational income would face a 15 percent minimum effective tax rate, regardless of where it was reported.

For those countries that want to tax corporate income, this agreement, as a practical matter, enhances tax sovereignty. Before the agreement it was very difficult to tax multinational income because there was the perception (fueled by corporate lobbyists) that U.S. companies would lack competitiveness relative to companies based in other countries if the foreign income of U.S. multinationals was taxed significantly more than the tax burdens assigned by other governments. I’ve argued elsewhere that these competitiveness concerns were more perception than reality, because U.S. tax burdens, even before the TCJA, were not out of line with those of U.S. trading partners, but these sorts of concerns are part of what drove the large corporate tax cuts in the TCJA.⁸

The one instance in which tax sovereignty is impaired is when the policy desire is to have rock-bottom effective tax rates on the most profitable companies in the world. For jurisdictions that served as tax sanctuaries for those companies (not the United States), the agreement represents a binding constraint, preventing them from siphoning off tax base and revenue from other countries. But in the eyes of most voters, as well as the governments that want to tax corporate income, that constraint is a feature, not a bug. From this more common perspective, reducing the pressures that drive a race to the bottom in corporate taxation is an enormously helpful thing to do. We owe gratitude to Treasury Secretary Janet Yellen and her peers abroad for negotiating what former Treasury Secretary Lawrence H. Summers referred to as “arguably the most significant international economic pact of the 21st century so far.”⁹

II. Other Countries Tax, Too

Despite the provincial perspectives of some observers, the United States is not the only country that has the right to tax U.S. multinationals that operate throughout the world. Before the recent international negotiations, the light taxation of U.S. multinationals generated widespread discontent abroad, and foreign governments often turned to self-help in the form of ad hoc tax regimes (including digital services taxes). Whatever the merits of the Office of the U.S. Trade Representative’s analysis of these measures, it ruled them as discriminatory, launching tariff threats, which in turn provoked threats of retaliation abroad, reducing the gains from trade for both sides and generating new sources of policy uncertainty and volatility for businesses and workers.¹⁰

Under that status quo, the U.S. current-law global intangible low-taxed income regime (a provision that taxes some foreign income of U.S. multinationals) raised revenue by providing some minimum tax on lightly taxed foreign income for U.S. multinationals. The GILTI design was flawed, as I’ve discussed elsewhere, but it did reduce profit shifting and benefit U.S. revenue, relative to a version of the TCJA without the GILTI in place.¹¹

However, Republican taxwriters are dreaming if they think there is a sustainable equilibrium in which the United States applies GILTI but other countries do not also protect their own corporate tax bases from international tax avoidance, implementing either ad hoc or coordinated tax measures in response. And foreign jurisdictions have every right to defend their own corporate tax bases.

When foreign governments also tax lightly taxed income, that will unsurprisingly, and mechanically, lower GILTI revenue. But at the same time, it will also stem the pressures of profit shifting that erode the corporate tax base. As the JCT analysis correctly notes, there is a great deal of uncertainty about what will happen in the new equilibrium when there is less pressure to shift profits, but regardless of the counterfactual location of these very lightly taxed profits, there is without question substantially less profit shifting.

The reduced pressures of tax competition will lower the elasticity of the corporate tax base, making it easier for governments (including the United States) to raise revenue from the corporate tax, even when companies are multinational. A long economics literature has concluded that corporate tax base elasticities are large and that greater tax rate differentials drive greater tax responses. The corporate tax base elasticity is also demonstrably nonlinear, with the largest tax responses at the lowest tax rates.¹² In this context, raising the bottom tax rate from 0 percent to 15 percent will undoubtedly reduce profit-shifting pressures and make it far easier for governments throughout the world to collect real revenue from the corporate tax, should they choose to do so.

III. Revenue Estimating Is Difficult

I was the lead economist for the Office of Tax Analysis (OTA) at Treasury in 2021-2022, and saw firsthand the effort that expert economists (with the help of the legal staff) at the OTA and JCT expend in developing their estimates. However, some estimates are easier than others. Ask the professional staff to estimate the effects of increasing the corporate tax rate from 21 percent to 22 percent, and they can give you a reasonably precise answer relatively quickly. Ask instead what would happen if some unspecified fraction of the world implements several entirely novel international tax provisions (the IIR (income inclusion rule); the UTPR (formerly known as the undertaxed payments rule); and qualified domestic minimum top-up taxes) and there will be a lot more uncertainty and modeling difficulty.

