Getting Serious About Cross-Border Earnings Stripping: Establishing an Analytical Framework

BY J. Clifton Fleming Jr., Robert J. Peroni, & Stephen E. Shay

Click here for PDF*

Clifton Fleming, Jr.,** Robert J. Peroni,*** & Stephen E. Shay****

A multinational enterprise (“MNE”) may use cross-border earnings stripping as a tax planning strategy. This strategy involves a higher-tax affiliate making deductible payments to a low- or zero-tax affiliate to reduce the MNE’s global effective tax rate and, in the process, erode the corporate tax bases of countries where its economic activity otherwise would be more highly taxed. Earnings stripping the U.S. corporate tax base is a major objective of U.S. corporations that engage in “inversion” transactions to become a subsidiary in a foreign-parented group. This Article explains how the earnings stripping problem extends beyond corporate inversions to U.S. subsidiaries of foreign-parent groups regardless of how the group was formed and is independent of the controversy regarding whether the United States should adopt a territorial system. The Article develops a theoretical framework for analyzing earnings stripping and identifying the scope of an appropriate response. The Article further grapples with developing workable rules to implement its theoretical policy prescriptions. The Article’s analysis supports limiting the deductibility of cross-border interest payments by a U.S. subsidiary with respect to intra-MNE debt to the subsidiary’s proportionate share of the MNE’s worldwide debt owed to unrelated parties. The Article’s analysis also supports treating the proper taxation of cross-border services fees as primarily a transfer pricing issue but justifies a source withholding tax as a device for dividing revenue between the source and residence countries. Finally, the Article advances understanding of the proper taxation of royalties paid by a U.S. subsidiary to a foreign member of an MNE but concludes that the ultimate resolution of that issue requires further work.

 

Introduction

A multinational enterprise (“MNE”) may use cross-border earnings stripping as a tax planning strategy. This strategy involves a higher-tax affiliate making deductible payments to a low- or zero-tax affiliate to reduce the MNE’s global effective tax rate and, in the process, erode the corporate tax bases of countries where its economic activity otherwise would be more highly taxed. Earnings stripping the U.S. corporate tax base is a major objective of U.S. corporations that engage in an “inversion” transaction to become a subsidiary in a foreign-parented group, thus allowing the deductible payments strategy to work without running afoul of U.S. anti-avoidance rules in Subpart F[1] and other provisions of the Internal Revenue Code.[2] This is because the deductible payments save tax by shifting U.S.-source income to low-tax countries, and corporate inversions are an effective way to create the U.S. subsidiary/foreign parent structure that makes the deductible payments possible. Hence, most of the current discussion of earnings stripping occurs in the context of how to tackle the inversion problem.

In this Article, we explain that although earnings stripping is appropriately an important part of the inversion conversation, it has additional critical effects. Those effects extend beyond corporate inversions to U.S. subsidiaries of foreign-parented groups regardless of how the group was formed. Those effects will continue to present a broad challenge to the U.S. system for taxing cross-border income even if inversions are curtailed. This Article also explains that earnings stripping is a threat to the U.S. tax base regardless of whether the United States (1) adopts a real worldwide income tax system that eschews deferral and substantial cross-crediting, (2) adopts an explicit territorial system, or (3) continues with its present bollixed system that appears to be a worldwide system but largely functions as an unduly complex, incoherent, and elective de facto territorial regime. Thus, the unresolved controversy regarding overall system design is not a reason to postpone developing remedies for the earnings stripping problem.

To identify appropriate remedies, however, it is first necessary to determine what is objectionable about earnings stripping, and that quest is inhibited by the fact that U.S.-source taxation, to which earnings stripping is a threat, is an under-theorized topic. Consequently, this Article develops a theoretical framework for analyzing earnings stripping and identifying the scope of an appropriate response to the earnings stripping problem. The Article then considers workable and policy-relevant ways to implement its theoretical prescriptions. In our judgment, the likelihood of the emergence of an effective multilateral approach to earnings stripping is remote. Thus, the United States must develop appropriate rules in light of its own tax system and interests, although seeking international cooperation in areas where it can be achieved will certainly further those interests.

The Article proceeds in Part I by providing a background that gives context to the earnings stripping issue both within and beyond the corporate inversion problem. Part II develops a theoretical framework for determining both the detriments of earnings stripping and the appropriate responses. Part III then confronts the problem of translating theoretical analysis into workable, real-world solutions. Part IV addresses the question of whether revenue realities allow us to ignore the earnings stripping problem and argues that they do not.

  1. Background
  2. Real Worldwide Taxation Revisited

In our prior work,[3] we have argued that the United States should adopt a real worldwide system for taxing the foreign income of a U.S. MNE.[4] Under a real worldwide system, the foreign-source and U.S.-source income of a U.S. parent corporation and all of its controlled foreign subsidiaries would be aggregated and subjected to a current, non-deferred U.S. income tax with a credit for foreign income tax paid thereon to the extent that the foreign tax did not exceed the applicable U.S. tax. In addition, foreign tax imposed by high-tax foreign countries in excess of U.S. tax could not be cross-credited against the U.S. residual tax on income earned in low-tax foreign countries. This approach would greatly ameliorate distortion of the choice between the United States and low-tax foreign countries as the location for business or investment activity,[5] would end the so-called lockout effect of current law,[6] would align U.S. international income taxation more closely with the ability-to-pay principle that underlies the choice of income as the primary U.S. federal tax base,[7] and would allow the United States to collect a residual tax[8] on income earned by U.S. MNEs in low-tax foreign countries at a time when the U.S. Treasury is in need of increased revenue. Stiff resistance to achieving these goals comes from opponents who seek to exclude foreign-source business income from the U.S. tax base, and a lively and important debate has resulted.[9]

  1. Corporate Inversions

Among the arguments made by opponents of the real worldwide taxation that we propose is that U.S. MNEs could readily defeat such a regime by reorganizing into foreign-controlled MNEs—a tactic referred to as corporate inversion.[10] The United States partially addressed this concern with the enactment of statutory rules discouraging inversions in 2004.[11] While these rules have had a substantial restrictive effect,[12] in recent years U.S. MNEs have sought to avoid the most burdensome aspect of those rules by merging with foreign corporations,[13] and there has been considerable discussion regarding the appropriate response.[14]

A major reason for U.S. MNEs to invert is to create large amounts of debt owed to foreign related parties and then to erode the U.S. income tax base by means of deductible interest payments on that debt, a maneuver included under the rubric of earnings stripping.[15] We have argued for broadening the definition of U.S. corporate tax residence, which would have a restraining effect on this tactic.[16] This argument, however, is not a complete response because the untoward effects of earnings stripping, which we describe below, are not dependent on the existence of an inverted corporation but also arise in the context of foreign-controlled MNEs that were not formed by inverting a U.S. MNE. Consequently, this Article will focus on developing an analytical framework for identifying responses to the principal earnings stripping strategies, regardless of whether a corporate inversion is involved.

  1. Illustrating the Earnings Stripping Challenge to Source Taxation

Foreign corporations that are not subsidiaries of U.S. corporations pay U.S. income tax only on their U.S.-source income (with a few limited exceptions).[17] Thus, with respect to these corporations, the principal revenue challenge is to ensure that their U.S.-source income is, in fact, taxed by the United States rather than being shifted out of the U.S. tax base by a variety of techniques,[18] including the strategy known as earnings stripping. Example 1 illustrates this point.

 

Example 1: USSub is a U.S. corporation and a wholly owned subsidiary of UKCo, a U.K. corporation that manufactures precision scientific instruments in the United Kingdom and whose shares are predominantly owned by nonresidents of the United States. UKCo sells the instruments to USSub, which markets them to U.S. customers. To a substantial extent, USSub’s operations are financed by loans from UKCo. These loans satisfy existing U.S. income tax law criteria for treatment as bona fide debt and they result in large market rate interest payments by USSub to UKCo. USSub also makes large market rate royalty payments to UKCo for the use of important marketing intangibles pursuant to contract terms that qualify as licenses rather than sales under U.S. income tax law. In addition, UKCo receives substantial fees from USSub for actually providing various accounting and management services from its U.K. offices. Although the fees are set at fair market value rates, they are substantially above UKCo’s actual cost of furnishing the services. USSub’s interest payments, royalty payments, and fees payments are all deductible expenses for U.S. income tax purposes[19] and have the effect of absorbing most of USSub’s income even if no transfer pricing abuse is involved, a tactic commonly referred to as “earnings stripping.” Consequently, there is no U.S. income tax on most of USSub’s profits from sales into the United States.[20] Moreover, there is no U.S. withholding tax on fees for UKCo’s management services performed from its U.K. headquarters[21] and the United States-United Kingdom Income Tax Treaty makes the U.S. withholding tax inapplicable to USSub’s interest and royalty payments to UKCo.[22]

With respect to payments by U.S. MNEs to controlled foreign corporations, the earnings stripping problem is mitigated by the application of the Subpart F foreign personal holding company rules, which, when they apply to the income from deductible U.S. payments, bring the amounts back into the U.S. tax base.[23] Avoiding the Subpart F rules in order to engage in U.S. earnings stripping is an important reason for corporate expatriations/inversions.[24] This clear intersection of residence and source taxation highlights a fundamental feature of international taxation. Taxation at source plays a critical role in protection of the residence tax base in that strong anti-earnings stripping rules reduce the incentive for a U.S. corporation to expatriate to become a foreign corporation.[25] In turn, comprehensive residence taxation of worldwide income protects against residents’ use of foreign corporations to erode the residence taxation base. Without application of the Subpart F foreign personal holding company rules, U.S. MNEs would engage in stripping the U.S. tax base just as they are stripping foreign tax bases.[26]

One important principle of tax system design is to not put pressure on margins the tax system cannot defend—what might be called weak margins.[27] Three weak margins in the current international tax rules are: corporate residence,[28] attributing income to a geographic source according to the legal category of the income,[29] and applying an arm’s-length transfer pricing principle in contexts in which there is not a reasonably close market benchmark.[30] As we will demonstrate, earnings stripping tax minimization strategies place pressure on tax systems across each of these margins.[31] Effective and coherent limitations on earnings stripping should reduce that pressure.

The earnings stripping problem shown in Example 1 would be even more exigent if the United States adopted a so-called territorial system. Such a system would typically employ characterization rules and arm’s-length transfer pricing principles to identify foreign-source business income of U.S. corporations and would then exclude such income from the U.S. tax base.[32] In that case, the business income tax base for U.S. corporations would be a narrow construct limited to U.S.-source business income.[33] Erosion of this narrow base through earnings stripping would have an even greater impact than if foreign-source business income of U.S. corporations were included in the base under true worldwide taxation or, at a minimum, under more robust Subpart F-type anti-avoidance rules. Thus, contrary to the arguments made by some proponents of territorial taxation,[34] adoption of a territorial system by the United States will not ameliorate the earnings stripping problem and will probably make it worse.[35] Stated differently, earnings stripping is a discrete issue that plagues both worldwide and territorial systems. Consequently, there is no good reason that earnings stripping reforms should be deferred until the United States determines whether its international income tax regime will be comprehensively transformed into a territorial system or a real worldwide system.[36]

  1. Aggressive Transfer Pricing Is Not Necessary

The facts of Example 1 stipulated that USSub’s deductible payments to UKCo were all at fair market value rates. Nevertheless, the effect of those payments was to move a substantial portion of USSub’s U.S.-source income out of the U.S. tax base and into the low-tax United Kingdom.[37] This fact illustrates the point that while aggressive transfer pricing magnifies the base erosion effect of earnings stripping (in some cases, greatly so), it is not a necessary element of the earnings stripping strategy. Earnings stripping can be effectively executed with arm’s-length payments.

Earnings stripping, in practice, however, is inextricably intertwined with transfer pricing rules and tax minimization strategies based on those rules. Under accepted interpretations of the arm’s-length principle, a generous range of prices may be found to be “arm’s length.”[38] Accordingly, in determining what is an arm’s-length payment for an item for which there is not a reliable comparable transaction, well-advised taxpayers naturally will adopt a price that is at whichever extreme end of the arm’s-length range that yields the most favorable after-tax result. So long as that price is within an arm’s-length range of prices, the current U.S. transfer pricing regulations immunize the pricing from adjustment.[39]

There is a further and more intractable transfer pricing problem. Transfer pricing rules, like source rules, operate on the basis of legal categories.[40] When payments are made for intangibles and services in a related party context in which the overall enterprise earns economic rents, it is possible to apply the transfer pricing rules either to transfer a portion of the rents through higher royalties or services payments or to leave the rents undisturbed in the income of the company that nominally earned the rents. There is no separate legal category for economic rents, except, to a limited extent, goodwill (the treatment of which from a legal perspective is underdeveloped). Which country is entitled to tax the rents is the central question of inter-nation income allocation and earnings stripping transfer pricing strategies are a flexible tool for taxpayers to use in making their own allocation of economic rents. As will be seen below, the problem of which related party should obtain the benefits of economic rents is a confounding issue in identifying a theoretically satisfying normative benchmark to evaluate earnings stripping.

  1. A Corporate Inversion Is Not Essential

In Example 1, there are several ways by which USSub might have become UKCo’s wholly owned subsidiary. USSub might have been a subsidiary created by UKCo to carry on the latter’s U.S. sales operation. Alternatively, UKCo and USSub might have been historically unrelated parties that came into a parent-subsidiary structure through a variety of methods. Generally, if UKCo is larger than USSub, UKCo would acquire USSub in an arm’s-length transaction. In the current market environment, if USSub is larger than UKCo, USSub might engage in a so-called inversion transaction in which a new UKCo parent corporation is organized with USSub and old UKCo as subsidiaries. USSub would engage in U.S. marketing activity for old UKCo.[41] Example 1 does not specify which of these paths the companies took to position USSub as a UKCo subsidiary since that is irrelevant for purposes of this Article. The critical point is that regardless of which transactional pattern was used to bring about Example 1, a foreign-owned U.S. subsidiary, such as USSub, was made available to serve as the facilitator of the earnings stripping strategy illustrated in that example. A corollary to this point is that restrictions on earnings stripping reduce both the incentive for U.S. corporations to invert and the relative advantage for a foreign corporation to acquire a U.S. corporation.

  1. What’s Bad About Earnings Stripping?

The preceding discussion leads to the question of exactly why earnings stripping is an apt object of tax policy concern. A major objection to earnings stripping is that it depletes the U.S. tax base.[42] This is shown in Example 1 where the U.S.-source income of the UKCo-USSub MNE was the only portion of that MNE’s worldwide income that was potentially taxable by the United States. Yet earnings stripping removed most of that income from the reach of the U.S. income tax even though the MNE generated the income by exploiting the U.S. market.[43]

A second objection to earnings stripping arises from the preceding point. U.S. resident corporations that compete with UKCo in the U.S. market, but that are not subsidiaries of foreign corporations, cannot generate income that escapes U.S. taxation through earnings stripping. This is so because they have no foreign parent to which they can make deductible payments.[44] Thus, these independent U.S. resident corporations find themselves at a tax-based disadvantage in the U.S. market vis-à-vis the UKCo-USSub MNE.[45]

The result of this competitive disadvantage is an incentive for foreign corporations to create or acquire U.S. corporations, or for U.S. corporations to create foreign parent corporations to invert their corporate structure, so that the earnings stripping advantage can be achieved.[46] We see no reason for the United States to provide a revenue-losing tax incentive for these corporate formations or acquisitions.[47]

  1. Tax Competition

To the extent that the earnings stripping limitations in § 163(j) are avoided or inapplicable for the reasons discussed later in this Article,[48] the earnings stripping illustrated in Example 1 allows the foreign investor, UKCo, to earn U.S. business income (i.e., earn income in the U.S. market) without paying U.S. income tax. This is clearly an incentive for UKCo and other foreign corporations to invest in U.S. business activity. So might an indulgent U.S. attitude towards earnings stripping attract large amounts of foreign investment that would produce a prodigious U.S. business expansion? Would this, in turn, yield domestic economic benefits sufficient to more than pay for the Treasury’s revenue loss resulting from a zero U.S. tax on the stripped income? A full answer to these questions would require a separate article that reviews the extensive literature regarding whether tax competition does or does not produce sufficient benefits to warrant low or zero tax rates. This literature is, at best, mixed regarding the effects of tax competition.[49] Moreover, the investment capital that tax competition seeks to attract is already abundantly available in the United States.[50] Consequently, we will not engage in a lengthy detour into the tax competition debate. For present purposes, we will assume, we believe realistically, that the Treasury’s revenue loss from earnings stripping will not yield a more-than-compensatory cornucopia of benefits.

The old-fashioned form of tax competition involved tax increases in the form of tariffs instead of tax subsidies.[51] In developing appropriate constraints on earnings stripping we are mindful that we do not want to turn to protectionism of U.S.-produced goods and services. Defense of the U.S. income tax base is not in any way protectionism, however. As shown above, it serves as a constraint on both U.S. as well as foreign resident taxpayers. Moreover, unlike a tariff that is imposed only on foreign producers, an income tax policy directed at moving toward neutral taxation of U.S. and non-U.S. controlled taxpayers should reduce, not increase, economic distortions. Under the principles we develop below, normal market returns are respected for foreign as well as U.S. business activity.

  1. Does the United States Have a Normative Source Taxation Right?

The purpose of this Article is to explore ways to eliminate or reduce the benefits of earnings stripping by the UKCo-USSub MNE in Example 1 and similarly situated foreign enterprises that sell into the U.S. market. This purpose assumes, of course, that the United States has a normative right to tax the U.S.-source income of the UKCo-USSub MNE. We have addressed that question in prior work and explained that the United States does, indeed, have such a right based on a benefits rationale, on the right of a nation-state to determine the conditions for access to its economy by nonresidents, and on considerations of fairness to its own residents who compete in the U.S. domestic economy against nonresidents.[52] Moreover, the right of a country to tax nonresidents on income earned within its borders is well established as a principle of customary international law.[53] Thus, it is appropriate for the United States to address the earnings stripping strategy illustrated in Example 1.

  1. Should the Focus Be Limited to Inversions and Interest Payments?

As shown in Example 1, earnings stripping can take place without being connected with a corporate inversion, and earnings stripping can be accomplished by means of royalty and services fees payments in addition to interest payments. Nevertheless, the 2007 Treasury report on earnings stripping seemed to endorse the view that because (1) “the use of related-party debt arguably is the most readily available method of shifting income out of U.S. corporations,”[54] (2) the “data on . . . [inverted corporations] strongly suggests that these corporations are shifting substantially all of their income out of the United States, primarily through interest payments,”[55] and (3) “the Treasury Department is unable to quantify accurately the extent of earnings stripping [by U.S. subsidiaries outside of the inversion setting],”[56] earnings stripping reforms should focus on interest deductions in the inversion scenario and action regarding other payments and scenarios should be deferred until further data is gathered.[57]

This view strikes us as unsatisfactory because the Treasury report also recognized that deductible payments other than interest can be used for earnings stripping[58] and that “opportunities for earnings stripping are not limited to inversion transactions . . . [but] are present in cases where a U.S. business is structured from the outset with a foreign parent and in cases where a foreign corporation acquires a U.S. operating group.”[59] Moreover, the aggressive behavior of corporations in using other tactics to shift profits out of the U.S. tax base[60] ipso facto suggests that earnings stripping is surely being employed for the same purpose in scenarios other than corporate inversions even if the confirming data has not yet been collected.[61]

The recent ScottishPower[62] case is illustrative. ScottishPower PLC structured its 1999 acquisition of PacifiCorp for $6.5 billion (and assumption of debt) to insert $4.896 billion of fixed and floating rate intra-MNE debt into the U.S. acquisition entity (resulting in approximately a 3:1 debt to equity ratio).[63] The Internal Revenue Service (“IRS”) lost its challenge to the deduction of $932 million of interest paid over three tax years, 2001 to 2003.[64] As another example, in 2009, the IRS withdrew its challenge to the deduction of interest paid on $13.5 billion of intra-MNE debt between GlaxoSmithKline Holdings (Americas) Inc. and a Swiss affiliate.[65] Neither of these cases involved inversions and many more examples could be adduced.[66] Consequently, we conclude that consideration of earnings stripping reforms should not be limited to interest payments that are connected with corporate inversions. Moreover, consideration should be given to protecting the U.S. tax base from other types of earnings stripping deductible payments (such as royalties and services fees) made to both related parties and tax indifferent third parties.[67]

Reform, however, requires identification of an underlying policy and that policy cannot be ascertained until a theoretical baseline is established. Part II deals with that matter.

