It may be possible to significantly improve the economics of the tax on global intangible low-taxed income by simply adjusting a few numerical values in the statute. It also may be possible to simplify our new minimum tax on foreign profits by eliminating the deduction that attempts to isolate excess foreign profit and instead allows expensing (or some variation of expensing) of foreign capital expenditures.
Those aren’t new ideas, but they deserve fresh consideration. Unlike the misguided tax deduction for foreign-derived intangible income and the unholy mess that is the base erosion and antiabuse tax, the tax on GILTI is here to stay. And there are rumblings that the notion of a formulaic minimum tax like GILTI may be under consideration outside the United States.
It has been said that GILTI is intended to provide an incentive for U.S.-resident multinationals to bring intangible assets back to the United States. For our purposes, such talk is best waved off as imprecise shorthand or the articulation of a mildly off-target objective. A more meaningful approach to guide analysis of the combined effect of the new tax on GILTI (section 951A), the new deduction from that income (section 250), and the reduced tax credits available for foreign taxes paid on that income (section 960(d)) is to view them as a brake on profit shifting generally and on the incentive resulting from that profit shifting to relocate tangible capital from the United States to non-U.S. locations.
Let’s focus first on three variables assigned precise values by Congress in the Tax Cuts and Jobs Act (P.L. 115-97). Yes, that’s right — only three. GILTI aficionados will be disappointed by the lack of detail here, but we are struggling to grasp the often-enigmatic economic effects of GILTI. If, for the sake of completeness, we examine all aspects of GILTI that rightly concern practitioners, even the brightest among us will be distracted from the big picture. We hope that any omissions do not leave a distorted impression of what we are trying to depict.
Why 10 Percent, 50 Percent, and 80 Percent?
QBAI percentage. The first variable to consider is what we’ll call the qualified business asset investment (QBAI) percentage. The GILTI minimum tax is imposed on excess profits of a U.S. multinational’s controlled foreign corporations. Excess profits in this context are those that exceed 10 percent of tangible depreciable capital (officially known as QBAI). This 10 percent is the QBAI percentage. Increasing the QBAI percentage would reduce U.S. minimum tax on foreign profits. Increasing the QBAI percentage would reduce the minimum tax’s brake on foreign investment. If it is sufficiently large — if the QBAI percentage is larger than the rate of return on foreign tangible capital and so excess profit (on the margin) is negative — the minimum could become an incentive to increase foreign investment.
GILTI percentage. After a multinational computes the excess of its CFC profit over QBAI, it is allowed a deduction equal to 50 percent (37.5 percent after 2025) of that excess profit. That 50 percent is the GILTI percentage. Raising the GILTI percentage would reduce U.S. tax (assuming excess profit is positive) and increase the incentive for foreign capital formation. On the other hand, if excess profit on the margin is negative, a larger GILTI percentage would increase U.S. tax on foreign profits and dampen the incentive for marginal low-profit investment outside the United States.
Increasing the GILTI percentage could have another significant economic effect that cannot be ascertained from the statute. Presumably, the dollar amount of profit shifted from the United States to a foreign location is a positive function of, among other things, (a) the excess of the U.S. statutory tax rate over (b) the sum of the effective foreign tax rate (taking into account U.S. credits on foreign tax) on shifted profit and the effective U.S. tax rate imposed by GILTI on shifted profit. The effective rate of U.S. tax imposed on the shifted excess profit is a negative function of the GILTI percentage. In this case, shifted profit is a positive function of the GILTI percentage (less tax on shifted profit increases shifted profit). Also, total tax paid per dollar of shifted profit is a positive function of the GILTI percentage (less tax on foreign capital income increases differential and benefit per dollar of shifting).
Foreign tax credit percentage. To provide U.S. multinationals with an incentive to minimize foreign taxes, only 80 percent of foreign taxes on GILTI income (before the section 250 deduction) are creditable. This 80 percent is the foreign tax credit percentage. Two effects of the FTC percentage concern us here. If this percentage was reduced, it would directly reduce the incentive to invest outside the United States by increasing the U.S. tax on foreign capital. Reducing the FTC percentage would also increase the total amount of shifting and the benefit per dollar of profit shifted out of the United States (by increasing the differential between effective rates of tax on investing in the United States and abroad).
It may be hard to keep all this stuff in your head, so here’s an easier way to think about it: You can eliminate artificial profit shifting and the incentives to move real investment out of the United States by moving toward a worldwide system. This can be done by moving the GILTI percentage toward zero (eliminating the section 250 deduction) and the FTC percentage toward 100 (allowing full creditability of foreign taxes). But this would make U.S. multinationals uncompetitive from a tax standpoint and encourage expatriation (as well as other manners of increasing foreign ownership of U.S. foreign affiliates). That could be offset with a generous incentive for tangible foreign capital formation. In the current law framework, this would be done by increasing the QBAI percentage.
Calculating Excess Profit and Expensing
Current law uses a method of calculating excess profit that differs from the groundbreaking idea suggested in 2013 by Harry Grubert and Rosanne Altshuler: treating excess return differently than normal return in the context of a minimum tax on foreign profits (Grubert and Altshuler, “Fixing the System: An Analysis of Alternative Proposals for the Reform of International Tax,” 66 Nat’l Tax J. 671 (2013)).