The JCT is a group of highly esteemed professional experts, and they do not shy away from describing the difficulties of this exercise, including listing many of the assumptions that were made, as well as some instances of assumptions that they chose not to make (emphasizing ranges of outcomes instead). Below are just a few of the many sources of uncertainty, which have also been recognized by other observers:

• As the JCT notes, there is enormous uncertainty about how much profit shifting will be reduced, and where those profits will ultimately be booked instead. The modified baseline assumes that adoption of the pillar 2 agreement by certain countries will substantially reduce profit shifting, and the JCT provides a range of scenarios whereby U.S. revenue effects could be either positive (up $224 billion over 2023-2033) or negative (down $174 billion), depending on the ultimate location of the shifted profits. In my prior work, I’ve assumed shifted profits ultimately belong (absent profit shifting) where the economic footprint (assets, employment, sales) of the multinationals are; the economic footprint of U.S. multinationals is typically about two-thirds in the United States, which would imply a heavier weight to the upper bound of the JCT analysis wherein the United States gains revenue.¹³ If one instead takes a simple average of these scenarios, the U.S. revenue pickup would be lower, and there may be nonlinearities in these calculations resulting from the role of foreign tax credits.

• There is uncertainty regarding how much low-taxed profit there is abroad because different sources have different magnitudes, and even the tax data are not completely transparent on this question.

• There are important interactions between the provisions that are difficult to model. As just one example, the base erosion and antiabuse tax affects the use of foreign tax credits in a way that could affect these revenue estimates, but it is unclear how the JCT modeled the BEAT.¹⁴ Likewise, the ways in which foreign-derived intangible income, the corporate alternative minimum tax, and other code provisions interact with these incentives is difficult to model.

• The behavior of foreign multinationals will also affect the U.S. corporate tax baseline because foreign multinationals operate in the United States, and pillar 2 means that these companies will have a reduced incentive to shift income out of the U.S. tax base. Because the JCT typically holds GNP fixed in its analyses, it is not able to account for the positive effects of reduced profit shifting by foreign-headquartered multinationals on the U.S. tax base.¹⁵

• There is uncertainty about the evolution of the policy abroad. Will those countries that are in the process fully adopt pillar 2? Will other countries that are not yet compliant adopt it down the road? Especially in the short to medium term, it is highly unlikely that the entire world will adopt pillar 2 (as is assumed in JCT scenarios one and two); in comparison, in the near term, JCT scenarios four and five (in which the currently compliant countries adopt) are more realistic.

IV. Baselines Matter

One way to frame the results of the JCT study is to note that its analysis suggests that pillar 2 adoption will reduce profit shifting. And, if the United States then does the bare minimum pillar 2 implementation, and every other country in the world also adopts it, the revenue consequences may be (close to) a wash because the revenue pickup caused by reduced profit shifting in the baseline is approximately offset by increased taxation of low-taxed income abroad. In that scenario, the United States has embraced the lowest common denominator approach regarding the taxation of foreign income. Yet in the past, the United States was always able to charge more corporate tax than the lowest-tax-rate country abroad because the U.S. economy comes with many advantages.

The existing U.S. minimum tax provisions constitute another important element of the baseline. Pillar 2 adoption wouldn’t mean a very big increase in the tax burden for U.S. companies because there is already a bevy of minimum tax provisions (GILTI, BEAT, and the corporate AMT) that are reaching the same tax base. Indeed, the most recent provision, the corporate AMT, plays an important role in the baseline, as some of the corporate AMT base overlaps with what the United States would have picked up under pillar 2 if Congress had not legislated the corporate AMT instead. If the baseline were set before the corporate AMT, the gains from pillar 2 adoption would have been larger, and U.S. multinationals would have also benefited from a tax regime that better aligned with what the rest of the world is doing. (Many tax experts would have preferred an outcome of pillar 2 compatible reform instead of the corporate AMT, which also suggests grounds for possible bipartisan compromise down the road.)¹⁶