  1. Establishing the Policy Baseline
  2. The Need for a Policy Baseline

The principal U.S. tool for restraining cross-border earnings stripping is § 163(j) of the Code. This provision, in highly simplified terms, limits a U.S. corporation’s current deduction for interest payments to related parties if the U.S. withholding tax rate on the payments is subject to a treaty reduction and if the U.S. corporation has a year-end debt-to-equity ratio greater than 1.5 to 1.[68] However, the deduction limitation applies only to the excess of the corporation’s net interest expense over 50% of its adjusted taxable income (i.e., taxable income calculated without taking net interest expense and certain other items into account),[69] and the amount disallowed is carried forward to later taxable years.[70] Section 163(j) does not affect corporations that can either keep their aggregate debt within the 1.5 to 1 ratio or keep their net interest expense below 50% of adjusted taxable income.[71] Moreover, importantly, § 163(j) does not apply to other deductible payments, including royalties and services fees, made by a U.S. corporation to a related party.[72] In addition, it is difficult to discern any coherent theory underlying § 163(j) tests for determining when to allow a deduction for interest paid to related parties. Thus, § 163(j), in its current form, is both under-theorized and not very effective in restraining earnings stripping. Although § 163(j) also applies to interest paid to tax-exempt organizations,[73] our focus is on the theoretical basis to combat cross-border earnings stripping.

For those concerned about earnings stripping, a seemingly obvious reform strategy might be to make § 163(j) less generous by extending it to royalty and services fees payments made to foreign related parties and by tightening the 50% of adjusted taxable income and 1.5 to 1 debt to equity benchmarks. We will examine those possibilities in Part III.B of this Article. Before we can do that, however, we must first deal with certain difficult foundational questions. What is the normative basis for limiting earnings stripping deductions? Should the limitations be less generous than current § 163(j), and to what extent? What is the correct degree of limitation on deductions for payments to related foreign parties? To answer those questions, it will be helpful to establish a policy baseline and develop a theoretically coherent approach for restraining earnings stripping. The remainder of Part II addresses that topic.[74]

  1. The No Deduction for Costless Payments Principle

Example 2 provides a useful place to begin by considering unrelated party financing.

Example 2: USCo, a U.S. corporation, borrows at a 10% per annum interest charge from an unrelated foreign bank and invests the proceeds in a U.S. business activity that yields a 10% annual return. Obviously, this produces an economic wash; the interest charge on the loan exactly equals the investment return, which means that there is no gain or loss because the interest charge is a cost of pursuing USCo’s business activity that must be taken into account to properly measure USCo’s economic result. Therefore, a tax deduction must be allowed for the interest charge so that USCo’s tax result will reflect economic reality. The same point applies to royalty payments and services payments that USCo makes to unrelated parties. These conclusions are based on the fact that the U.S. income tax is a levy on net income.[75] Thus, when business-connected interest payments,[76] royalty payments, and services payments are made to independent payees, they should be deducted for income tax purposes because they are costs of earning income and the income tax base cannot be properly measured unless these costs are subtracted.

Now compare that fact scenario with Example 3 where a third-party lender is not involved.

Example 3: USSub is a wholly owned U.S. subsidiary of UKCo, a U.K. corporation with no external debt. USSub borrows funds representing retained earnings from UKCo at 10% and invests the proceeds in a U.S. business activity that yields a 10% annual return. In this situation, if a deduction were allowed, it would disguise the fact that the MNE’s U.S. activity yielded a 10% return that effectively has been internally allocated from USSub to UKCo by means of interest payments that do not reflect a real offsetting interest expense.[77] Stated differently, the interest payment is not an expense that produces an economic wash in relation to the U.S. income tax base; it effectively is a distribution of income within the UKCo-USSub MNE. Considered this way, USSub should not be allowed to deduct its interest payments to UKCo because they effectively are distributions in respect of equity.

  1. Real Expenses vs. Costless Foreign Related Party Payments

If UKCo borrowed from an independent party to obtain the funds to loan to USSub in Example 3, then the interest paid to the independent lender would be a real cost rather than an internal distribution. Thus, to that extent, USSub’s interest payments to UKCo would be reimbursements for a real cost. If the real cost is attributable to U.S. business activity, which we assume at this stage of the discussion, the reimbursed cost should be a deductible expense for U.S. tax purposes in determining the U.S. taxable income of the UKCo-USSub MNE.[78] Likewise, USSub should be allowed to deduct royalty payments to UKCo to the extent they reimburse UKCo for the cost of developing its own intangibles or for the cost of acquiring the use of unrelated party intangibles that UKCo sublicenses to USSub and that USSub actually uses in its business. Finally, UKCo’s actual labor costs incurred in providing management services from its headquarters to USSub should also be deductible for U.S. tax purposes, assuming that they benefit the U.S. business, because these are real expenses rather than internal income distributions. These conclusions do not, however, apply to amounts charged to USSub in excess of UKCo’s actual intangible development and acquisition costs and actual labor costs, without regard to whether the charges exceed an arm’s-length amount.

One might protest that even if USSub’s payments to UKCo do no more than reimburse UKCo for actual expenses incurred in transactions with third parties for USSub’s benefit, they nevertheless erode the U.S. tax base if they are treated as deductible. Speaking generally, however, the U.S. tax base is not gross income; it is net income (i.e., gross income minus the expenditures allowed as a deduction).[79] USSub’s reimbursement payments to UKCo are real costs to the UKCo-USSub MNE of producing income through activity in the United States. They are not costless transfers and their deduction is essential for purposes of properly measuring USSub’s net income. Stated differently, deductions for these payments measure, but do not transfer, USSub’s net income and, therefore, these deductions do not erode the U.S. income tax’s net income base. Thus, allowing them to be deductible does not amount to tolerating base erosion. To hold otherwise would lead to the conclusion that no deduction should be allowed for any business expenses paid to foreign unrelated parties in arm’s-length transactions. This would amount to an economically indefensible definition of the tax base and could have a deleterious effect on free trade.[80]

But, to recapitulate, up to this point our analysis suggests that (i) interest charges paid by USSub to UKCo for use of the latter’s own capital, (ii) royalties that exceed the related development costs paid by USSub to UKCo for use of the latter’s self-developed intangibles, and (iii) services fees paid by USSub to UKCo that exceed UKCo’s actual cost of providing those services are all internal income distributions that should not generate U.S. income tax deductions for purposes of properly measuring USSub’s net income because they do not represent real costs.[81] To reflect this point, we will use the terms “costless foreign related party expenses” and “costless foreign related party payments” throughout this Article as a short-hand reference to items that merely shift income within an MNE. In contrast, we use the terms “real expenses” and “real payments” in referring to outlays that are not costless foreign related party expenses or costless foreign related party payments.[82]

We will discuss several further points below. First, should UKCo earn a normal return on its expenditures? Second, should UKCo earn more than a normal return on its expenditures or otherwise receive payments for costless foreign related party expenses? If economic rents are earned by the UKCo-USSub MNE, when should they appropriately be allocated away from USSub to UKCo (an inter-nation income allocation question), and when is a deductible payment the appropriate legal mechanism for the transfer of income?

  1. Combined Reporting vs. Separate Accounting

The foregoing conclusions assume that it is analytically correct to treat UKCo and USSub as a single economic enterprise. This seems indisputably so. The separate corporate existence of UKCo and USSub for income tax purposes is a legal fiction maintained for reasons of convenience[83] that can be curtailed or eliminated when the costs of the fiction outweigh the benefits.[84] Such would be the case if UKCo were permitted to remove income from the U.S. tax base merely by creating documents between itself and USSub, a controlled party that has no mind and will of its own.[85] Moreover, UKCo and USSub are treated as a single enterprise with respect to financial reporting of profits and losses,[86] and formulary apportionment systems with combined reporting would impose the same result.[87] In addition, if USSub were the parent and UKCo were the subsidiary, the two corporations would be effectively treated as one for purposes of the indirect foreign tax credit.[88] Finally, treating UKCo and USSub as a unit is consistent with the contemporary view that the corporate members of an MNE constitute an integrated enterprise in which the separate entity status of the members is a legal fiction with no economic substance.[89]

Those who object to treating UKCo and USSub as a combined enterprise will point out that under the arm’s-length approach embedded in U.S. transfer pricing law, and in our tax law generally,[90] they are treated as separate entities whose contracts with each other generally are respected for purposes of allocating income and costs.[91] However, this aspect of U.S. international taxation exists primarily because its defenders insist that it is the best practical solution to international allocation problems and not because it has a sound normative basis.[92] Indeed, the treatment of parents and subsidiaries as legally distinct entities for international income tax purposes, an approach known as separate accounting, is widely and severely criticized for being based on a legal formality (separate incorporation) that has no economic substance and that exposes the theoretical incompleteness of the arm’s-length transfer pricing regime.[93] That regime does not provide a normative basis for treating UKCo and USSub as separate entities. Indeed, the case for treating them as a single MNE appears unassailable.

  1. Affirming Nondeductibility

This analysis seems to lead to the tentative conclusion that in Example 1, the payments from USSub to UKCo should not be deductible for U.S. income tax purposes to the extent they are costless foreign-related party expenses. Instead, it would appear that such expenses should be ignored and USSub’s deductions should be limited to real expenses. Speaking more broadly, it would seem that in principle there should be no U.S. deductions for earnings stripping payments made to related foreign parties except to the extent the payments represent real expense. This tentative conclusion will be revisited in Part II.C., but, first, we will consider certain aspects of a rule of complete nondeductibility.

  1. Conflict with Existing Domestic Law and Treaties


Notwithstanding the theoretical analysis above, existing U.S. federal income tax law does, in fact, treat even wholly owned subsidiaries and their parent corporations as separate taxpayers and generally regards payments between them as real.[94] Thus, in Example 1, § 162 would allow USSub to deduct its royalty and services fees payments and § 163 would allow USSub to deduct its interest payments, regardless of whether they represent real costs under the analysis of Example 1.[95] Section 482 would limit all of those deductions to arm’s-length amounts, but the Example 1 facts state that this restriction would not affect USSub’s deductions because none of its payments exceed the arm’s-length standard.[96] Thus, implementing the tentative conclusions regarding nondeductibility in the preceding paragraphs of this Part II.B. would require a change to current U.S. tax law.

This discussion also raises important questions regarding the articles dealing with associated enterprises, interest, royalties, and nondiscrimination in the typical U.S. bilateral income tax treaty. Virtually all such treaties became effective after 1977.[97] Thus, they are based on either the currently applicable U.S. Model Income Tax Treaty of 2006,[98] or the earlier U.S. Models of 1996,[99] 1981,[100] or 1977.[101] The salient provisions addressing associated enterprises, interest, royalties, and nondiscrimination are substantially identical in all four of these U.S. Model versions.[102] Consequently, the extent of any conflict between currently effective U.S. income tax treaties and rules disallowing deductions for payments of interest, royalties, and services fees to foreign-related parties can conveniently focus on the relevant language of the current U.S. Model Treaty.[103]

Articles 11 and 12 of the current U.S. Model provide that treaty-created withholding tax reductions with respect to interest and royalties received by nonresidents are inapplicable to the extent that the interest and royalties exceed arm’s-length amounts.[104] These disqualifications, however, are irrelevant to the earnings stripping problem illustrated in Example 1 because the related party payments of interest and royalties in that example involve only arm’s-length amounts and Articles 11 and 12 do not apply to the fees for services in that example.[105] Moreover, these Articles do not address the issue of whether deductions can be denied solely on the ground that expenses are costless. Thus, these provisions of the U.S. Model can be dismissed for purposes of this discussion. Nevertheless, this is not the end of treaty questions.

Article 9 of the current U.S. Model allows treaty countries to make profit adjustments between related parties to correct transfer pricing abuses, including abuses involving interest, royalties, and services fees. It then adds that when one treaty party makes such an adjustment, “that other State shall make an appropriate adjustment to the amount of the tax charged therein on those profits.”[106] Does this language carry the negative implication that if dealings between related parties do not involve a transfer pricing abuse, there can be no adjustments that would affect the tax liability of one of the related parties, not even a denial of deductions for costless expenses? Although nothing in the language of Article 9 can be fairly interpreted to create such an implication, the U.S. Treasury Department’s Technical Explanation of the U.S. Model states with respect to Article 9 that “[i]f the conditions of the transaction are consistent with those that would be made between independent persons, the income arising from that transaction should not be subject to adjustment under this Article.”[107] In our view, this quoted statement does not impose a blanket prohibition against making adjustments to arm’s-length transactions. The italicized language indicates that the quotation applies only to adjustments based on the authority of Article 9. It says nothing about denying deductions for expenses on the ground that the expenses are costless. Thus, a statutory rule that prohibits the deduction of costless expense payments to foreign related parties would not, in our judgment, conflict with Article 9 of the typical U.S. income tax treaty.

That conclusion brings us to the current U.S. Model’s Article 24,[108] which is the nondiscrimination provision contained in U.S. income tax treaties. The portions of Article 24 that are relevant to adoption of a U.S. rule denying deductions for costless expense payments by U.S. resident corporations to foreign related parties are Paragraphs 24.1, 24.4, and 24.5.[109] Paragraph 24.1 provides that nationals of treaty partners (including corporations)[110] shall not be subjected to more burdensome taxation than U.S. nationals “in the same circumstances.”[111] Arguably, this has no impact on the disallowance of deductions for costless expense payments by U.S. corporations to foreign related parties because the disallowance increases the U.S. tax liability of the U.S. corporation, not the U.S. tax liability of the foreign related party. The counterargument would be that the foreign related party may bear some of the incidence of the increased tax on the U.S. corporation[112] and that this should make Paragraph 24.1 applicable.[113] There is no authority directly on point but the fact that Paragraph 24.4 deals explicitly with denial of deductions suggests that the more general language of Paragraph 24.1 does not conflict with a domestic rule denying deductions for costless expenses paid to foreign related parties by U.S. corporations.[114]

More importantly, if the United States were to adopt a rule prohibiting deductions for costless expenses paid by U.S. corporations to related foreign parties, but not to related domestic parties, Paragraph 24.1 states that such a rule would not be a violation thereof unless the foreign and domestic payees were “in the same circumstances.”[115] Regarding this point, the U.S. Model’s Technical Explanation states that “if one person is taxable in a Contracting State on worldwide income and the other is not, . . . distinctions in treatment would be justified under paragraph [24.]1.”[116] Since U.S. resident payees of costless expenses are taxable by the United States on their worldwide incomes but foreign payees of such expenses are usually taxable by the United States only on U.S.-source income, it would seem that a U.S. rule prohibiting deductions for costless expense payments by U.S. corporations to foreign related parties but not to U.S. related parties would not conflict with Paragraph 24.1 of the standard U.S. bilateral income tax treaty.[117] This conclusion is supported by the fact that unless a related domestic payee is a tax-exempt entity, deductible payments to domestic related payees remain in the U.S. tax base while deductible payments to foreign related payees that are not controlled foreign corporations are excluded from the U.S. tax base.[118]

There is, however, more to discuss regarding Article 24. Paragraph 24.5 states:

Enterprises of a Contracting State, the capital of which is wholly or partly owned or controlled directly or indirectly, by one or more residents of the other Contracting State, shall not be subjected in the first-mentioned State to any taxation or any requirement connected therewith that is more burdensome than the taxation and connected requirements to which other similar enterprises of the first-mentioned State are or may be subjected.[119]

For purposes of this provision, the term “enterprises” has been interpreted to include business activity carried on by a corporation.[120] Thus, in Example 1, USSub is an enterprise of the United States within the meaning of Paragraph 24.5. However, Paragraph 24.5 only applies to domestic rules that discriminate on the basis of foreign ownership.[121] A U.S. provision that prohibits deductions for costless expense payments made to foreign related parties, but not to U.S. resident related parties, will be acceptable under Paragraph 24.5 so long as it makes no distinction between U.S. payors that are U.S.-owned and U.S. payors that are foreign-owned.[122] This is precisely the type of rule that is suggested by the analysis in this Part II.B., and, therefore, there would be no conflict with Paragraph 24.5 of the current U.S. Model.

The preceding analysis also leads to a solution for the problem posed by Paragraph 24.4. This paragraph provides that arm’s-length amounts of “interest, royalties, and other disbursements paid by a resident of a Contracting State to a resident of the other Contracting State shall, for the purpose of determining the taxable profits of the first-mentioned resident, be deductible under the same conditions as if they had been paid to a resident of the first-mentioned State.”[123] Paragraph 24.4 does not provide an exception to this requirement for cases where foreign and domestic payees are in different circumstances.[124] Therefore, a U.S. rule prohibiting deductions for arm’s-length amounts of costless expense payments made to foreign related parties, but not to U.S. resident related parties, would appear to conflict with Paragraph 24.4.[125] When Congress enacted § 163(j), which limits a U.S. corporation’s current deductions for interest payments made to related parties but nevertheless allows substantial amounts of such interest to be deducted within the limitation,[126] Congress argued that § 163(j) did not conflict with U.S. treaty obligations.[127] Two of its arguments, evidently made with Paragraph 24.4 in mind, were that § 163(j) is both a rule for identifying non-arm’s-length interest amounts and a thin capitalization rule for detecting dividends disguised as interest payments,[128] both types of rules being free from conflict with Paragraph 24.4.[129] Even if these arguments are valid,[130] however, they surely cannot be stretched far enough to legitimate a rule that would deny deductions for arm’s-length interest payments by adequately capitalized U.S. corporations to foreign related parties but not to U.S. resident related parties. Moreover, these arguments have no application to a rule that limits deductions for other kinds of payments to foreign related parties but not to the U.S. resident related parties.

Fortunately, these difficulties can be overcome by adopting a U.S. rule that prohibits deductions for payments of costless expenses to all related parties, both domestic and foreign. Under this approach, payments to residents of treaty countries will be deductible “under the same conditions as if they had been paid to a resident” of the treaty partner country.[131] This approach involves a broader deduction prohibition than the rule that solved the challenge posed by Paragraph 24.5. That rule was limited to costless payments made to foreign related parties—an approach that does not work under Paragraph 24.4.

In summary, the potential for conflict between the U.S. bilateral treaty network and a U.S. rule that prohibits deductions for costless expense payments made to related foreign parties is most acute with respect to Paragraph 24.4 of the current U.S. Model. The conflict, however, can be eliminated with a rule that applies regardless of whether the related party is U.S. or foreign.