Expensing is equivalent in present value to exemption of the normal return on capital over the life of an asset. By allowing expensing on foreign tangible capital, those normal returns are exempt from U.S. tax under the Grubert-Altshuler minimum tax. In other words, excess profit in this case is the residual measured foreign profit after expensing rather than depreciating foreign tangible capital expenditures. As the authors perceptively point out, normal returns can justifiably be taxed at a lower effective rate than excess returns because normal returns are the component of total profits that are critical in the determination of investment behavior.
Normal returns are usually generated by commodity-like products and services in industries for which there are low barriers to entry. So in the Grubert-Altshuler scheme, foreign investment by U.S. multinational companies with normal returns would compete on equal terms with foreign investment by foreign-owned businesses. On the other hand, investment that generates excess return, which is less vulnerable to cross-border competition, is subject to full U.S. tax under the Grubert-Altshuler minimum tax plan.
It is instructive to compare GILTI with the Grubert-Altshuler minimum tax with expensing. First, it is critical to understand that the term “excess profit” has several meanings (which are too often conflated). In economics, “excess return” means anything above normal returns. These normal returns are not competed away because the investing company has some sort of market power (for example, it is a monopoly or part of an oligopoly or it has successfully differentiated its product from those of potential competitors).
In international tax, “excess profit” usually applies to any profit earned by a component of a multinational that is above what would be earned by that component if all transactions with related parties were conducted as if with third parties at arm’s length. So, for example, tangible capital might have a normal return of 5 percent and an additional economic excess return of 3 percent. This 8 percent return can be generated with no profit shifting. If 8 percent is the total arm’s-length return, any amount exceeding that would be shifted profit. In our example, the Grubert-Altshuler plan would shelter only the 5 percent from tax.
Intangibles in Focus
If we assume that intangible capital generates the same normal and excess return as tangible capital, the 5 percent normal return on intangibles is also exempt because (most) intangible development costs are expensed. This simple but critical observation receives almost no attention in commentary on minimum taxation of foreign profits. (One implication of this is that if the QBAI return equals the actual return on foreign investment, intangible-heavy industries like banking and technology are not disadvantaged by GILTI.)
With this in mind, we can plainly state that the 3 percent economic excess return on both tangible and intangible capital is taxed under the Grubert-Altshuler minimum tax even though there is no profit shifting. If preventing profit shifting is the goal, this plan overreaches. This problem could be remedied by making expensing more generous or providing a tax credit for foreign tangible investment. (These extra deductions or additional investment credits should vary by the useful life of the assets to maintain neutrality across tangible asset classes.)
Under the GILTI regime, tax is imposed on 10 percent of tangible capital. This may or may not exceed the sum of normal and excess returns on that capital. As with Grubert-Altshuler, normal intangible profits under this plan are effectively exempt from U.S. tax because intangible development costs are expensed.
In addition to the method of identifying excess profit, a major difference between Grubert-Altshuler and GILTI is that excess profits under Grubert-Altshuler are taxed at the full U.S. rate while excess profits under current law are effectively taxed at half the U.S. rate. This means profit shifting remains an issue (although far less than under prior law). If the amount of profit shifted is significantly affected by differentials between domestic and foreign effective tax rates, there might be considerable room for improving a minimum tax by adjusting those effective rates.
Expensing current capital expenditures is far simpler and less susceptible to abuse than calculating 10 percent of the four end-of-quarter average adjusted bases of foreign tangible capital. And in effect, expensing automatically calculates the normal rate of return. Despite those advantages — and we civilians can only guess why — Congress shunned the expensing approach for two possible reasons: First, it wasn’t sufficiently generous. Second, it likely would have generated large GILTI losses in the early years after enactment (which is especially problematic given the prohibition of loss carryforwards in the calculation of GILTI).
As we have suggested, the first problem could be addressed by juicing up expensing with extra investment incentives for foreign tangible capital. If a benefit greater than expensing was added, this would tilt the playing field in favor of tangible capital, which would have more than economic excess returns exempt from tax. This unequal treatment could be remedied by devising a markup on intangible development cost. Ideally, this markup would exempt an amount of intangible income from tax equivalent to the exempt excess profit tangible capital. But in practice the best we could hope for would be an acceptable approximation of the extra benefit bestowed on tangible capital.
The problem of large upfront losses is real but could be remedied for the most part by allowing a fresh start. Under this option, all depreciation deductions for tangible capital purchased in prior years would be disallowed. Besides smoothing out depreciation, this disallowance of deductions on old capital has little direct effect on new investment. The top line in the figure shows depreciation deductions under pure expensing that begins in 2020. (The figure assumes 6 percent annual growth and straight-line depreciation over a five-year useful life.) The middle line shows expensing with a fresh start. The bottom line shows deductions with no expensing.
A Better Mousetrap?
Combining those observations suggests that the options worth exploring are those in which tangible capital receives a simple, upfront benefit that is more generous than expensing combined with a significantly reduced or eliminated GILTI percentage. Reducing the GILTI percentage would have the twin benefit of reducing the incentive to invest in foreign capital (when excess profits are negative, which they are more likely to be with a large QBAI percentage or other extra investment incentive for foreign tangible capital) and reducing profit shifting. Also, adjusting the FTC percentage to reduce profit shifting should be added to the mix.
A more precise evaluation of these options will greatly depend on how sensitive profit shifting is to changes in effective tax rates that result from changing the minimum tax parameters. This will be attempted in a sequel to this article — with equations.