In general, both OECD analyses and those by outside academics show strong revenue gains from pillar 2 for the countries that adopt. For example, the global minimum tax is estimated by the OECD to raise about $220 billion per year, benefiting many countries.¹⁷ But, because the United States already employs many minimum taxes that are near the pillar 2 rates, it should be unsurprising that its revenue pickup from adopting a pillar 2-compliant reform is modest. However, the agreement ultimately supports the ability of the United States to have a more robust corporate tax base. It would be easier to take the GILTI rate to 21 percent, as the Biden administration has proposed doing, if foreign minimum tax rates are 15 percent instead of 0 percent, because there would no longer be important competitiveness concerns. (Perhaps this is why Republican taxwriters are so alarmed by the agreement.) That would also enable the regular corporate rate to be higher in tandem. The revenue benefits of a 28 percent corporate rate, alongside a 21 percent country-by-country GILTI, would be enormous, and U.S. multinationals would remain competitive.¹⁸ Rethinking large FDII preferences would provide another large source of revenue.

V. It Was Never About Competitiveness

Republican lawmakers have long cautioned about raising taxes on the foreign income of multinationals, arguing that it would put U.S. companies at a disadvantage when facing competitor companies from lower-tax jurisdictions.¹⁹ Now the same lawmakers are disappointed that other countries are raising their tax rates, reducing the pressures that have long led to a race to the bottom in corporate taxation. By their own prior logic, these moves abroad should be applauded, as they make it more feasible to protect the U.S. corporate tax base without risking cherished metrics of competitiveness. Indeed, with the rest of the world’s minimum rate at 15 percent, U.S. companies are more competitive at 21 percent than they would be under prior law when the rest of the world’s minimum rate was 0 percent and our GILTI charged 10.5 percent.

Protecting the corporate tax rate is important for a multitude of reasons. It is not just important for revenue needs (though those are certainly important). The corporate tax is an also important part of a fair tax system, because more than 70 percent of U.S. equity income goes untaxed at the individual level, and the capital income that is taxed at the individual level often benefits from deferral or even escapes taxation entirely because of the step-up in basis at death.²⁰ Because capital income is far more concentrated than labor income, the adequate taxation of business income is an important part of a progressive tax system.²¹

Taxing multinationals is also an important part of a fair, efficient playing field. Recent decades have been characterized by the rising market power of large businesses, and multinationals have many advantages relative to smaller companies, including lower tax rates on their foreign income.²² The U.S. corporate tax base is very concentrated. IRS data show that less than one-half of 1 percent of U.S. corporations account for 87 percent of the U.S. corporate tax base, with about 350 of these companies accounting for 69 percent of the tax base.²³

Finally, narrowing the gap between the tax treatment of foreign and domestic income serves another, more important, definition of competitiveness — reducing the advantage that our tax code provides for foreign income over domestic income. This change would reduce the tilt in the tax playing field that favors job creation and investment abroad relative to jobs and investments at home.

VI. Bottom Line

As the JCT notes, there are large sources of uncertainty in estimates of what will happen under different hypothetical scenarios of policy adoption. However, simple common sense tells us that the agreement is good for any country that wants to raise revenue from the corporate tax. If the rest of the world agrees to tax multinational income at some base level, that helps the revenue potential of the corporate tax because it lessens the pressures of tax competition and profit shifting. The effect of the agreement on the ability of countries to collect revenue from the corporate tax is not the policy exercise that the JCT has been asked to consider in its recent report, but it is a highly relevant policy question. (The JCT’s preliminary analysis of the baseline supports these same conclusions.)

Of course, the U.S. revenue potential of the corporate tax is highest if the United States taxes multinational income throughout the world and other countries refrain from doing so. But that outcome is not realistic. The U.S. corporate community would not support those divergent tax burdens for competitiveness reasons, and foreign governments would not be content to cede all taxing rights to the United States.

The international tax agreement solves an important global collective action problem, facilitating the collection of adequate tax revenue from highly mobile multinational corporate income. This allows governments throughout the world the space to make important revenue-raising corporate tax policy changes, and it will result in a better alignment between where economic activity occurs and where taxes are paid. In the United States, it provides policy room to raise the GILTI rate, raise the corporate rate, and reconsider other tax preferences. These corporate tax policy changes have the potential to fund important fiscal priorities, lower tax rates elsewhere in the system, and reduce deficits. Simply put, the agreement helps the United States tax multinationals — if it has the courage to do so.