  1. Taxpayer and IRS Gains

A general rule of nondeductibility would have two desirable collateral effects. First, the pressure on the debt/equity distinction would be reduced because interest payments to related foreign parents that did not represent real expenses would be treated the same as dividends; they would be nondeductible. Thus, the taxpayer planning costs and the IRS enforcement resources devoted to the debt/equity distinction would be reduced, and the distortive incentive to over-leverage U.S. subsidiaries would also be reduced.

Second, if all foreign related party payments by U.S. corporations would be nondeductible, the importance of transfer pricing enforcement with respect to outbound interest, royalty, and services fees payments would be greatly lessened.[132] Consequently, the related taxpayer compliance and planning costs would be reduced, and the IRS administrative resources presently used to police this area of transfer pricing could be devoted to other purposes. In a case where the MNE earns economic rents, however, this approach would leave unresolved the issue of whether the rents properly reside in the company where the income attributable to rents is located under the MNE’s accounting methodology. We will take up this question in the next Part.

  1. The Investment Location Primacy Principle

In spite of the strong case for treating all costless foreign related party payments as nondeductible, there are countervailing considerations as illustrated by Example 4.

  1. Intangibles

Example 4: As in Example 1, USSub is the wholly owned U.S. marketing subsidiary of UKCo, a U.K. corporation that has incurred considerable costs in the United Kingdom to develop a marketing intangible that is licensed to USSub for use in its U.S. marketing activities. To aid in clear analysis, we make the simplifying assumption that 100% of USSub’s royalty payments to UKCo equals both a source country market return on UKCo’s cost investment in the intangible as well as an arm’s-length amount. Because the intangible was developed in the United Kingdom, there is a plausible argument for allowing that country an exclusive right to tax the market return on the development cost (which would include recovery of that cost).[133] That market return happens to be 100% of USSub’s royalty payments to UKCo under the facts of this example. We will refer to this as the Investment Location Primacy Principle. It would allow USSub to fully deduct the royalty payments so as to remove them from the U.S. tax base.

Professor Lawrence Lokken has, however, suggested a different conclusion.[134] In 1981, he published an article in which he argued that for purposes of determining the source of a royalty payment, both the country where the related intangible was created (the United Kingdom in Example 4) and the country where the intangible is used (the United States in Example 4), have legitimate source claims.[135] Nevertheless, since he viewed split sourcing as impractical,[136] he concluded that the royalty payment should be sourced entirely to the country of use because that country’s legal protections make exploitation of the related intangible possible and, therefore, give that country the stronger source claim.[137]

Although Professor Lokken’s analysis was developed for purposes of applying the U.S. rules for taxing U.S.-source income of nonresident aliens and foreign corporations,[138] it is relevant by analogy to the inter-nation allocation of the income represented by royalty payments received by a foreign parent corporation from a U.S. subsidiary. If his analysis were applied to Example 4,[139] it would reach the opposite of our conclusion in that it would give the United States the primary taxing claim with respect to the income represented by USSub’s royalty payments. That would mean that the payments would be nondeductible by USSub. The royalty payments would be characterized for U.S. tax purposes as distributions with respect to UKCo’s USSub stock, and the United Kingdom would have the obligation to relieve double taxation.[140]

We respectfully disagree with this approach. At this point, our discussion is purely theoretical, and if USSub pays a royalty to UKCo in Example 4 that is limited to a normal market return on UKCo’s U.K. creation cost, it seems to us that, in theory, this royalty represents a return on the U.K. development work that should be allocated exclusively to the U.K. tax base[141] to the extent that it does not exceed an arm’s-length amount.[142] In contrast, income resulting from using UKCo’s intangible to exploit the U.S. market is appropriately allocated to, and should be taxed by, the United States. In fact, that income is included in the profits of USSub that are taxable by the United States after deductions are allowed for USSub’s royalty payments to UKCo.[143] Thus, in our view, the royalty should be deductible by USSub, up to the lesser of a normal market return on UKCo’s creation cost or an arm’s-length amount, and the deductible amount should be taxed exclusively by the United Kingdom.[144]

However, if the United States applies the No Deduction for Costless Payments Principle developed in Part II.B., USSub will not be allowed to deduct the part of its royalty payments to UKCo in Example 4 that exceeds a recovery of UKCo’s development cost. This is so because the excess payments are simply costless transfers within an MNE.[145] Under that approach, the nondeductible portion of the royalty payments will effectively be included in the U.S. tax base (the opposite of our conclusion just stated above) instead of being allocated exclusively to the U.K. tax base. In contrast, if USSub were allowed to fully deduct its royalty payments to UKCo, the deduction would eliminate the payments from the U.S. tax base, and they would be exclusively in the U.K. tax base.[146] This is the correct result if one accepts the argument that the United Kingdom has the primary taxing right on all returns to the cost of developing the intangible in the United Kingdom. Obviously, in Example 4, there is a conflict between the No Deduction for Costless Payments Principle outlined in Part II.B. and the Investment Location Primacy Principle. We will address that conflict below.[147]

Before doing so, however, we must point out that our argument for assigning royalties to the development country is subject to qualifications. First, our argument that the Investment Location Primacy Principle supports a taxing right on the part of the development country would not justify an income allocation to the development country greater than the lesser of a normal return on the development cost or an arm’s-length amount.[148] Thus, in Example 4, the Investment Location Primacy Principle would allow USSub to deduct only royalty payments that do not exceed that benchmark.

Payments in excess of that ceiling are costless intra-MNE transfers and, therefore, should be nondeductible by USSub for U.S. income tax purposes. Under this approach, supranormal returns, i.e., economic rents, would be taxable by the United States. It is not clear, however, that this is the correct approach. In the context of a particular intangible, economic rent exists because an exuberant consumer preference allows the supplier to charge a price in excess of what is justified by any possible objective superiority of the related good or service. The necessary consumer reaction to the intangible is a consequence of both development work (suggesting allocation to the development country) and consumer mindset (suggesting allocation to the market country). Thus, allocating rent from intangibles between development and market countries is an issue for which there is no easy answer. Supranormal returns or rents are the key point of intersection of earnings stripping and transfer pricing. Our approach to earnings stripping would isolate this critical issue and force taxpayers and tax authorities to identify returns that exceed normal levels and develop explicit methods for dividing these returns for taxation purposes.

It is beyond the scope of this Article to reach final conclusions on the important and difficult issue of how to allocate economic rents, i.e., residual profits, of an MNE; however, this is not a new problem. We have explained in other work why we disfavor formulary apportionment as a method for allocating the entire taxable income from a business.[149] There, we favored use of a residual profit split method and also accepted that there is room for use of factors as a basis on which to allocate income in particular circumstances.[150] The allocation of economic rents is such a circumstance.

The allocation of residual profit also underlies the methodology of the OECD’s authorized approach under Article 7 of the OECD Model Convention.[151] That method emphasizes taking into account the functions performed by, and the assets and risks borne in, each country.[152] Consistent with our analysis of costless payments within an MNE, we are highly skeptical of risk shifting within an MNE. Thus, we would focus on real functions performed by actual employees using real assets. Before formulating our concrete conclusion regarding the difficult problem of allocating rents, however, we look forward to learning the outcome of the rethinking of these issues in connection with the OECD BEPS project.[153] We hope that this Article will be helpful to that project.

Our analysis calls for the application of transfer pricing rules to determine the portion of royalty payments that does not exceed the normal return/arm’s-length ceiling and the excess portion. Although the transfer pricing law on this point is not always precise, our restriction of an allowable royalty deduction to a normal return on an expenditure greatly simplifies this aspect of the calculation. Moreover, we have explained in other work how these methods can be made less vulnerable to taxpayer gaming by allowing the IRS to make retrospective adjustments.[154]

Finally, it should be noted that the most extensive erosion of the U.S. tax base by means of deductible royalty payments is not accomplished by foreign owners of U.S. subsidiaries who use royalty expense deductions to shift income out of the United States.[155] Instead, the most extensive degradation of the U.S. tax base results from U.S. MNEs moving their intangibles to tax-haven corporations in countries where royalties are subject to little or no foreign tax, remain unrepatriated for long periods of time, and then can be sheltered from U.S. taxation by cross-crediting.[156] In other words, it is regrettable that transfer pricing law appears to be the principal tool for limiting earnings stripping through deductible royalties, but that is the lesser portion of the problem of the U.S. tax base being eroded by transfers of intangibles returns. In addition, as we have often observed, a significant part of the overall problem would be substantially curtailed by U.S. adoption of real worldwide taxation.[157]

There is a second qualification to the argument that the Investment Location Primacy Principle gives the country where an intangible was developed an exclusive right to tax the lesser of a normal market return on the development costs or an arm’s-length royalty. It is that, as a practical matter, this principle can be applied only if the ultimate licensor of the intangible is subject to the taxing jurisdiction of the development country, as is the case in Example 4.[158]

Although the foregoing analysis has identified limitations to the Investment Location Primacy Principle, it has left a substantial core of that principle intact with respect to intangibles. That is, the country where an intangible is developed should have a primary right to tax foreign-source intangibles royalties received by a licensor that is subject to its taxing jurisdiction but only to the extent that the royalties do not exceed the lesser of a normal market return on the licensor’s development costs (which includes recovery of those costs) or an arm’s-length amount.[159] Thus, in Example 4, USSub should get a U.S. deduction for its royalty payments that are not greater than that benchmark. The deduction will shift the deducted amount out of the U.S. tax base and make it exclusively subject to taxation by the United Kingdom. In addition, royalties paid by USSub that effectively reimburse UKCo for its costs of acquiring the use of unrelated party intangibles that it has sublicensed to USSub are real costs to USSub rather than costless intra-MNE transfers. Therefore, USSub should be allowed to deduct such payments.

  1. Relevance of a Foreign Tax

What if the USSub royalty payments that we have just argued should be deductible for U.S. tax purposes are not taxed by the United Kingdom, are taxed by the United Kingdom at a rate that is much lower than the U.S. rate, or are paid through the United Kingdom as a conduit to an affiliate in a low-tax country? To the extent that USSub’s deductions are allowed in order to either properly allocate income to the U.K.’s taxing jurisdiction or to recognize real costs borne by UKCo for the benefit of USSub, it may be argued that U.S. tax deductions should be allowed regardless of the extent to which a U.K. tax is imposed on UKCo’s receipt of the royalties.

We have argued in earlier work that income should not be treated as foreign-source for purposes of applying a territorial system[160] unless the income bears a substantial foreign tax.[161] That is because the purpose of the exemption for foreign-source income under a territorial system is to mitigate double taxation that would otherwise chill international trade and investment.[162] However, application of the U.S. income tax to foreign-source income cannot cause double taxation unless that income bears a significant foreign tax.[163] For that reason, we have argued that an exemption for foreign income should not be available if the United States were to adopt a territorial system unless the relevant foreign-source income is subject to a meaningful foreign tax.[164]

Double taxation is not, however, the issue regarding the deductible royalties paid by USSub to UKCo in Example 4. There the objective is to (i) allocate income representing a normal return to the appropriate country or (ii) to take account of real costs borne by USSub that must be recognized in order to properly measure USSub’s U.S.-source net income. In that context, U.S. deductions for USSub’s royalty payments should be allowed to the extent that UKCo has borne real third-party costs for the benefit of USSub, regardless of the U.K.’s tax treatment of those payments. Allocating “normal return” income, however, is an issue of inter-nation revenue allocation and not measurement of income.[165] Thus, it is necessary to determine whether source rules and transfer pricing rules are sufficiently robust that they should be relied upon to allocate even normal returns between countries in the absence of a meaningful foreign tax on the income.[166]

As observed above, attributing income to a geographic source according to the legal category of income is a “weak margin.”[167] That is, it is not a strong basis on which to determine important tax consequences because source of income does not rest on a strong normative basis and, in a range of cases, is susceptible to manipulation.[168] The Investment Location Primacy Principle can be relied upon to identify which entity should earn income with respect to a third-party cost, but, if there is no such third-party cost, the “investment location” is a function of non-arm’s-length arrangements among related parties and their dependent agents.[169] It is, thus, possible to inappropriately manipulate the investment location. A subject-to-tax test adds a level of adversity with which to test the reality of the claimed investment location. Without the friction of a meaningful foreign tax, normal returns may be artificially shifted and the source country denied revenue that is taxed nowhere or, if at all, at a low rate.

It is internationally recognized that the source country is entitled to tax income arising within its borders and owned by foreign investors.[170] As a matter of inter-nation equity, this principle permits the source country to derive its share of revenue from the foreign-owned capital in its country.[171] Moreover, because of the “weak margin” concerns, the source country should be allowed to assert its claim to revenue by denying deductions if the investment location country does not tax the income.[172] In other words, we would allow inter-nation revenue allocation concerns to take precedence over the Investment Location Primary Principle in circumstances where the normal return would not be subject to a meaningful foreign tax. We next consider how to apply this approach in the context of hybrid mismatches and taxation under another country’s controlled foreign company rules.

Hybrid mismatches present a situation of current concern where the existence, or not, of a foreign tax liability is relevant. From the U.S. standpoint, a hybrid mismatch is presented when (1) a payment from a U.S. taxpayer to a foreign resident bears no U.S. tax because the payment is deductible for U.S. purposes and (2) the payment also escapes foreign tax because it is excludable from the foreign payee’s income under the applicable foreign tax regime.[173] The result of this asymmetry is income that is taxed nowhere, also referred to as “double non-taxed income,” “homeless income,” and “stateless income.”[174]

Hybrid mismatches that are relevant to U.S. earnings stripping concerns can arise when a U.S. corporation issues a financial instrument that is treated by the U.S. tax system as debt producing deductible interest payments but that is characterized by the foreign holder’s residence country as an equity investment yielding exempt dividend income.[175] Obviously, the U.S.-source income that funds the U.S.-deductible interest payments is not taxed in either the United States or the holder’s residence country.

Hybrid mismatches can also be created by the use of so-called hybrid entities—i.e., business organizations that are treated as taxpayers in one country but disregarded as such in another country.[176] In the U.S. earnings stripping context, this scenario arises when a U.S. corporation pays a business expense to a foreign organization that is recognized as a corporate entity under U.S. law but that is ignored by the foreign country and treated there as an unincorporated branch of the U.S. payor.[177] In such a case, the payment bears no U.S. tax because it is deductible under U.S. law and it bears no foreign tax because the foreign jurisdiction regards the payment as a meaningless internal cash transfer from one part of the U.S. corporation to another component of the same corporation.[178]

The problem with these results is that they yield income that is taxed nowhere. Thus, U.S. taxpayers have a distortive incentive to inefficiently pursue this income in lieu of taxable alternatives that are economically superior from a before-tax perspective.

The significance of this hybrid mismatch issue and the full range of appropriate responses are matters that are outside the scope of this Article.[179] For our purposes, it is sufficient to note that the Obama Administration has proposed legislation that, generally speaking, would deny deductions to U.S. taxpayers for interest and royalty payments if the involvement of a hybrid instrument or a hybrid entity resulted in no income inclusion by the foreign related payee under the relevant foreign tax regime.[180] The OECD is engaged in considering a similar proposal.[181]

These proposals would be consistent with our analysis that the source country should be permitted to tax income that otherwise would be subject to double non-taxation by reason of these structured arrangements. They are consistent with our view that USSub’s deductions have a normative basis when representing a real cost that must be taken into account to properly measure net income. In contrast, there is no normative basis for determining when a business organization should be treated as a separate taxpayer that can receive deductible payments and when it should be treated as an integral component of its economic owner so that payments to it by the owner are disregarded.[182] Likewise, there is no normative guide for determining whether an investor should be treated as a creditor or an equity holder.[183] These decisions are based primarily on considerations such as the relative convenience of collecting a tax from a single source (the business organization) instead of from multiple sources (the owners) and judgments about what kinds of investor rights and priorities are necessary to allow borrowers to adequately raise capital.[184] When making such prudential decisions, it is appropriate to weigh the policy and revenue costs that result from permitting the creation of double non-taxed income. When the costs are too high, denying deductions for payments is an appropriate response.

When considering the effect of disqualifying non-taxation, an important practical question is whether indirect taxation under controlled foreign company rules should be treated as taxation for purposes of a subject-to-tax test. This is an additional element in the inter-nation allocation of income. The issue generally arises when a payment is made to a company that is a brother-sister affiliate of a common parent company. If the payment is not subject to a meaningful tax in the hands of the company receiving the payment but is included in the common parent company’s income and subjected to tax under controlled foreign company rules applicable to the parent company, should the source country allow the deduction? In this circumstance, the conditions for the allocation of income to the investment location should be considered satisfied even though the tax is imposed by a third country on the common parent and not on the income recipient. This position is consistent with our approach of recognizing the single economic nature of the multinational business enterprise. In that circumstance, we would allow a deduction so that income constituting a normal return may be taxed in the investment location country. For this purpose, we are in essence treating the “rest of the world” as a single country.

  1. Fees and Interest

To what extent does the preceding analysis regarding intangibles also apply to the interest and management fees that USSub pays to UKCo in Example 1? It strikes us as appropriate to treat UKCo’s wage payments to its headquarters employees who furnish services to USSub as the equivalent of an investment with a U.K. location. Thus, the United Kingdom should have the primary right to tax a normal market return on that investment.[185] This means that USSub should be allowed a deduction for that return in addition to its deduction, as explained in Part II.B.,[186] for the portion of the fees that equal UKCo’s actual labor cost. However, USSub’s deduction should not exceed the sum of UKCo’s actual labor cost plus a normal markup on that cost because nothing in excess of that sum can be characterized as deductible under the Investment Location Primacy Principle.[187] Presumably, the deductible sum is the same as an arm’s-length amount. This means that transfer pricing law must carry the burden of determining the market rate return that should be included in USSub’s deduction.

The internal cash that UKCo lends to USSub in Example 1 is also an investment, and UKCo can expect a return on that investment in the form of interest at a market rate. USSub’s use of the borrowed funds in the United States, however, produced that interest. Thus, the United States has the primary taxing claim to that interest,[188] and the Investment Location Primacy Principle would not require that the United States give USSub a deduction for its interest payments to UKCo.[189] In other words, both the No Deduction for Costless Payments Principle and the Investment Location Primacy Principle support the conclusion that USSub’s interest expense in Example 1 should be nondeductible for U.S. tax purposes.

  1. Which Principle Prevails in the Case of a Conflict?

In the immediately preceding discussion, the Investment Location Primacy Principle and the No Deduction for Costless Payments Principle gave identical answers regarding USSub’s interest payments on loans made by UKCo out of retained earnings—namely, no U.S. deduction by USSub. There is, however, a conflict between these principles with respect to royalties and fees for headquarters services. The No Deduction for Costless Payments Principle posits that a U.S. subsidiary’s deductions for use of its foreign parent’s self-developed intangibles should be limited to the parent’s actual development costs. It also suggests that the subsidiary’s deduction for the cost of services provided by the foreign parent should not exceed the parent’s actual labor costs.[190] In contrast, the Investment Location Primacy Principle would allow a subsidiary to deduct royalty payments for use of its foreign parent’s self-developed intangibles up to the lesser of a market return on the parent’s development costs or an arm’s-length amount.[191] It also would allow the subsidiary’s deduction for services fees to be the sum of the parent’s actual labor costs plus a normal market return thereon.[192]

Thus, the preceding conflict must be resolved. Article 9 of the widely adopted OECD Model Income Tax Treaty states the overarching principle and international norm that where related corporations deal with each other, as in the case of USSub and UKCo in Example 1, then their incomes from their dealings with each other shall be calculated as if they had been independent parties dealing at arm’s length.[193] The purpose of the deductions that are allowed to USSub under the Investment Location Primacy Principle is to give the United Kingdom the same primary right to taxing jurisdiction that it would have if UKCo were dealing with USSub at arm’s length and insisting on a normal market rate of return with respect to the services and intangible property provided by UKCo to USSub (after application of the subject-to-tax limitation on the principle).[194] The Investment Location Primacy Principle is a form of jurisdiction-to-tax rule whereas the No Deduction for Costless Payments Principle is a base defining rule.[195] Generally, jurisdictional rules should trump base defining deduction disallowance rules.[196] Consequently, we conclude that the allowance of deductions under the Investment Location Primacy Principle should trump a disallowance required by the No Deduction for Costless Payments Principle. When, however, an expense of USSub represents a real cost, a U.S. deduction should be allowed even if the effect is to allocate income to another country. Thus, interest paid by USSub to unrelated parties in Example 2 should be deductible by USSub even though the related debt represents an investment by the lender in the United States.[197] This result is appropriate because the U.S.-source tax on corporate profits is a net income tax, and deductions for real costs are required to measure net income.[198]

  1. Summary

Thus, the end result of the effort to develop a policy baseline is the conclusion that the United States should allow deductions for payments for expenses (but not capital expenditures) made by U.S. subsidiaries to foreign parent corporations if the deduction is necessary to properly allocate income between the United States and the foreign parent’s residence country. If a deduction is not required under this principle, then it should be allowed only to the extent that it is for a real current expense as opposed to a costless foreign related party payment, a distinction explained in Part II.B.[199]

  1. Double Taxation

But what if the United Kingdom taxes USSub’s payments of interest, royalties, and fees to UKCo? To the extent that these items were nondeductible for U.S. tax purposes, they would also be part of USSub’s U.S. taxable income.[200] The resulting double taxation would have a chilling effect on UKCo’s participation in the U.S. market.[201] There is, however, a well-settled solution to this problem—customary international law requires the United Kingdom to resolve the double tax problem with a credit for the U.S. tax or an exemption for the interest, royalties, and fees.[202]

III. Searching for Proxies and Rough Justice Rules

  1. The Problem

The preceding discussion of a normative baseline suggests that a theoretically correct approach to earnings stripping requires the following difficult allocations regarding payments by foreign-controlled U.S. subsidiaries to related foreign parties:

  1. Interest payments must be divided between (i) interest on loans to a U.S. subsidiary from purely intra-MNE funds (nondeductible interest) and (ii) interest on loans to the U.S subsidiary from funds supplied to the MNE by third-party creditors (deductible interest to the extent that the arm’s-length benchmark is not exceeded);
  2. Royalty payments must be divided between (i) a normal return on the related foreign developer’s costs (deductible), (ii) royalties that do no more than reimburse the MNE for royalties paid to an unrelated party for intangibles licensed by the MNE to the U.S. subsidiary (deductible), (iii) economic rents (unresolved treatment), and (iv) all other royalty payments by the U.S. subsidiary to the MNE (nondeductible); and
  3. Payments for services provided by the MNE must be divided between (i) a normal return on services costs (deductible), (ii) payments that reimburse the MNE for actual services costs (deductible), and (iii) any amount in excess of the arm’s-length ceiling (nondeductible).

As discussed above,[203] we would separately allocate economic rents formally earned in the country of source.

These are difficult, complex allocations and they rely in part on transfer pricing law that is widely and properly regarded as problematic.[204] These problems are magnified in the case of a large MNE with many subsidiaries and many intra-MNE transactions.[205] Thus, rules intended to achieve theoretically correct results may be objectionable on the grounds that their enforcement and compliance costs are too high and that sufficient accuracy cannot be achieved at any cost. Moreover, a rule of complete disallowance for interest payments on loans of purely intra-MNE funds may be politically impossible. Consequently, it is appropriate to search for proxy rules and rough justice approaches that are an acceptable compromise between practical realities and theoretical correctness. The remainder of Part III discusses this topic.

  1. Section 163(j) As a Rough Justice Provision
  2. Identifying the Loopholes

Although earnings stripping can be effected to some extent through payments of royalties and services fees, the largest amount is accomplished through interest payments.[206] We have previously explained that § 163(j) is intended to constrain this type of earnings stripping by limiting interest deductions but that the § 163(j) constraint is so loose that it is not significantly effective.[207] Perhaps a practical solution for the largest part of the earnings stripping problem is to tighten the parts of § 163(j) that trigger its application. We will now investigate that approach.

The present version of § 163(j) prevents a U.S. corporation from currently deducting its disqualified interest expense for any year with respect to which its year-end debt-to-equity ratio is greater than 1.5 to 1.[208] For this purpose, disqualified interest expense is calculated by computing the part of each related party interest payment during the year that is proportional to any treaty reduction in the U.S. withholding tax on that payment and then aggregating those parts.[209] However, the deduction disallowance cannot exceed the corporation’s excess interest expense for the year.[210] Under § 163(j), excess interest expense is the amount by which net interest expense (i.e., disqualified interest payments plus all other interest expense for the year minus the year’s includible interest income) exceeds 50% of adjusted taxable income.[211] For this purpose, adjusted taxable income is taxable income for the year increased by net interest expense and certain other items.[212]

The current § 163(j) limitations are generally intended to mimic the cash flow debt payment coverage ratios that routinely are used by lenders in credit agreements to assure that the borrower has sufficient cash generating power to repay the loan.[213] The carryover of suspended interest allows the borrower flexibility and, assuming constant cash flows, allows suspended deductions to be recouped as debt is paid down.[214] The limit of 50% of adjusted taxable income is an arbitrary percentage that has not been reconsidered in light of events occurring since its enactment in 1989 (including the 2008 financial crisis) and simply takes as given the remarkable latitude afforded taxpayers to insert debt into MNE groups.[215] Our prior analysis demonstrates that, in essence, the § 163(j) rules elevate a weak proxy for determining an arm’s-length interest amount over considerations of inter-nation equity.[216]

The preceding summary shows that in the following situations, § 163(j) allows U.S. subsidiaries to deduct interest payments that should not be deductible under the policy baseline that was developed in Part II of this Article:

(i)    The U.S. subsidiary manages to keep its debt-to-equity ratio from exceeding 1.5 to 1, a fact that is irrelevant under the Part II policy baseline with respect to interest;

(ii)   The U.S. subsidiary manages to keep its disqualified interest expense from being greater than a safe harbor ceiling defined as the point at which the corporation’s net interest expense for the year exceeds 50% of its adjusted taxable income for the year, another fact that is irrelevant under the policy baseline; and

(iii)  The U.S. subsidiary’s interest payments are not subject to a treaty-imposed reduction in the U.S. withholding tax, a point that is also meaningless under the policy baseline.

In addition, § 163(j) allows an affected corporation to carry nondeductible interest forward to future years when its deductibility will continue to be governed by § 163(j).[217] Finally, § 163(j) applies only to interest deductions and does not impose any restrictions on a U.S. subsidiary’s deductions for payments of royalties and services fees to foreign related parties.[218] Indeed, its design is based solely on a debt paradigm.[219]

Section 163(j) is under-inclusive in several respects and, in those cases, it needs to be strengthened. On the other hand, the present version of § 163(j) is also over-inclusive in that it denies deductions for interest payments that do not come within the loopholes described in (i)–(iii) above but that do represent a real expense as explained in Part II.B. Accordingly, in considering whether modifications that strengthen § 163(j) might make it an acceptable approach, its over-inclusiveness as well as its under-inclusiveness must be taken into account. We now turn to that issue.

  1. Strengthening Section 163(j)

To explore approaches for strengthening § 163(j), it will be helpful first to illustrate the effects of the current version of that provision. Examples 5, 6, and 7 serve this purpose. In each case, we assume for simplicity that the USSub stock is the only asset of the foreign parent.

Example 5: ForCo, a foreign corporation, owns all the shares of USSub, a U.S. resident corporation. ForCo paid $100,000 for the USSub stock and provided USSub with the remainder of its capital by borrowing $900,000 from an unrelated bank and lending this money to USSub on arm’s-length terms. All debt formalities are observed. For year one, USSub has adjusted taxable income of $100,000, and it pays $90,000 of interest on the loan from ForCo. This payment constitutes USSub’s net interest expense for the year, and it bears no U.S. withholding tax because of an applicable treaty. ForCo pays the entire $90,000 interest receipt to its bank lender.

USSub’s debt-to-equity ratio is considerably in excess of 1.5 to 1. In addition, the tax-free interest payment is disqualified interest. Therefore, the $40,000 amount by which the $90,000 interest payment exceeds $50,000 (i.e., 50% of USSub’s adjusted taxable income) is nondeductible interest for year 1 under the present version of § 163(j), even though it was attributable to debt owed to an independent party. In this factual context § 163(j) is over-inclusive.

In our view, this result is incorrect. The $90,000 of interest paid to ForCo and then paid onward to the bank lender was a real expense rather than a costless foreign related party expense, and deduction thereof should have been fully allowed instead of being limited to $50,000. The $40,000 deduction disallowance would be correct only if there were some policy standard that objectively characterized $40,000 of USSub’s $90,000 of interest expense as normatively excessive even though the facts of the example state that the full $90,000 was an arm’s-length charge entirely attributable to unrelated party debt and that all debt formalities were observed. We are not aware of any such objective policy standard.[220] Thus, present § 163(j) is over-inclusive in Example 5.

Example 6: If, however, ForCo’s $900,000 loan to USSub had come entirely from ForCo’s retained earnings, then USSub’s $90,000 interest payment to ForCo would have been a costless foreign related party expense for which no deduction should be allowed. Nevertheless, present § 163(j) would allow USSub to deduct the $50,000 of its interest payment that did not exceed 50% of USSub’s adjusted taxable income. In this factual context, § 163(j) is under-inclusive.

Now consider a middle ground version of Examples 5 and 6.

Example 7: The facts are identical to Example 5 except that one-half of the $900,000 loaned by ForCo to USSub came from ForCo’s retained earnings, and the other one-half was borrowed on arm’s-length terms from an independent bank with both debts bearing the same 10% interest rate. Again, USSub’s deduction for its $90,000 interest payment would be limited to $50,000 by present § 163(j), but that is $5,000 too much. In principle, only $45,000 would be properly deductible because only that portion of the interest was allocable to third-party debt. The remaining amount is a costless foreign related party payment that should be nondeductible. Thus, in this scenario, current § 163(j) got the right answer with respect to $45,000 of USSub’s $90,000 interest payment and the wrong answer regarding $5,000 of the payment.

No principled rationale explains the results of these scenarios. Clearly, current § 163(j)’s operation in Examples 5, 6, and 7 is dependent on the largely unprincipled interaction of the chief components of the § 163(j)(1)(A) excess interest expense limitation—net interest expense and 50% of adjusted taxable income—[221]neither of which has any necessary connection to the theoretically correct approach for determining the deductibility of interest payments by U.S. subsidiaries to foreign parents. We could present multiple additional examples, but, in our judgment, those we have offered show that current § 163(j) is too arbitrary in its operation to function as an acceptable rough justice substitute for a principled approach to limiting earnings stripping.

We can now consider addressing these problems by strengthening § 163(j) along the various margins identified above. A useful place to begin is with the feature of § 163(j) that excuses all corporate debtors who prevent their debt-to-equity ratios from exceeding 1.5 to 1.[222] If that ratio were lowered or even eliminated, more corporate debtors would be subject to the § 163(j) interest deduction limitation.[223]

Note, however, that the 1.5 to 1 ceiling did not play any role in the problematic results of Examples 5, 6, and 7, because the debtor’s debt-to-equity ratio was greater than 1.5 to 1 in all three cases. In other words, the erratic results in those three examples would have occurred even if the debt-to-equity requirement of § 163(j) had been reduced to less than 1.5 to 1 or eliminated. Thus, in our view, tweaking § 163(j)’s leverage ceiling is not sufficient to cure the defects illustrated in Examples 5, 6, and 7.[224]

An additional possibility is to reduce present § 163(j)’s 50% of adjusted taxable income safe harbor to something less—say 30%.[225] If that had been the case in Examples 5 and 7, then the safe harbor ceiling in both examples would be $30,000 instead of $50,000. Thus, in Example 5, $60,000 of USSub’s interest payment would be nondeductible, and only $30,000 would be allowed. But this is even less correct than the $40,000 deduction disallowance/$50,000 allowance result under the 50% ceiling that applied in the original facts of Example 5 because, as previously explained, the correct answer under the policy baseline for this “real expense” interest is zero deduction disallowance/$90,000 allowance on account of the related debt being owed to an independent creditor.

Example 7 presents a similar problematic outcome when a 30% (instead of 50%) safe harbor ceiling is assumed. Again, USSub’s interest deduction would be disallowed to the extent of $60,000, and $30,000 would be allowed. However, because one-half of the $90,000 interest expense is a costless foreign related party payment as explained in Part II, the nondeductible amount should have been $45,000, and the deductible amount should have been the remaining $45,000. Here, the deduction disallowance is $15,000 too large, and the allowed amount is $15,000 too small.

The preceding observations should likely be disregarded if empirical data can be found that shows that the interest deductions properly disallowed by the current version of § 163(j) are overwhelmingly greater than inappropriately disallowed deductions. In that event, the over-inclusive effect of a tightened safe harbor ceiling might be small enough to constitute an acceptable cost. That is an empirical question that should be investigated. Although reductions in its existing 50% safe harbor ceiling would provide enhanced protection to the U.S. tax base, they would not cure erratic effects of current § 163(j).

A final margin that suggests a possibility for strengthening § 163(j) is its feature that makes it inapplicable to interest payments in the cross-border context unless they enjoy a treaty-based reduction in the 30% U.S. withholding tax.[226] This exception seems inconsistent with § 163(j)’s anti-earnings stripping purpose because even if there were no withholding rate reduction, so that the full 30% U.S. tax applied to an interest payment, a U.S. subsidiary with a 35% U.S. marginal income tax rate could still create a meaningful five percentage point tax advantage by incurring a 30% withholding tax on deductible interest payments to a foreign parent. This feature of § 163(j) seems ripe for modification, or, alternatively, the withholding tax rate could be increased.

Nevertheless, the withholding tax treaty reduction requirement played no role in creating the aberrant § 163(j) results that are illustrated in Examples 5, 6, and 7 because those examples assumed that an applicable treaty had reduced the U.S. withholding tax to zero, thereby making this requirement irrelevant. Thus, § 163(j) cannot be amended to achieve the appropriate results by tinkering with or eliminating the presently existing treaty withholding tax reduction requirement.

In our judgment, a version of § 163(j) that is tightened on any or all of the three margins discussed above would still produce results that are erratic in terms of the Part II policy baseline. In addition, there seems to be no way to make § 163(j)’s structure applicable to royalty and services fees payments.[227] Indeed, the only thing that can be said for the present iteration of § 163(j) is that its continued existence may be better than allowing totally unrestrained earnings stripping involving interest deductions. Nonetheless, it is difficult to see how § 163(j), even if amended, will have any significant role to play in developing a comprehensive and coherent approach to dealing with earnings stripping.

  1. An Offsetting Withholding Tax As an Alternative to Deduction Disallowance (The Wells/Lowell Proposal)

Commentators Bret Wells and Cym Lowell have proposed addressing the earnings stripping problem by imposing a U.S. “base-protecting surtax” on all deductible payments to foreign related parties.[228] Although they do not fully state the details, they do indicate that the U.S. payor would withhold the tax at the time of the payment, and the tax would be calibrated “to collect upfront the expected U.S. tax that should be due with respect to the residual profits that are represented by the base erosion payment.”[229]

The two authors acknowledge that actually calibrating the tax so that it does no more, and no less, than remove the reward of earnings stripping is a serious challenge.[230] Over-inclusiveness and under-inclusiveness are both potential problems with their proposal. That challenge is avoided by adopting the nondeduction approach towards earnings stripping payments that we have generally advocated in this Article.

Of equal importance is the point that if the withholding tax is to have a sound, principled basis, it will not apply to payments identified as deductible under the analysis in Part II.B. Thus, the Wells and Lowell proposal will involve the same allocation complexities as the nondeduction approach. Thus, although we acknowledge that the Wells and Lowell proposal would be an improvement over existing law and is more coherent than current law § 163(j), we conclude that, on balance, the nondeduction approach that we have developed in this Article is preferable.

  1. Should Nondeduction Be Limited to Payments into Low-Tax Countries?

Notwithstanding the foregoing comprehensive theoretical analysis, we recognize that the earnings stripping problem is most acute with respect to deductible payments made to foreign related parties that are resident in low-tax countries.[231] Given that fact, would a rule making deductibility contingent on the imposition of a meaningful foreign residence country tax be a good alternative to a more theoretically correct nondeductibility rule that requires difficult allocations? This alternative rule would seem to be over-inclusive because it would presumably apply to all payments to parties resident in low-tax countries even if the payments should be deductible in principle under the analysis in Part II.B. More importantly, this alternative also would be under-inclusive because it would require the United States to allow deductions for payments into high-tax countries even though the deductions erode the U.S. tax base. Thus, in our view, this approach would be an improvement over present law, but, before adopting it, other more precise and comprehensive proposals should be seriously considered.

  1. Lodin’s Proposal

Swedish Professor Sven-Olof Lodin has developed another approach for dealing with earnings stripping techniques involving interest and royalty payments.[232] Under Professor Lodin’s proposal, the deductions for interest or royalties of legal entities would be replaced by an equivalent inverted tax credit.[233] In the case of interest or royalties paid to unrelated payees, the credit amount would be based on the current top domestic Swedish corporate tax rate of 26.3% (i.e., that rate multiplied by the borrower’s or licensee’s interest or royalty costs).[234] In the case of interest or royalties paid to related payees, the credit amount would be at the rate that normally applies to the lender’s or licensor’s interest or royalty income (i.e., that rate multiplied by the borrower’s or licensee’s interest or royalty costs), but not in excess of the current top domestic Swedish corporate tax rate of 26.3%.[235] The proposal would allow a company that reports a loss for the year in which the interest or royalty costs are incurred to carry the credit forward in the same manner as losses are carried forward under Swedish law.[236] The proposal also would apply to interest on loans that are indirectly made “by associated companies through guarantees or loans, or their equivalent, to the direct lender.”[237] The purpose of this proposal is to eliminate the tax arbitrage on related party interest or royalty payments that are deductible by the payor but not taxable or taxable at only a low rate in the hands of the related payee.[238]

This proposal, if adopted by the United States, would be a significant improvement over current U.S. tax law. However, several defects undercut its effectiveness. First, as Professor Lodin acknowledges, a serious weakness of this proposal, as applied to interest deductions, is the ease with which companies using intermediaries in high-tax countries that had not adopted a similar proposal could circumvent this proposal.[239] In those cases, it would be difficult for the tax authorities to follow the chain of intermediaries to determine the final beneficiary of the interest payments.[240] Second, unlike our nondeduction for costless payments approach discussed above, the Lodin proposal continues to allow income to be stripped from the domestic tax base, but it reduces the incentive to do so by limiting the tax benefit of the U.S. tax deduction to no more than the foreign related party’s tax cost.[241] Thus, unless the related party’s tax cost were zero, deductible interest or royalty payments to foreign related parties would still result in a revenue loss to the United States, even though the payments do not represent a real cost. Third, to implement its limitation on the tax benefits of the earnings stripping interest and royalties expense deductions, the Lodin proposal seems to require calculation of the related foreign payee’s “normal” effective tax rate with respect to such items.[242] This approach seems to introduce the complexities of determining effective tax rates under Code provisions such as § 954(b)(4).[243] Finally, unlike our proposed analytical framework in this Article, Professor Lodin’s proposal does not deal with services fees payments between related parties. Thus, on balance, we conclude that the approach that we have developed in this Article is a more comprehensive and effective way of dealing with the earnings stripping problem.

  1. The Obama Administration’s 2016 Budget Proposal Regarding Thin Capitalization

As previously explained, § 163(j) does not limit a U.S. corporation’s current deductions for related party interest payments unless the corporation’s year-end debt-to-equity ratio exceeds 1.5 to 1 and the corporation’s net interest expense breaches a 50% of adjusted taxable income ceiling for the year.[244] Thus, so long as a U.S. corporation’s debt level and interest expense are kept within these limits, borrowing by related corporations that comprise a foreign MNE can be structured so that the MNE’s debt and related interest deductions are disproportionately located in high-taxed U.S. subsidiaries instead of low-taxed foreign subsidiaries. The Obama Administration’s 2016 fiscal year revenue proposals include a recommendation to limit the possibility of disproportionately leveraging the U.S. subsidiaries of a foreign MNE.[245]

The Administration’s proposal provides that the U.S. interest expense deduction of a member of a group of related corporations covered by consolidated financial statements would be limited to the sum of two components: (1) the member’s interest income and (2) the proportion of the group’s worldwide net interest expense, computed under U.S. tax principles, which is the same as the member’s proportion of the group’s worldwide earnings (computed by adding back the group’s net interest expense and certain other items).[246] Although the proposed rule’s coverage would exclude financial services entities and groups of related corporations with less than $5 million of net interest expense,[247] coverage would be very broad because MNEs generally use consolidated financial statements, which is the principal touchstone for determining when the new proposal would apply to a non-excluded entity.[248] However, the proposal provides that corporations subject to the new rule would be exempt from the § 163(j) limitation.[249] Thus, the effect of the proposal would be to substantially curtail the application of § 163(j), without actually repealing or amending it.

On the positive side, the proposal would strengthen the restraints on interest deductions in that it would apply to corporations that presently escape § 163(j) because their debt is managed so that it avoids a breach of at least one of the § 163(j) limits referred to above. Most inverted U.S. corporations seem to fall into this category,[250] and so the proposal would likely have a restraining effect on tax-driven inversions.

The proposal’s deduction limitation would affect interest paid on debt to unrelated parties and would conflict with the Part II policy baseline discussed above.[251] This may be justified on inter-nation revenue allocation grounds, where the third-party debt is “over allocated” to the United States to take advantage of the higher value of the deduction at current U.S. corporate tax rates.

The first component of the proposal’s limitation effectively allows tracing to the extent that the U.S. subsidiary has interest income. By privileging interest income in this way, it permits higher levels of indebtedness in relation to the non-interest bearing assets of the U.S. subsidiary borrower. Although this does not involve shifting income to other countries, it would, nevertheless, reduce the U.S. tax base by increasing the debt allowable in relation to non-interest bearing business assets that generate higher margins. Thus, this first component of the proposal reduces its effectiveness in remedying earnings stripping involving interest deductions.

The proposal also would allow U.S. subsidiaries to elect into an alternative interest deduction limitation defined as 10% of adjusted taxable income.[252] This election, like any other, is a one-way ratchet that would create gaming opportunities for well-advised taxpayers; the proposal would probably be improved by eliminating this election. Finally, the proposal does not apply to services fees payments or royalties,[253] so, of course, it does not even attempt to do anything about earnings stripping involving those types of payments.

On balance, the Administration’s proposal appears to be an improvement over current law § 163(j). However, it is not clear that the proposal need supersede § 163(j) nor is it clear that the proposal is a more effective limitation on earnings stripping interest payments than § 163(j) strengthened along the lines suggested in Part III.B.2.

  1. Proportional Allocation

A reasonably simple and workable solution to the investment allocation required by Part II.B.1 may lie in the principles that “money is fungible and that interest expense is attributable to all activities and property regardless of any specific purpose for incurring an obligation on which interest is paid.”[254] Using this principle and the principles of § 864(e)(2) and the temporary regulations under § 861,[255] a U.S. subsidiary’s share of a foreign-controlled MNE’s worldwide interest expense on debt owed to independent parties could be allocated to the U.S. subsidiary by referring to the ratio of the subsidiary’s assets to the MNE’s worldwide assets, and then deducted by the subsidiary. The subsidiary’s interest expense in excess of this allocation would then be considered interest on intra-MNE debt that is nondeductible. The subsidiary, however, would not be allowed to deduct more than its actual interest payments. Anti-abuse rules would be required to disregard spurious third-party debt involving loans from a putative independent lender to the MNE followed by a loan back to the putative independent lender or involving offsetting funds deposited with the putative independent lender by the MNE. These anti-abuse rules would admittedly increase the complexity of this allocation approach, but workable rules of this variety have been developed in other areas.[256]

Under this approach, all U.S. subsidiary interest payments would be classified as deductible or nondeductible in the correct amounts and the over- and under-inclusive results of current § 163(j), as illustrated in Examples 5, 6, and 7, would be avoided. This approach could be designed to work in conjunction with rules for allocating interest expense of the U.S. branch of a foreign corporation. In our view, this approach is the preferred way to deal with earnings stripping interest payments.

  1. Do Revenue Considerations Make All of the Above Substantially Pointless?

If U.S. corporations receive more interest from foreign related parties than foreign related parties receive from U.S. corporations, would the United States be better off by adopting a treaty negotiation policy that treats interest payments by domestic corporations to foreign related parties as fully deductible for purposes of taxing corporate profits in exchange for treaty partner reciprocity? Indeed, the United States has already adopted a similar approach to the withholding tax on cross-border interest payments by pursuing a treaty negotiation policy of reciprocally reducing that tax to zero,[257] even in the case of interest payments that are merely costless intra-MNE transfers under the analysis explained in Part II.B.

Suppose that it were the case that U.S. corporations received more interest from foreign related parties than vice versa and the United States agreed by treaties to allow a full deduction for interest paid by U.S. corporations to foreign related parties, thus removing the interest from the U.S. corporate tax base, in exchange for foreign treaty partners doing the converse. In such a case, the United States could fully tax the inbound interest payments without having to allow a foreign tax credit. The resulting revenue would, in theory, exceed the revenue lost from the United States allowing unlimited deductions for interest payments by U.S. corporations to foreign related parties. This might leave the United States in a better revenue position than if it and its treaty partners took vigorous steps to limit deductions for earnings stripping interest payments. Although that is an empirical question that we do not resolve, the imprecise data that is available indicates that U.S. corporations probably do not receive more interest from foreign related parties than they pay to such parties.[258]

Moreover, even if the United States would achieve a net revenue gain from modifying its treaty network to achieve unlimited reciprocal deductions for earnings stripping interest payments, important negative aspects of earnings stripping explained in Section I.F. would remain. Specifically, U.S. resident corporations that compete with foreign MNEs in the U.S. market, but that are not subsidiaries of foreign corporations, cannot generate income that escapes U.S. tax by earnings stripping.[259] This is the case because these independent U.S. resident corporations have no foreign parent to which they can make deductible payments. Thus, they are at a disadvantage in the U.S. market with respect to foreign MNEs, and this disadvantage would be increased if the United States allowed unlimited earnings stripping. This disadvantage would depress the profits of purely domestic U.S. corporations. It also would create a diminution of the U.S tax base that would pro tanto offset the revenue gain that might arise from reciprocal unlimited earnings stripping. Regardless of the size of that offset, the discrimination suffered by purely domestic U.S. corporations would be problematic.

There is an additional consequence from the foregoing disadvantage to independent U.S. corporations that do business exclusively in the United States. That consequence is a tax-subsidized incentive for U.S. corporations to invert and for foreign corporations to create or acquire U.S. corporations so that they can reap the earnings stripping benefit that is unavailable to purely U.S. domestic corporations. This incentive would be increased by a U.S. policy of unlimited deductions for earnings stripping interest payments and the result would be additional base erosion that would pro tanto offset the hoped-for revenue gain from unlimited reciprocal earnings stripping deductions.

In our judgment, the preceding points mean that the United States should not move to a treaty policy that accommodates reciprocal unlimited earnings stripping deductions. Even if there were quantitative evidence of a revenue gain, we would be reluctant to endorse a policy that systemically favored MNEs over domestic businesses.

Conclusion

The term “corporate inversion” is used to identify several transactional forms by which U.S. resident corporations are converted into foreign corporations or into U.S. subsidiaries of foreign corporations. These transactions are currently a large concern to U.S. tax policy makers, and a lively debate is in progress regarding the best way forward.

From a tax standpoint, corporate inversions are driven by the triple objectives of (1) enabling inverting U.S. corporations to escape U.S. taxation of their foreign-source income, (2) enabling U.S. corporations to effectively repatriate foreign-source income without paying a U.S. tax on such income, and (3) enabling those U.S. corporations to move U.S.-source income out of the U.S. tax base by means of deductible expense payments—a tactic known as cross-border earnings stripping. In previous work, we have explained how the first two of these objectives could be forestalled if the definition of a U.S. domestic corporation were broadened to include a shareholder ownership test and if the U.S. international income tax system were changed into a real worldwide system. In this Article, we addressed ways to forestall the third objective by imposing limits on earnings stripping.

Focusing on inversions, however, results in a view of earnings stripping that is far too narrow. A principal emphasis of this Article has been that earnings stripping presents challenges to the U.S. tax base that are much broader than corporate inversions, and so we have developed an analytical framework for identifying responses to the full menu of earnings stripping tactics, of which inversions are only a part.

That framework showed that deductions for interest payments on intra-MNE debt, which are the largest contributor to earnings stripping, are also the most vulnerable to criticism from a policy standpoint. Consequently, we examined various approaches to limiting earnings stripping interest deductions and concluded that the best promise lay in employing a proportionate allocation approach to distinguishing between interest expenses that are deductible as real costs and interest expenses that should be nondeductible because they are costless foreign related party payments that do not effect a proper inter-nation income allocation.

We concluded that distinguishing between properly deductible and properly nondeductible royalty payments and payments for services is much more difficult and requires reliance on transfer pricing law. Fortunately, these items represent the smaller part of the earnings stripping challenge.

Most importantly, cross-border earnings stripping is devastating to the tax bases of both worldwide and territorial international tax systems. Thus, action needs to be taken to curtail the use of earnings stripping to erode the U.S. tax base without waiting to resolve the controversy over whether the United States should adopt a territorial system or instead significantly strengthen its badly flawed worldwide system.

                    *   Copyright © 2015 by J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay. All rights reserved. For helpful comments on earlier drafts of this Article, we thank Yariv Brauner, C. Bryan Cloyd, Lisa De Simone, Omri Marian, Stuart Singer, and participants in a taxation law panel of the Law & Society Association annual meeting in Minneapolis in May 2014, participants at a taxation in a global economy research symposium, cosponsored by the University of Texas McCombs School of Business and the Waterloo Centre at the University of Waterloo, in Austin in September 2014, participants in a seminar at the Institute for Austrian and International Tax Law of the Vienna University of Economics and Business in Vienna in October 2014, and participants at a tax policy colloquium presentation at the University of Minnesota Law School in February 2015.

                **   Ernest L. Wilkinson Chair and Professor of Law, J. Reuben Clark Law School, Brigham Young University.

            ***   Fondren Foundation Centennial Chair for Faculty Excellence and Professor of Law, The University of Texas School of Law. Professor Peroni dedicates this Article to his loving parents, Betty Peroni and the late Emil Peroni for their tremendous support and inspiration over the years. Professor Peroni also thanks the University of Texas School of Law for its generous research support.

        ****   Professor of Practice, Harvard Law School.

         [1].   See generally I.R.C. §§ 951–64 (2012) (detailing the anti-deferral rules relating to controlled foreign corporations).

         [2].   See, e.g., I.R.C. § 163(j) (2012) (containing earnings stripping deduction limitations only for interest payments and only if certain triggers are met, discussed infra text accompanying notes 206–27).

         [3].   See, e.g., J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay, Perspectives on the Worldwide vs. Territorial Taxation Debate, 125 Tax Notes 1079, 1081 (2009) [hereinafter Fleming, Peroni & Shay, Perspectives]; Robert J. Peroni, J. Clifton Fleming, Jr. & Stephen E. Shay, Getting Serious About Curtailing Deferral of U.S. Income Tax on Foreign Source Income, 52 SMU L. Rev. 455, 458 (1999) [hereinafter Peroni, Fleming & Shay, Getting Serious].

         [4].   The most basic MNEs are a parent/subsidiary structure or a pair of corporations whose stock is owned by the same party or parties (i.e., the brother/sister structure). There are virtually endless combinations and permutations of these basic patterns. For purposes of this Article, an MNE is a parent/subsidiary group of corporations that functions as an economic unit with at least one member being engaged in business activity outside the country in which the MNE parent is a resident for tax purposes.

         [5].   See Cong. Budget Office, Options for Taxing U.S. Multinational Corporations 20 (2013) [hereinafter CBO, Options]; Staff of J. Comm. on Taxation, JCX-42-11, Present Law and Issues in U.S. Taxation of Cross-Border Income 103 (2011) [hereinafter J. Comm., Cross-Border Income]; Fleming, Peroni & Shay, Perspectives, supra note 3, at 1084–85.

         [6].   See CBO, Options, supra note 5, at 20; J. Comm., Cross-Border Income, supra note 5, at 103; J. Clifton Fleming, Jr. & Robert J. Peroni, Eviscerating the Foreign Tax Credit Limitations and Cutting the Repatriation Tax – What’s ETI Repeal Got to Do with It?, 104 Tax Notes 1393, 1406, 1414 (2004) [hereinafter Fleming & Peroni, Eviscerating].

         [7].   See Fleming, Peroni & Shay, Perspectives, supra note 3, at 1091–1101; J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay, Fairness in International Taxation: The Ability-to-Pay Case for Taxing Worldwide Income, 5 Fla. Tax Rev. 299, 328–33 (2001) [hereinafter Fleming, Peroni & Shay, Fairness]; see also Ajay K. Mehrotra, Making the Modern American Fiscal State 8–15 (2013) (tracing the rise of the ability-to-pay concept as a fundamental pillar of taxation in the United States).

         [8].   A residual tax is the U.S. income tax liability that exceeds any allowable credit for foreign income tax payments on foreign-source income.

         [9].   See, e.g., Mihir A. Desai & James R. Hines, Jr., Old Rules and New Realities: Corporate Tax Policy in a Global Setting, 75 Nat’l Tax J. 937, 957 (2004); Business Roundtable & Business Council, Policy Burdens Inhibiting Economic Growth 17–18 (June 21, 2010), available at http://businessroundtable.org/sites/ default/files/20100621_Letter_to_OMB_Director_Orszag_from_BRT_and_BC_with_Attachments.pdf; J.D. Foster & Curtis S. Dubay, Obama International Tax Plan Would Weaken Global Competitiveness, Heritage Found. (May 5, 2009), http://www.heritage.org/Research/Reports/2009/05/Obama-International-Tax-Plan-Would-Weaken-Global-Competitiveness; see also J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay, Designing a U.S. Exemption System for Foreign Income When the Treasury is Empty, 13 Fla. Tax Rev. 397, 426–27 (2012) [hereinafter Fleming, Peroni & Shay, Designing] (arguing that if the U.S. international income system is reworked into an exemption or territorial system, it should be designed to protect revenue); Daniel N. Shaviro, Evaluating the Case for 1986-Style Corporate Tax Reform, 145 Tax Notes 1267, 1274 (2014) (“[A]t present there is ‘no evidence’ to support assertions that U.S. companies generally face higher effective tax rates, whether by reason of the higher statutory rate or having a nominally worldwide system for taxing foreign source income.”).

       [10].   See Eric Solomon, Corporate Inversions: A Symptom of Larger Tax System Problems, 136 Tax Notes 1449, 1449 (2012).

       [11].   See I.R.C. § 7874 (2012); Martin A. Sullivan, Lessons from the Last War on Inversions, 143 Tax Notes 861, 861 (2014); Bret Wells, Corporate Inversions and Whack-a-Mole Tax Policy, 143 Tax Notes 1429, 1429–32 (2014) [hereinafter Wells, Whack-a-Mole].

       [12].   See Edward D. Kleinbard, The Lessons of Stateless Income, 65 Tax L. Rev. 99, 160–61 (2011) [hereinafter Kleinbard, Lessons]; Lee A. Sheppard, Inversions Continue, but Are More Difficult, 141 Tax Notes 355, 355 (2013) [hereinafter Sheppard, Inversions Continue].

       [13].   See Donald J. Marples & Jane G. Gravelle, Cong. Research Serv., Corporate Expatriation, Inversions, and Mergers: Tax Issues 10–11 (2014); Mindy Herzfeld, What’s Next in Inversion Land?, 143 Tax Notes 1225, 1226–27 (2014); Sheppard, Inversions Continue, supra note 12, at 355; Wells, Whack-a-Mole, supra note 11, at 1432.

       [14].   See U.S. Treasury Dep’t, General Explanations of the Administration’s Fiscal Year 2016 Revenue Proposals 37–38 (2015) [hereinafter U.S. Treas. Dep’t, General Explanations]; Lindsey McPherson & Andrew Velarde, Democrats Disagree on Legislative Fix for Inversions, 143 Tax Notes 876, 876–77 (2014); Stephen E. Shay, Mr. Secretary, Take the Tax Juice Out of Corporate Expatriations, 144 Tax Notes 473, 479 (2014) [hereinafter Shay, Mr. Secretary]; Martin A. Sullivan, Short-Term Inversion Fix May Be Necessary, 143 Tax Notes 1090, 1092 (2014); Andrew Velarde & Lindsey McPherson, Schumer Seeks Interest Deduction Limit for Inverters, 144 Tax Notes 379, 379–80 (2014).

       [15].   See U.S. Treasury Dep’t, Earnings Stripping, Transfer Pricing and U.S. Income Tax Treaties 21–22 (2007) [hereinafter U.S. Treas. Dep’t, Earnings Stripping]; Dana Mattioli, Acquirers Plot Escape from a Turn on Taxes, Wall St. J., July 7, 2014, at C1, C2 (quoting Kevin Rinker, a partner at Debevoise & Plimpton LLP, as saying, “Because the [tax] benefit the buyer will get from an inversion is so important to a deal, if there were a change in [tax] law, the original price wouldn’t make sense for the acquirer”). As stated by Willard Taylor:

Inversions are not, and never were, solely about voting-with-your feet, or electing into, an exemption system of taxation for foreign business income. . . . A dollar earned in the United States is taxed at a 35% rate but if paid as interest to the new Irish parent is taxed at a 12.5% rate at most (and is not subject to withholding under the U.S.-Ireland Treaty). This has little to do with whether or not the United States adopts an exemption or territorial system . . . .

Willard B. Taylor, Letter to the Editor, A Comment on Eric Solomon’s Article on Corporate Inversions, 137 Tax Notes 105, 105 (2012). This Article only addresses earnings stripping in a cross-border context. Earnings stripping techniques also may be used in connection with tax-exempt taxpayers, taxpayers with carryover losses, and in other contexts, each of which would require separate technical and policy analyses that are beyond the scope of this Article.

       [16].   See J. Clifton Fleming, Jr., Robert J. Peroni & Stephen E. Shay, Formulary Apportionment in the U.S. International Income Tax System: Putting Lipstick on a Pig?, 36 Mich. J. Int’l Law 1, 7 (2014) [hereinafter Fleming, Peroni & Shay, Formulary Apportionment]; Fleming, Peroni & Shay, Perspectives, supra note 3, at 1098–1100.

       [17].   See I.R.C. §§ 11(d), 881(a), 882(a), 864(c) (2012). Foreign corporations are subject to U.S. income tax on certain specified items of foreign-source income that are effectively connected with a trade or business conducted within the United States. See id. §§ 864(c)(4), 882(a). As discussed in the text, a U.S. MNE parent of a foreign corporation that is a controlled foreign corporation is taxed currently on most interest, dividends, and royalties stripped from the United States under the foreign personal holding company rules of Subpart F of the Code. See id. §§ 951–52, 954, 957–58. Under the Foreign Account Tax Compliance Act (“FATCA”) rules, certain nonparticipating foreign financial institutions and nonfinancial foreign entities that fail to comply with specified reporting obligations will be subject to withholding on withholdable payments without regard to the technical geographical source of the income. See id. §§ 1471–1474.

       [18].   Other strategies include aggressive transfer pricing, loss importation, and “double dip” financing. These strategies have been targeted with mixed success by several Internal Revenue Code provisions. See id. §§ 334(b)(1), 362(e)(1), 482, 1503(d). We do not consider in this Article the effect on earnings stripping of structural reforms to the corporate income tax. Recent proposals would (i) shift the corporate tax to shareholders, see, e.g., Eric Toder & Alan D. Viard, Major Surgery Needed: A Call for Structural Reform of the U.S. Corporate Income Tax 1 (2014) (report funded by the Peter G. Peterson Foundation), available at http://www.aei.org/publication/major-surgery-needed-a-call-for-structural-reform-of-the-us-corporate-income-tax/; (ii) convert the corporate tax into a form of cash flow consumption tax, see, e.g., Alan J. Auerbach, A Modern Corporate Tax 9 (2010), available at http://www.hamiltonproject.org/ papers/a_modern_corporate_tax/; and (iii) adopt a cost of capital deduction and shareholder inclusion in lieu of an interest deduction, see, e.g., Edward D. Kleinbard, Rehabilitating the Business Income Tax 5 (2007), available at http://www.hamiltonproject.org/papers/rehabilitating_the_business_income_tax1/. Each of these proposals remains a work in progress and requires further specification before being in a form capable of adoption as legislation. Consideration of whether earnings stripping remedies would be necessary in the event of a structural reform of the corporate income tax can wait until such a proposal has some serious prospect of being adopted.

       [19].   See I.R.C. §§ 162(a), 163(a); see also J. Comm., Cross-Border Income, supra note 5, at 54 (“A domestic corporation with a foreign parent may reduce the U.S. tax on the income derived from its U.S. operations through the payment of deductible amounts such as interest, rents, royalties, premiums, and management services fees to the foreign parent or other foreign affiliates that are not subject to U.S. tax on the receipt of such payments. Generating excessively large U.S. tax deductions in this manner is known as ‘earnings stripping.’ ”); U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 7 (“Although payments of other deductible amounts by a U.S. corporation to tax-exempt or partially exempt related parties also provide an opportunity to shift income out of a U.S. corporation, the use of related-party debt arguably is the most readily available method for shifting income out of U.S. corporations.”); Taylor, supra note 15, at 106 (stating that proposals to deal with earnings stripping “should not be limited to interest but should also consider reinsurance premiums . . . and the treatment of payments to related foreign persons for the use of intangible or other property and for off-shore services”). Note that because USSub makes actual payments to UKCo, § 267(a)(3) does not prevent USSub from deducting the interest payments. See Treas. Reg. § 1.267(a)-3(b)(4), ex.1 (1993).

       [20].   UKCo has no U.S. taxable income from its sales to USSub because the sales are not made through a U.S. permanent establishment. See Convention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income and on Capital Gains, U.S.-U.K., arts. 5, 7, July 24, 2001, T.I.A.S. No. 13,161 [hereinafter U.S.-U.K. Treaty]. We assume for purposes of this discussion that UKCo is a “qualified person” and is entitled to all of the benefits of the United States-United Kingdom Income Tax Treaty. See id. at art. 23.

       [21].   See I.R.C. §§ 862(a)(3), 882(a) (2012).

       [22].   See U.S.-U.K. Treaty, supra note 20, at arts. 11, 12. U.S. income tax treaties typically follow this pattern. See United States Model Income Tax Convention of November 15, 2006, arts. 11, 12 [hereinafter 2006 U.S. Model], available at www.treasury.gov/resource-center/tax-policy/treaties/Documents/model006.pdf.

       [23].   I.R.C. § 954(a), (c); see Charles H. Gustafson, Robert J. Peroni & Richard Crawford Pugh, Taxation of International Transactions 496 (4th ed. 2011). For a detailed discussion of the Subpart F provisions of the Code, see 1 Joel D. Kuntz & Robert J. Peroni, U.S. International Taxation ch. B3 (1992 & Cum. Supp. No. 3 2014).

       [24].   See Shay, Mr. Secretary, supra note 14, at 474; Taylor, supra note 15, at 105.

       [25].   Strong earnings stripping rules interfere with the tactic of a U.S. corporation transforming itself into a foreign corporation with a U.S. subsidiary that makes deductible payments of interest, services fees, and royalties to the foreign corporation that reduce the subsidiary’s U.S. tax base. By limiting these deductible payments and protecting the U.S. subsidiary’s U.S. tax base, properly designed earnings stripping rules reduce the tax savings to be achieved in the corporate expatriation transaction.

       [26].   See Committee of Public Accounts, HM Revenue & Customs: Annual Report and Accounts 2011–12, 2012–13, H.C. 716, ¶ 8, Ev 21–Ev 50 (U.K.) (containing oral evidence taken from Troy Alstead, Starbucks Global Chief Financial Officer; Andrew Cecil, Amazon Director of Public Policy; and Matt Brittin, Google Vice President for Sales and Operations, Northern and Central Europe, on Nov. 12, 2012); Lee A. Shepperd, Luxembourg Lubricates Income Stripping, 145 Tax Notes 1071, 1071–72 (2014); Tom Bergin, Burger King Has Maneuvered to Cut U.S. Tax Bill for Years, Reuters (Sept. 2, 2014), http://uk.reuters.com/article/2014/09/02/uk-burger-king-tax-insight-idUKKBN0GX0AK20140902?feedType=RSS&feedName=GCA-GoogleNewsUK (reporting that Burger King had pre-tax losses in Germany for two years on over $500 million in sales after payments of 5% of the sales turnover to Swiss affiliate); Leslie Wayne, Kelly Carr, Marina Walker Guevara, Mar Cabra & Michael Hudson, Leaked Documents Expose Global Companies’ Secret Tax Deals in Luxembourg, Int’l Consortium of Investigative Journalists (Nov. 5, 2014, 4:00 PM), http://www.icij.org/project/luxembourg-leaks/leaked-documents-expose-global-companies-secret-tax-deals-luxembourg (disclosing Luxembourg tax rulings for U.S. and non-U.S. MNE structures that shifted profits through earnings stripping and other tax-saving maneuvers in Luxembourg that were authorized under the rulings).

       [27].   In contrast to the tax system designer/tax policy analyst, the tax planner seeks out weak margins in order to exploit them to avoid taxes. The idea of weak margins may be attributed in part to Harry Grubert, who emphasizes aligning design of policy instruments with the behavioral margins to be affected. See Harry Grubert, Tax Credits, Source Rules, Trade, and Electronic Commerce: Behavioral Margins and the Design of International Tax Systems, 58 Tax L. Rev. 149, 150 (2005).

       [28].   In other work, we have suggested directions for strengthening the definition of corporate residence, see Fleming, Peroni & Shay, Formulary Apportionment, supra note 16, at 23, and one of us has very preliminarily explored how to reduce the significance of corporate residence in the taxation of shareholders, see Stephen E. Shay, Theory, Complications, and Policy: Daniel Shaviro’s Fixing U.S. International Taxation, 9 Jerusalem Rev. Legal Stud. 104, 106 (2014). Other commentators have explored alternative residence tests. See, e.g., Omri Marian, Jurisdiction to Tax Corporations, 54 B.C. L. Rev. 1613, 1618 (2013) (considering basing U.S. residency on U.S. public listing or central management and control in the United States); George K. Yin, Letter to the Editor, Stopping Corporate Inversions Sensibly and Legally, 144 Tax Notes 1087, 1087 (2014) (suggesting that residence of an enterprise should be based on location of the enterprise’s principal customer base).

       [29].   See Stephen E. Shay, J. Clifton Fleming, Jr. & Robert J. Peroni, The David Tillinghast Lecture “What’s Source Got to Do with It?” Source Rules and U.S. International Taxation, 56 Tax L. Rev. 81, 83 (2002) [hereinafter Shay, Fleming & Peroni, What’s Source Got to Do with It?].

       [30].   See Fleming, Peroni & Shay, Formulary Apportionment, supra note 16, at 3; Stephen E. Shay, Commentary on Ownership Neutrality and Practical Complications, 62 Tax L. Rev. 317, 327–30 (2009).

       [31].   Earnings stripping is also a major target of the G20’s efforts to fight base erosion and profit shifting, often referred to as the “BEPS project” or “BEPS plan.” See Org. for Econ. Cooperation & Dev., Action Plan on Base Erosion and Profit Shifting 15–17 (2013) [hereinafter OECD, Action Plan]; Yariv Brauner, What the BEPS?, 16 Fla. Tax Rev. 55, 88–91 (2014); Kristin A. Parillo, Amanda Athanasiou, Margaret Burow, Ajay Gupta, David D. Stewart & Stephanie Suong Johnston, OECD Releases BEPS Report Ahead of G-20 Meeting, 144 Tax Notes 1347, 1347 (2014).

       [32].   See Kleinbard, Lessons, supra note 12, at 138 (stating that “a territorial tax system means in practice that every country, including the residence country, is just another source country”).

       [33].   See CBO, Options, supra note 5, at 3; see also Wells, Whack-a-Mole, supra note 11, at 1433 (“The U.S. subpart F regime has served as a means to backstop the U.S. tax base when this skewed application of our transfer pricing and treaty rules creates inappropriate profit-shifting opportunities.”).

       [34].   See, e.g., William McBride, Letter to the Editor, Earnings Stripping and the Complexity of the Corporate Tax, 144 Tax Notes 1086, 1086 (2014); Scott A. Hodge, IRS Data Contradicts Kleinbard’s Warning of Earnings Stripping Inversions, Tax Pol’y Blog (Sept. 2, 2014), http://taxfoundation.org/blog/irs-data-contradicts-Kleinbard-s-warnings-earnings-stripping-inversions.

       [35].   See Edward D. Kleinbard, “Competitiveness” Has Nothing to Do With It, 144 Tax Notes 1055, 1056 (2014); Paul Caron, The Tax Foundation and Ed Kleinbard Debate the Impact of Inversions on the Corporate Tax Base, Tax Pol’y Blog (Sept. 7, 2014), http://taxprof.typepad.com/taxprof_blog/2014/09/tax-foundation.html (providing Kleinbard’s response to criticisms raised by Hodge, supra note 34).

       [36].   See Martin A. Sullivan, Will Tax Reform Stop Inversions?, 144 Tax Notes 530, 531 (2014).

       [37].   In 2015, the U.K. general corporate tax rate will be 20%. William McBride, Tax Reform in the UK Reversed the Tide of Corporate Tax Inversions, Tax Found. (Oct. 14, 2014), http://taxfoundation.org/article/tax-reform-uk-reversed-tide-corporate-tax-inversions.

       [38].   See Treas. Reg. § 1.482-1(e) (as amended in 2013).

       [39].   See id. § 1.482-1(e)(1).

       [40].   See Gustafson, Peroni & Pugh, supra note 23, at 721–57; 1 Kuntz & Peroni, supra note 23, ¶ A3.01[1].

       [41].   There are multiple forms of inversion transactions. For descriptions and explanations, see Marples & Gravelle, supra note 13, at 3–5; David L. Brumbaugh, Cong. Research Serv., RL31444, Firms That Incorporate Abroad for Tax Purposes: Corporate “Inversions” and “Expatriation” 3–4 (2007); Steven H. Goldman, Corporate Expatriation: A Case Analysis, 9 Fla. Tax Rev. 71, 73, 74–76 (2008); Solomon, supra note 10, at 1204–05; Steven M. Surdell, Inversions 2014—Self Help International Tax Reform for U.S. Multinationals?, 92 Taxes 63, 64–80 (2014); Bret Wells, Cant and the Inconvenient Truth About Corporate Inversions, 136 Tax Notes 429, 430–36 (2012) [hereinafter Wells, Inconvenient Truth]; Bret Wells, What Corporate Inversions Teach About International Tax Reform, 127 Tax Notes 1345, 1349–51 (2010) [hereinafter Wells, Corporate Inversions].

There is a large body of literature discussing corporate inversions and the legislative and administrative attempts to deal with them. See, e.g., Brumbaugh, supra, at 6, 9–14; Solomon, supra note 10, at 1206–10; Wells, Inconvenient Truth, supra, at 430–36; Wells, Corporate Inversions, supra, at 1349–51. Because this Article focuses on the earnings stripping problem, which does not require a corporate inversion, see generally J. Comm., Cross-Border Income, supra note 5, at 50; U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 9–10; Taylor, supra note 15, at 105, we do not discuss this literature or the details of the corporate inversion topic.

       [42].   See U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 21–23.

       [43].   The 2007 U.S. Treasury report on earnings stripping stated that “[t]he effect of earnings stripping on the U.S. tax base is unclear, because we cannot accurately quantify earnings stripping by [foreign controlled domestic corporations] generally.” Id. at 26. Examining a data set of companies that inverted before 2004, the report found clear evidence of earnings stripping of U.S. affiliates of foreign companies that inverted. Id. at 21. The fact that the effect of earnings stripping could not be precisely quantified does not, however, mean that the effect is insignificant. The OECD has concluded that the effect of earnings stripping on the tax bases of member countries is, indeed, significant. See OECD, Action Plan, supra note 31, at 16–17. An important item in the BEPS action plan is Item 11 (“Establish methodologies to collect and analyze data on BEPS and the actions to address it.”), which should lead to improvements in the empirical data on these issues. See id. at 21–22. See generally Edoardo Traversa, Interest Deductibility and the BEPS Action Plan: nihil novi sub sole?, 2013 Brit. Tax Rev. 607 (2013) (critiquing the portion of the Action Plan that deals with intra-group interest payments).

       [44].   If a U.S. subsidiary makes deductible interest, royalty, and fees payments to a U.S. parent corporation, the deductions are removed from the subsidiary’s taxable income, but they remain in the U.S. tax base as income items of the parent corporation. See I.R.C. § 61 (2012). If a U.S. corporation makes deductible interest, royalty, and fees payments to a foreign subsidiary of a U.S. parent corporation, the payments are likely to be included in the U.S. tax base as Subpart F income that is currently taxable to the U.S. parent corporation as a constructive inclusion. See id. §§ 951–952, 954, 957–958.

       [45].   See U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 23, 26; see also OECD, Action Plan, supra note 31, at 8 (“[C]orporations that operate only in domestic markets, including family-owned businesses or new innovative companies, have difficulty competing with MNEs that have the ability to shift their profits across borders to avoid or reduce tax.”).

       [46].   See Wells, Whack-a-Mole, supra note 11, at 1432.

       [47].   See generally Kimberly A. Clausing, Should Tax Policy Target Multinational Firm Headquarters?, 63 Nat’l Tax J. 741, 761 (2010) (discussing several disadvantages to these types of tax incentives).

       [48].   The significantly limited coverage of § 163(j) and the substantial avoidance opportunities that result are discussed infra Part III.B.

       [49].   Indeed, in many cases it will be unclear whether the optimal response to a tax rate change by one country is for another country to increase or decrease its tax rate. As stated by two commentators:

Intuition might suggest, in particular, that the best response to a reduction in some other country’s tax rate will be for [country] i to reduce its own rate too; meaning that tax rates are strategic complements. But this is not, in general, assured (even in the case of symmetric countries). A lower tax rate in some other country j, for instance, moves capital out of country i, and so reduces its tax revenue and public spending; whether the best response to this is for i to raise or lower its tax rate depends, among other things, on how large an increase in the marginal value of public spending this implies (being more likely the greater is that increase).

Michael Keen & Kai Konrad, The Theory of International Tax Competition and Coordination, in Handbook of Public Economics 258, 267–68 (2013). A cynic might argue that if a country does not know the correct sign of the response, tax competition either is truly stupid or is a consequence of capture by interested parties. See Int’l Monetary Fund, Spillovers in Int’l Corporate Taxation 21–23 (2014) [hereinafter IMF, Spillovers] (stating that alleged benefits of tax competition “generate heated dispute, but remain essentially uninformed by empirical knowledge”); see also Thushyanthan Baskaran & Mariana Lopes de Fonseca, The Economics and Empirics of Tax Competition: A Survey and Lessons for the EU, 7 Erasmus L. Rev. 3, 3–12 (2014) (surveying the economics literature on tax competition and concluding that it is ambiguous regarding whether or not tax competition causes a race to the bottom).

       [50].   This is evidenced at present by the very low U.S. interest rates reported regularly in the financial press, see, e.g., Biggest 1,000 Stocks, Wall St. J., Oct. 15, 2014, at C6; Ben Edwards, Apple’s Euro Bond Sells at Record-Low Rate, Wall St. J., Nov. 5, 2014, at C8, and by frequent financial press stories regarding the large cash accumulations of U.S. corporations, see, e.g., Theo Francis & Ted Mann, An Irony of U.S. Policy: Record Cash, Record Debt, Wall St. J., Aug. 1, 2014, at B1; Vipal Monga, New Rules, New Puzzle: Where to Stash the Cash?, Wall St. J., Nov. 4, 2014, at B1. These low interest rates also facilitate funding of the federal budget deficit. If interest rates increase during a period of continued U.S. budget deficits, the negative impact of the revenue loss would be exacerbated.

       [51].   See Harry G. Brainard, International Economics and Public Policy 93–103 (1954); Yariv Brauner, International Tax and Trade Agreements May Be Coordinated, but Not Reconciled, 25 Va. Tax Rev. 251, 253 (2005).

       [52].   See Shay, Fleming & Peroni, What’s Source Got to Do with It?, supra note 29, at 88–106. Professor Daniel Shaviro has argued that purely as a prudential matter, “zero is likely to be the optimal U.S. tax rate for truly inbound investment earning a merely normal rate of return which the [foreign] investors could and would match elsewhere.” Shaviro, supra note 9, at 1273. This statement, however, assumes that the relevant foreign countries tax their residents’ U.S.-source income at a positive rate without allowing an adequate credit for U.S. income tax. See id. In reality, the major trading partners of the United States generally address U.S.-source business income by either exempting it or taxing it with a full foreign tax credit for U.S. tax, and they generally deal with U.S.-source investment income by taxing it and allowing a full foreign tax credit for U.S. tax. Moreover, Professor Shaviro’s quoted statement does not challenge the normative case for U.S. taxation of U.S.-source income earned by foreign persons.

       [53].   See Restatement (Third) of the Foreign Relations Law of the United States §§ 411–12 (1986); Reuven S. Avi-Yonah, International Tax as International Law: An Analysis of the International Tax Regime 27 (2007) [hereinafter Avi-Yonah, International Tax as International Law]; Kim Brooks, Global Distributive Justice: The Potential for a Feminist Analysis of International Tax Revenue Allocation, 21 Can. J. Women & L. 267, 280 (2009); Michael S. Kirsch, The Role of Physical Presence in the Taxation of Cross-Border Personal Services, 51 B.C. L. Rev. 993, 999 (2010). For a discussion of the connection between source-based taxation and the international law concept of sovereignty, see Diane M. Ring, What’s at Stake in the Sovereignty Debate?: International Tax and the Nation-State, 49 Va. J. Int’l L. 155, 157–233 (2008).

       [54].   U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 7.

       [55].   Id. at 21; see also Jim A. Seida & William F. Wempe, Effective Tax Rate Changes and Earnings Stripping Following Corporate Inversion, 57 Nat’l Tax J. 805, 825 (2004) (finding that “a typical inversion firm experiences an 11.6 percentage point reduction in its effective tax rate . . . [and that a] large portion of the reduction is attributable to the stripping of U.S. earnings via intercompany interest payments”).

       [56].   U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 26.

       [57].   Id. at 29, 31.

       [58].   Id. at 7.

       [59].   Id. at 29; see also Staff of J. Comm. on Taxation, 111st Cong., Present Law and Background Related to Possible Income Shifting and Transfer Pricing, JCX-37-2010, at 108–09 (Comm. Print 2010) [hereinafter Joint Comm., Income Shifting and Transfer Pricing].

       [60].   See Harry Grubert, Foreign Taxes and the Growing Share of Multinational Company Income Abroad: Profits, Not Sales, Are Being Globalized, 65 Nat’l Tax J. 247, 274 (2012); Martin A. Sullivan, Pfizer’s Tax Picture Dominated by U.S. Losses, Repatriation, 140 Tax Notes 103, 105–10 (2013); Martin A. Sullivan, Transfer Pricing Abuse Is Job-Killing Corporate Welfare, 128 Tax Notes 461, 461 (2010); Martin A. Sullivan, U.S. Multinationals Paying Less Foreign Tax, 118 Tax Notes 1177, 1177–83 (2008); Martin A. Sullivan, U.S. Multinationals Shifting Profits Out of the United States, 118 Tax Notes 1078, 1078 (2008); see also U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 55–61; Transfer Pricing Issues: Hearing Before the H. Comm. on Ways & Means, 111th Cong. 1–6 (2010) (statement of Stephen E. Shay, Deputy Assistant Secretary for Int’l Tax Affairs, U.S. Treasury Dep’t) (updating data reported in 2007 study and finding substantial income shifting to lower-tax countries, including evidence from company tax data of margin increases correlated inversely with effective tax rates).

       [61].   We agree with the following analysis by the Staff of the Joint Committee on Taxation:

[S]ome argue that, as a matter of tax policy, the earnings stripping rules should be strengthened for all foreign-controlled domestic corporations (including expatriated entities) because they all have the same incentives and capabilities to erode the U.S. tax base, and may do so in the same manner. . . . Proponents of stricter across-the-board earnings stripping rules also argue that there is sufficient evidence of earnings stripping to justify implementing such a regime, and that significant erosion of the U.S. tax base will continue until earnings stripping rules are strengthened for all foreign-controlled domestic corporations.

  1. Comm., Cross-Border Income, supra note 5, at 61.

       [62].   NA Gen. P’ship & Subsidiaries v. Comm’r, 103 T.C.M. (CCH) 1916 (2012).

       [63].   See id. at 1917.

       [64].   See id. at 1919, 1924.

       [65].   See Stipulation, GlaxoSmithKline Holdings (Americas) Inc. v. Comm’r., Nos. 18940-08, 18941-08 (T.C. Nov. 19, 2009).

       [66].   For example, reportedly, 20% to 30% of all large cases open at the end of 2012 involved a debt-equity classification issue. Andrew Velarde, Recent IRS Litigation Presents Opportunities for Taxpayers in Debt-Equity Cases, Tax Notes Today, Nov. 15, 2013, available at Lexis, 2013 TNT 221-4.

       [67].   See, e.g., Sven-Olof Lodin, Intragroup Royalties As a Vehicle for International Tax Arbitrage, 71 Tax Notes Int’l 1317, 1318 (2013) [hereinafter Lodin II] (“Most countries have introduced limitations of the total interest cost deductible. . . . This reduction of the room for using interest payments for income shifting has meant that other income-shifting methods are expected to be used more frequently. Royalty payments have become the fastest-growing substitute method for profit shifting.”); Taylor, supra note 15, at 106 (“ ‘Earnings stripping’ in this context should not be limited to interest but should also consider reinsurance premiums . . . and the treatment of payments to related foreign persons for the use of intangible or other property and for off-shore services.”).

       [68].   See I.R.C. § 163(j)(1)-(3) (2012).

       [69].   A body of case law allows the IRS to re-characterize nominal corporate debt as an equity investment and to deny any deduction for interest payments on such debt. See, e.g., Philip G. Cohen, Testing for Thin Capitalization Under Section 163(j): A Flawed Safe Harbor, 67 Tax Law. 67, 77–81 (2013); David C. Garlock & Amin N. Khalaf, Debt vs. Equity: Myths, Best Practices and Practical Considerations for U.S. Tax Aspects of Related-Party Financings, Taxes, Aug. 2014, at 35, 36–49; William T. Plumb, Jr., The Federal Income Tax Significance of Corporate Debt: A Critical Analysis and a Proposal, 26 Tax L. Rev. 369, 370 (1971). However, when the formalities of debt are observed between the parties, this case law has allowed highly leveraged debtor corporations to deduct interest on related party debt. See Baker Commodities, Inc. v. Comm’r, 48 T.C. 374, 401 (1967) (upholding debtor corporation’s interest deductions in spite of 700:1 debt-equity ratio), aff’d, 415 F.2d 519 (9th Cir. 1969), cert. denied, 397 U.S. 988 (1970). This case law has not been effective in restraining earnings stripping. See generally NA Gen. P’ship & Subsidiaries v. Comm’r, 103 T.C.M 1916 (2012).

       [70].   See I.R.C. § 163(j)(1)(B).

       [71].   See id. § 163(j)(1), (2)(A)–(B).

       [72].   See id. § 163(j)(1)(A).

       [73].   See id. § 163(j)(3)(A).

       [74].   The general background rule of the Internal Revenue Code is that capital expenditures are not currently deductible unless an exception applies. See id. § 263(a)(1); Joseph M. Dodge, J. Clifton Fleming, Jr. & Robert J. Peroni, Federal Income Tax: Doctrine, Structure, and Policy 43–48 (4th ed. 2012). Thus, in the remainder of this Article, all expenditures identified as deductible are assumed to be current expenses and not capital expenditures.

       [75].   See I.R.C. § 63 (defining taxable income as net income). See generally id. § 1 (using the taxable income of individuals to determine tax levy due); id. § 11(a) (using the taxable income of corporations to determine tax levy due).

       [76].   See H. David Rosenbloom, Banes of an Income Tax: Legal Fictions, Elections, Hypothetical Determinations and Related-Party Debt, 32 Tax Notes Int’l 989, 997–98 (2003) (“[I]nterest is generally deductible in the unrelated party context because it is a legitimate cost of the funds needed to operate a business.”); Jonathan Talisman, Do No Harm: Keep Corporate Interest Fully Deductible, 141 Tax Notes 211, 211–12 (2013) [hereinafter Talisman, Do No Harm]. See generally Am. Law Inst., Federal Income Tax Project: Reporter’s Study Draft: Subchapter C (Supplemental Study) 80–83 (June 1, 1989) [hereinafter ALI, Supplemental Study] (explaining the need for a corporate interest deduction with respect to unrelated party debt).

       [77].   See IMF, Spillovers, supra note 49, at 30 (explaining that intra-MNE interest payments are a method for shifting profits to low-tax countries and concluding that “[i]t is reasonable to ask why, in principle, any deduction at all should be given for interest paid to related parties”). As stated by H. David Rosenbloom:

[R]elated-party debt [interest] is generally not compensation for money lent from one person to another. Rather it is a transfer of funds from one incorporated pocket to another. . . . As noted, interest is generally deductible in the unrelated-party context because it is a legitimate cost of the funds needed to operate a business. The application of this rationale in a related-party context, where no funds are being raised, is one of the tax miracles of our time.

Rosenbloom, supra note 76, at 997–98; see also Johanna Hey, Base Erosion and Profit Shifting and Interest Expenditure, 68 Bull. for Int’l Tax’n 332, 342 (2014).

       [78].   To hold otherwise would force USSub to borrow directly from the third-party lender and forgo the possibility that even if UKCo guaranteed USSub’s debts, UKCo could borrow at a lower cost than USSub because of the nuances of applicable debtor-creditor law, UKCo’s greater size, and other factors. There is no apparent reason for the income tax law to force subsidiaries like USSub into direct borrowing from third parties when nontax market factors make it more efficient to have the borrowing done by a foreign parent. As stated by Professor Kleinbard: “Capital markets ordinarily prefer parent-level financing, because all of the group’s operations then support the loan, and because the agency costs associated with policing parent-subsidiary transfer pricing and transactions are irrelevant.” Kleinbard, Lessons, supra note 12, at 163; see also Hey, supra note 77, at 342. If, however, UKCo maintained a corresponding deposit with its lender, UKCo’s real expense would be its net cost of borrowing. These and other avoidance schemes will be considered in formulating workable implementing rules, discussed infra Part III.

       [79].   See I.R.C. §§ 1, 11, 63 (2012).

       [80].   See ALI, Supplemental Study, supra note 76, at 80–83.

       [81].   This is also the approach generally taken with respect to payments between a corporation’s branches and its headquarters. See Am. Law Inst., Federal Income Tax Project: International Aspects of United States Income Taxation: Proposals on United States Taxation of Foreign Persons and of the Foreign Income of United States Persons 121 (1987) [hereinafter ALI, International Tax Proposals].

       [82].   This is an expansion of an approach advocated by H. David Rosenbloom. See Rosenbloom, supra note 76, at 999–1000 (“[I]n the context of an ostensible debt owed by one such [related corporate] party to another, it strains credulity to say that ‘repayment absolutely and in all events’ is envisioned or provided for. Two pieces of paper owned ultimately by a single economic interest cannot do business together no matter what legal sophists say.”).

       [83].   See Daniel N. Shaviro, Decoding the U.S. Corporate Tax 10–14 (2009) (explaining the administrative advantages of collecting income tax from corporations instead of from numerous shareholders); Charles I. Kingson, Can a Piece of Paper Earn Billions?, 71 Tax Notes Int’l 243, 243 (2013) (stating that a “corporation is essentially a piece of legal paper”); Rosenbloom, supra note 76, at 992 (stating that “[b]ecause corporations are a convenient mechanism for the collection of tax, the U.S. system considers them income tax payers”); see also U.S. Treasury Dep’t, Integration of the Individual and Corporate Tax Systems: Taxing Business Income Once 29 (1992) (stating that “tax is more likely to be collected if paid at the corporate level”).

       [84].   See Rosenbloom, supra note 76, at 992 (“[T]here does not appear to be a fundamental inconsistency between collecting tax from the corporation and according it something less than full ‘person-hood’ equivalent to individuals.”); see also George K. Yin & David J. Shakow, Am. Law Inst., Federal Income Tax Project: Taxation of Private Business Enterprises—Reporters’ Study 13–15 (1999) (proposing a mandatory system of conduit taxation that would include taxing the owners of closely held corporations directly on their shares of corporate income regardless of whether the income is actually distributed); Alvin C. Warren, Jr., Am. Law Inst., Federal Income Tax Project: Integration of the Individual and Corporate Income Taxes—Reporter’s Study of Corporate Tax Integration 1 (1993) (recommending that corporate income be subjected to a shareholder tax and that the corporate income tax be converted into a withholding mechanism for the shareholder tax). As stated by another commentator:

[T]here is no general principle under international public law that countries have to simply accept the legal existence of a domestic or foreign corporate entity as a “shield” against any look-through approach from either side. Domestic legislation (including domestic legislation in the country of the parent company) decides which entities are “transparent” for tax purposes and which are not. Secondly, it cannot be denied that the shareholders of the parent company are the final recipients of profits earned by a subsidiary, even if these profits are not yet repatriated for the time being. Thus, it depends only on good policy reasons whether a look-through approach should be adopted, e.g. in the context of an integration system for individual and corporate income tax applied by the host country or in the context of anti-deferral measures (CFC legislation) established by the home country.

Wolfgang Schön, International Tax Coordination for a Second-Best World (Part I), World Tax J. Sept. 2009, at 67, 92 [hereinafter Schön, Part I].

       [85].   See Edward D. Kleinbard, Stateless Income, 11 Fla. Tax Rev. 699, 709 (2011) [hereinafter Kleinbard, Stateless Income] (describing as “fantastic” the notion that “a wholly-owned subsidiary has a mind of its own with which to negotiate ‘arm’s-length’ contractual terms with its parent”). Of course, this depletion of the tax base is allowed in the domestic sphere where the related payee is a tax-exempt organization or a loss corporation. This, however, raises tax policy and tax theory issues outside the scope of this Article.

       [86].   See Staff of J. Comm. on Tax’n, Present Law and Background Relating to the Interaction of Federal Income Tax Rules and Financial Accounting Rules 31–32 (2012); Jasper L. Cummings, Jr., Consolidating Foreign Affiliates, 11 Fla. Tax Rev. 143, 154–56 (2011).

       [87].   See Sol Picciotto, Tax Justice Network, Towards Unitary Taxation of Transnational Corporations 10 (2012), available at www.taxjustice.net/cms/upload/pdf/Towards_Unitary_Taxation_1-1.pdf; Rosanne Altshuler & Harry Grubert, Formula Apportionment: Is It Better Than the Current System and Are There Better Alternatives?, 63 Nat’l Tax J. 1145, 1145 (2010); Reuven S. Avi-Yonah & Ilan Benshalom, Formulary Apportionment—Myths and Prospects, World Tax J., Oct. 2011, at 371, 380; Reuven S. Avi-Yonah, Kimberly A. Clausing & Michael C. Durst, Allocating Business Profits for Tax Purposes: A Proposal to Adopt a Formulary Profit Split, 9 Fla. Tax Rev. 497, 508 (2009); Lorence L. Bravenec, Connecting the Dots in U.S. International Taxation, 27 Tax Notes Int’l 845, 848–49 (2002); Cummings, supra note 86, at 148–49; Brian Lebowitz, Profit Sharing as a New World Order in International Taxation, 52 Tax Notes Int’l 585, 589–90 (2008); Julie Roin, Can the Income Tax Be Saved? The Promises and Pitfalls of Adopting Worldwide Formulary Apportionment, 61 Tax L. Rev. 169, 214 (2008); H. David Rosenbloom, Angels on a Pin: Arm’s Length in the World, 38 Tax Notes Int’l 523, 523 (2005); Lee A. Sheppard, Is Transfer Pricing Worth Salvaging?, 136 Tax Notes 467, 469 (2012).

       [88].   See Gustafson, Peroni & Pugh, supra note 23, at 367–71.

       [89].   See Picciotto, supra note 87, at 10; Daniel N. Shaviro, Fixing U.S. International Taxation 197 (2014); Avi-Yonah, Clausing & Durst, supra note 87, at 501; Cummings, supra note 86, at 148–49; Kleinbard, Stateless Income, supra note 85, at 709 (rejecting the view that “a multinational enterprise that exists as a global platform to exploit a core set of intangible assets best is analogized to wholly independent actors taking on limited and straightforward roles in a vertical chain of production or a horizontal array of distribution of a product”); Rosenbloom, supra note 76, at 992 (“The economic justification for regarding a [related] group of 10 corporations as 10 separate persons is hardly compelling. There may well be nontax justifications for the creation of the group, but it is not foreordained that tax laws should respect the results.”). See generally J. Comm., Income Shifting and Transfer Pricing, supra note 59, at 12–17 (describing intercompany transactions and their income tax consequences); Barbara Angus, Tom Neubig, Eric Solomon & Mark Weinberger, The U.S. International Tax System at a Crossroads, 127 Tax Notes 45, 50–53 (2010) (discussing the rise of global companies and the increased importance of their interconnectivity).

       [90].   See Treas. Reg. § 1.482-1(b) (2013).

       [91].   See Moline Props. v. Comm’r, 319 U.S. 436, 438–39 (1943); see also, e.g., Org. for Econ. Cooperation & Dev., Guidance on Transfer Pricing Aspects of Intangibles 40–42 (2014) [hereinafter OECD, Guidance on Intangibles] (“Legal ownership and contractual relationships serve simply as reference points for identifying and analysing controlled transactions relating to the intangible and determining the appropriate remuneration to members of the controlled group with respect to the transactions.”).

       [92].   See, e.g., OECD, Action Plan, supra note 31, at 14 (disfavoring formulary apportionment as a replacement for the arm’s-length approach to transfer pricing because “it is . . . unclear that the behavioral changes companies might adopt in response to the use of a formula would lead to investment decisions that are more efficient and tax-neutral than under a separate entity approach”); id. at 19–20 (arguing that “[i]n many instances, the existing transfer pricing rules, based on the arm’s length principle, effectively and efficiently allocate the income of multinationals among taxing jurisdictions” and “rather than seeking to replace the current transfer pricing system, the best course is to directly address the flaws in the current system”).

       [93].   See Shaviro, supra note 89, at 39–40; Fleming, Peroni & Shay, Formulary Apportionment, supra note 16, at 18–20; Rosenbloom, supra note 76, at 995–96. See generally Stanley I. Langbein, The Unitary Method and the Myth of Arm’s Length, 30 Tax Notes 625 (1986) (criticizing the arm’s-length standard).

       [94].   See Moline Props., 319 U.S. at 438–39; Bausch & Lomb, Inc. v. Comm’r, 92 T.C. 525, 525 (1989), aff’d, 933 F.2d 1084 (2d Cir. 1991).

       [95].   See, e.g., Estate of Paxton v. Comm’r, 44 T.C.M. (CCH) 771, 771 (1982); Lausman v. Comm’r, 37 T.C.M. (CCH) 1740, 1743 (1978).

       [96].   See supra text accompanying notes 19–22.

       [97].   See IRS, Withholding of Tax on Nonresident Aliens and Foreign Entities: Publication 515, at 59–60 tbl.3 (2013).

       [98].   2006 U.S. Model, supra note 22.

       [99].   See Daniel M. Berman, Covering the World: The Expanding U.S. Tax Treaty Network, 74 Taxes 1064, 1071–83 (1996).

     [100].   1981 U.S. Model Income and Capital Tax Convention [hereinafter 1981 U.S. Model].

     [101].   1977 U.S. Model Income and Capital Tax Convention [hereinafter 1977 U.S. Model].

     [102].   Compare 2006 U.S. Model, supra note 22, at arts. 9, 11, 12, 24, with Berman, supra note 99, at 1075–77, 1081, 1981 U.S. Model, supra note 100, at arts. 9, 11, 12, 24, and 1977 U.S. Model, supra note 101, at arts. 9, 11, 12, 24.

     [103].   Indeed, in the treaties between the United States and Ireland and the United States and the United Kingdom, two of the most attractive destinations for earnings stripping payments, the salient provisions dealing with associated enterprises, interest, royalties, and nondiscrimination are substantially identical to the corresponding provisions in the current U.S. Model. See Convention for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income and on Capital Gains, U.S.-Ir., arts. 9, 11, 12, 25, July 28, 1997, 2141 U.N.T.S. 167; U.S.-U.K. Treaty, supra note 20, at arts. 9, 11, 12, 25.

     [104].   See 2006 U.S. Model, supra note 22, at art. 11, para. 5, art. 12, para. 4.

     [105].   See supra text accompanying notes 19–22.

     [106].   2006 U.S. Model, supra note 22, at art. 9, para. 2.

     [107].   United States Model Technical Explanation Accompanying the United States Model Income Tax Convention of November 15, 2006, art. 9, para. 1 [hereinafter U.S. Explanation], available at www.treasury.gov/resource-center/tax-policy/treaties/Documents/TEMod006.pdf (emphasis added).

     [108].   See 2006 U.S. Model, supra note 22, at art. 24.

     [109].   In the U.S. context, Paragraph 24.2 applies only to permanent establishments (unincorporated branches) operated in the United States by nonresidents. See U.S. Explanation, supra note 107, at art. 24, para. 2. Paragraph 24.3 applies only to nonresidents who are individuals. See U.S. Explanation, supra note 107, at art. 24, para. 3.

     [110].   See 2006 U.S. Model, supra note 22, at art. 3, para. 1(j)(ii).

     [111].   Id. at art. 24, para. 1.

     [112].   Regarding the uncertain incidence of the corporate income tax, see U.S. Treasury Dep’t, Office of Tax Analysis, A Review of the Evidence of the Incidence of the Corporate Income Tax 2 (2007) (concluding that labor may bear a substantial burden from the corporate tax); Jane G. Gravelle & Thomas L. Hungerford, Cong. Research Serv., RL34229, Corporate Income Tax Reform: Issues for Congress 15–29 (Cong. Res. Serv. 2011) (concluding that most of the corporate tax’s burden falls on capital).

     [113].   See 2006 U.S. Model, supra note 22, at art. 24, para. 1.

     [114].   See Org. for Econ. Coop. & Dev., Model Tax Convention on Income and Capital, Commentary on Article 24, paragraph 1 [hereinafter OECD, Commentary] (“Article [24] should not be unduly extended to cover so-called ‘indirect’ discrimination.”); Mary C. Bennett, The David R. Tillinghast Lecture—Nondiscrimination in International Tax Law: A Concept in Search of a Principle, 59 Tax L. Rev. 439, 468 (2006) (stating that in the few relevant cases, the “courts showed no willingness to look behind the terms of the relevant law to determine whether its overwhelming practical effect would be to disfavor members of the protected class”).

     [115].   2006 U.S. Model, supra note 22, at art. 24, para. 1.

     [116].   U.S. Explanation, supra note 107, at art. 24.

     [117].   A U.S. rule prohibiting deductions of costless expense payments to foreign related parties would be mainly applicable to payments by U.S. subsidiary corporations to affiliated foreign corporations—the scenario illustrated in Example 1. With respect to that situation, the following comment by a prominent international income tax law practitioner is relevant:

In the case of corporations, Article 24(1) . . . has essentially no impact for U.S. income tax purposes, because the dividing line between those corporations subject to U.S. tax on their worldwide income and those not so subject is . . . whether the corporation is incorporated under U.S. laws. Thus, a corporation that is not a U.S. “national” is, virtually by definition, not “in the same circumstances” as a corporation that is a U.S. national.

Bennett, supra note 114, at 446; see also OECD, Commentary, supra note 114, at para. 17 (stating that “two companies that are not residents of the same state for purposes of the Convention . . . are usually not in the same circumstances for purposes of paragraph [24].1”); Philip F. Postlewaite & David S. Makarski, The A.L.I. Tax Treaty Study—A Critique and a Modest Proposal, 52 Tax Law. 731, 794 (1999) (“Non-residents are seldom in the same circumstances as that of a resident because they are generally not subject to tax on their worldwide income.”).

     [118].   See infra note 146.

     [119].   2006 U.S. Model, supra note 22, at art. 24, para. 4.

     [120].   In UnionBanCal Corp. v. Comm’r, 113 T. C. 309, 325–26 (1999), aff’d, 305 F.3d 976 (9th Cir. 2002), the courts treated the term “enterprises” in Paragraph 24(5) of the United States-United Kingdom income tax treaty, which was virtually identical to Paragraph 24.5 of the current U.S. Model, as including business corporations.

     [121].   See Square D Co. & Subs. v. Comm’r, 438 F.3d 739, 748 (7th Cir. 2006), aff’g 118 T.C. 299 (2002) (stating that Paragraph 24.5 “hinges on the nationality of the related party to whom the payment goes and does not fluctuate based on nationality of the ultimate owner [of the payor]”).

     [122].   See UnionBanCal Corp. v. Comm’r, 305 F.3d 976, 987 (2002) (stating that “discrimination against foreign-owned subsidiaries is all that the nondiscrimination clause at issue protected it against”).

In Square D Co. & Subs., 438 F.3d 739, a U.S. resident subsidiary corporation claimed 1991 and 1992 deductions under the accrual method of accounting for interest then owed to its French parent corporation but not paid until 1995 and 1996. 438 F.3d at 740. Treas. Reg. § 1.267(a)-3 generally provides that interest owed to a foreign related party cannot be deducted until actually paid. Relying on this regulation, the IRS disallowed the U.S. subsidiary’s 1991 and 1992 deductions and asserted that they had to be deferred until actual payment was made in 1995 and 1996. Id. The U.S. subsidiary contended that the regulation violated the portion of the United States-France income tax treaty that was substantially identical to Paragraph 24.5 of the current 2006 U.S. Model. Id. at 743. The U.S. subsidiary’s argument was that because the deductions would have been allowed in 1991 and 1992 if the parent corporation had been a U.S. resident, denial of the deductions until 1995 and 1996 on account of the parent corporation being a French resident amounted to more burdensome taxation on a French corporation’s U.S. subsidiary in violation of the treaty. Id. The Seventh Circuit affirmed the Tax Court and upheld the validity of the regulation, even though it discriminated, because it did not create the type of discrimination that was prohibited by Paragraph 24.5 of the U.S. Model. Id. at 748. The pertinent portion of the Seventh Circuit opinion states that

[Prohibited] discrimination is absent here. The regulation requires that all interest payments to a foreign related party must use the cash method of accounting without regard to the nationality of the owner [of the payor]. The regulation does not impose the cash method simply because of foreign ownership [of the payor], which would be prohibited, but rather for payments to a foreign related party. Even if a corporation were owned by a United States parent, it still appears all interest payments to one of these foreign related parties would lead to the use of the cash method. The requirement, therefore, hinges on the nationality of the related party to whom the payment goes and does not fluctuate based on nationality of the ultimate owner [of the payor]. It is merely fortuitous that, in this case, the foreign related party to which the payment was made also happened to be the owner. The regulation does not discriminate based on foreign ownership [of the payor], and thus, does not violate the nondiscrimination clause.

Id. at 748 (emphasis added); see also Square D Co. & Subs., 118 T.C. at 313–15 (adopting the Seventh Circuit’s decision).

[123].   2006 U.S. Model, supra note 22, at art. 24, para. 4.

     [124].   The Technical Explanation of the U.S, Model Treaty states, in the introductory paragraphs of the explanation of Article 24, that there is an exception to Paragraph 24.4 where foreign and domestic payees are in dissimilar situations. However, that is a printing error and the statement clearly is intended to refer to Paragraph 24.5. See U.S. Explanation, supra note 107, at art. 24.

     [125].   See U.S. Explanation, supra note 107, at art. 24, para. 4.

     [126].   For an explanation of the loose constraint imposed by Section 163(j), see infra Part III.B.

     [127].   See H.R. Rep. No. 101-386, at 569 (1989).

     [128].   See id.; Am. Law Inst., Federal Income Tax Project: International Aspects of United States Income Taxation II—Proposals on United States Income Tax Treaties 281 (1992) [hereinafter ALI, Treaty Proposals].

     [129].   Paragraph 24.4 does not apply to nondeductible dividends and is also inapplicable to other payments to the extent that they are non-arm’s-length.

     [130].   For prominent criticism of these arguments, see ALI, Treaty Proposals, supra note 128, at 281.

     [131].   2006 U.S. Model, supra note 22, at art. 24, para. 4.

     [132].   See IMF, Spillovers, supra note 49, at 33–34 (suggesting limitations on deductions for intra-MNE interest payments as a partial alternative to transfer pricing enforcement based on the arm’s-length principle).

     [133].   See Avi-Yonah, International Tax as International Law, supra note 53, at 44–45; Lawrence Lokken, The Sources of Income from International Uses and Dispositions of Intellectual Property, 36 Tax L. Rev. 233, 241–42 (1981); Shay, Fleming & Peroni, What’s Source Got to Do with It?, supra note 29, at 143.

     [134].   See Lokken, supra note 133, at 241–43.

     [135].   See id.

     [136].   See id. at 242 (“[A] typical royalty cannot feasibly be split between the place of the licensee’s use of the licensed property and the place of the activities by which the licensor created the property.”); see also Wolfgang Schön, International Tax Coordination for a Second-Best World (Part II), World Tax J., Feb. 2010, at 65, 91 [hereinafter Schön, Part II].

     [137].   See Lokken, supra note 133, at 242–43 (“Overall, the country in which a licensee uses intellectual property therefore has the strongest claim of origin.”).

     [138].   See I.R.C. §§ 864, 871, 881, 882 (2012).

     [139].   Professor Lokken did not, however, argue for this extension of his analysis. See Lokken, supra note 133, at 241–43.

     [140].   Professor Lokken addressed the operation of the U.S. rules for taxing the U.S.-source income of nonresident aliens and foreign corporations. He did not expressly deal with the inter-nation allocation of corporate profits or the consequences of denying deductions when computing corporate net income (i.e., profits). However, his analysis can be fairly understood as suggesting the outcome stated in the text. See Lokken, supra note 133, at 241–43.

     [141].   See Schön, Part II, supra note 136, at 92–93.

Economists generally hold that the development of many intangible property items (excluding logos, brand names, and the like) creates uncontrollable spillover effects that can be enjoyed without cost by free riders. As a result, developers are said to be unable to capture the full benefit of their costs and so have a tendency to underinvest in creating this type of property. See Cong. Budget Office, Federal Support for Research and Development 1 (2007). A frequent prescription for solving this problem is a tax incentive for research and development. See, e.g., id.; Robert D. Atkinson, The Case for the Research Tax Credit, 142 Tax Notes 1226, 1226–27 (2014); Martin A. Sullivan, Time to Scrap the Research Credit, 126 Tax Notes 891, 891 (2010). In this Article, we take no position on this matter. We do note, however, that if a tax incentive is proper, the country where development occurs seems to be the appropriate governmental authority to provide the incentive. This point is not specifically applicable to the marketing intangible that is featured in Example 4 because that type of intangible is unlikely to create uncontrollable spillover benefits.

Additionally, assigning USSub’s royalty payments to the U.K.’s taxing jurisdiction as a return on UKCo’s development costs is consistent with the view that intangible income should be allocated to the countries where value creation occurs. See OECD, Action Plan, supra note 31, at 18, 20.

Professor Adam Rosenzweig has recommended the adoption of rules that would allocate intangible income to lesser developed countries to assist them in providing public goods to their residents. See Adam H. Rosenzweig, Building a Framework for a Post-BEPS World, 74 Tax Notes Int’l 1077, 1080–81 (2014). This raises broader questions about proposals for delivering foreign aid through the tax system. As explained in our earlier work, if the tax system is to be used to deliver economic assistance to needy countries, the best approach is to employ bilaterally negotiated arrangements rather than generally applicable statutory rules. See Fleming, Peroni & Shay, Fairness, supra note 7, at 344–49.

     [142].   In general, a normal market return on UKCo’s creation cost would be an arm’s-length royalty. If, however, market dynamics cause a difference to occur, we believe that a nuanced application of the arm’s-length benchmark such as that under development by the OECD should be the standard for allocating income between related parties. See OECD, Guidance on Intangibles, supra note 91, at 40–42.

     [143].   Professor Schön has demonstrated that in a complex intangible licensing situation, up to four different countries may have plausible taxing claims with respect to income generated by the intangible in question. See Schön, Part II, supra note 136, at 92. In such a case, a more nuanced approach to international income allocation will be required than the one examined in this Article. However, we believe that the simple binary allocation issue that we have discussed is adequate to demonstrate that the Investment Location Primacy Principle requires careful analysis and that the country where the intangible is put to use can satisfy its taxing claim even while allowing a deduction for the related royalty payment.

     [144].   Not only do deductions serve to define the net income tax base, they also shift income when coupled with a linked income inclusion that is imposed on another taxpayer. A prominent example in U.S. federal income tax law is § 215, which defines an alimony payor’s tax base by allowing a deduction for alimony payments, but which also shifts income from the alimony payor to the alimony payee because of the linked income inclusion imposed on the payee by § 71. Similarly, the allowance of U.S. deductions for USSub’s royalty payments shifts income to the U.K. tax base by leaving the United Kingdom with the exclusive right to tax those payments.

If UKCo had acquired the intangible by license or purchase, then the royalty would be a return on UKCo’s investment in the intangible. In this scenario, however, UKCo would not have engaged in any U.K. development activity, and USSub’s royalty payments to UKCo would be entirely the result of exploiting the U.S. market. Thus, the Investment Location Primacy Principle would not require that an income shifting deduction be allowed for the royalty payments.

     [145].   See supra text accompanying notes 78–82. We acknowledge the sometimes difficult practical issues of identifying which costs relate to development of an intangible and how to allocate the costs of unsuccessful research, but these issues are well known and regularly addressed in a variety of contexts. Moreover, our proposal to allow a normal return on costs subsumes these issues. See infra text accompanying notes 148 and 159.

     [146].   To illustrate the after-tax effects of these two scenarios, assume that USSub pays a $100 royalty payment to UKCo. Also assume that USSub’s marginal U.S. income tax rate is 35%. If the United States does not allow a deduction for the royalty payment, the payment will come from funds bearing a 35% U.S. tax. Thus, the after-tax cost to USSub of the $100 payment would be $154 ($154 × [1-.35] = $100). In contrast, if the United States does allow a deduction for USSub’s royalty payment to UKCo, the deduction will shelter the payment from U.S. tax so that only $100 of cash would be required to make the royalty payment.

     [147].   See infra text accompanying notes 190–98.

     [148].   See supra notes 141–42.

     [149].   See Fleming, Peroni & Shay, Formulary Apportionment, supra note 16, at 56–57.

     [150].   See id. at 158.

     [151].   Org. for Econ. Coop. and Dev., Model Tax Convention on Income and Capital, art. 7 (2010) [hereinafter OECD, Model Convention].

     [152].   See id. at art. 7, para. 2.

     [153].   See OECD, Action Plan, supra note 31, at 19–20; OECD, Guidance on Intangibles, supra note 91, at 9–12 (proposing revisions to OECD transfer pricing guidelines on intangibles); Brauner, supra note 31, at 86–103.

     [154].   See Fleming, Peroni & Shay, Formulary Apportionment, supra note 16, at 55.

     [155].   See generally Kleinbard, Lessons, supra note 12 (explaining how U.S. MNEs move royalty income to low-tax foreign countries); Kleinbard, Stateless Income, supra note 85 (same).

     [156].   See generally Kleinbard, Lessons, supra note 12, at 124–29 (explaining how U.S. MNEs use cross-crediting to shelter royalties from U.S. taxation); Kleinbard, Stateless Income, supra note 85, at 727–50 (same).

     [157].   See, e.g., Fleming, Peroni & Shay, Perspectives, supra note 3, at 1081; Peroni, Fleming, & Shay, Getting Serious, supra note 3, at 468–70, 497–501, 507–16.

     [158].   The development country generally cannot tax income paid from a second country to a recipient that is located in a third country and that is not otherwise subject to the development country’s taxing jurisdiction.

     [159].   See supra text accompanying notes 147–58.

     [160].   See Fleming, Peroni & Shay, Designing, supra note 9, at 413–15.

     [161].   See id.

     [162].   See id. at 401–03.

     [163].   See id. at 413–15.

     [164].   See id.

     [165].   See supra text accompanying notes 133–44.

     [166].   For purposes of this discussion, we are continuing to assume that the payments are arm’s length. If that assumption is relaxed, the case for requiring a meaningful foreign tax as a condition to allowing the deduction becomes stronger.

     [167].   See supra text accompanying notes 27–31.

     [168].   See Kleinbard, Stateless Income, supra note 85, at 750–52; Shay, Fleming & Peroni, What’s Source Got to Do with It?, supra note 29, at 137–38.

     [169].   In other words, related corporations can agree among themselves to have intangibles inappropriately considered to be developed in a low-tax country because the work done and the risks assumed in the low-tax country are overstated.

     [170].   See Fleming, Peroni & Shay, Designing, supra note 9, at 401.

     [171].   See Peggy B. Musgrave, Sovereignty, Entitlement, and Cooperation in International Taxation, 26 Brook. J. Int’l L. 1335, 1341–42 (2001).

     [172].   In this situation, the lack of taxation by the investment location country substantially increases the likelihood that the investment location is artificial and arranged for tax-avoidance reasons.

     [173].   See Michael L. Schler, OECD vs. D/NI: Ending Mismatches on Hybrid Instruments, Part I, 75 Tax Notes Int’l 485, 488–89 (2014); OECD, Action Plan, supra note 31, at 15–16.

     [174].   See Kleinbard, Stateless Income, supra note 85, at 701–02; Bret Wells & Cym Lowell, Tax Base Erosion and Homeless Income: Collection at Source Is the Linchpin, 65 Tax L. Rev. 535, 538 n.10 (2012) [hereinafter Wells & Lowell, Tax Base Erosion].

     [175].   See generally Lee A. Sheppard, BEPS Effects Without Implementation, 142 Tax Notes 775 (2014) (explaining the use of debt/equity hybrid instruments).

     [176].   See OECD, Action Plan, supra note 31, at 15–16; J. Comm., Income Shifting and Transfer Pricing, supra note 59, at 48.

     [177].   See generally David D. Stewart, OECD Drafts Call for Domestic Law Solutions to Hybrid Mismatches, 142 Tax Notes 1315 (2014) (explaining OECD concerns regarding hybrid entities).

     [178].   Id.

     [179].   For discussions of this issue, see generally id. (explaining the hybrid mismatch problem); Amanda Athanasiou, Hybrid Mismatch Proposals: Practical Problems Remain, 74 Tax Notes Int’l 1083 (2014) (same).

     [180].   See U.S. Treas. Dep’t, General Explanations, supra note 14, at 10.

     [181].   See OECD, Action Plan, supra note 31, at 15–16.

     [182].   See Schön, Part I, supra note 84, at 92.

     [183].   See generally Wolfgang Schön, The Distinct Equity of the Debt-Equity Distinction, 66 Bull. for Int’l Tax’n 490 (Sept. 2012) [hereinafter Schön, Debt-Equity] (explaining that financial instruments are distributed along a debt/equity continuum that lacks sharp borders between debt and equity).

     [184].   Id.

     [185].   This is consistent with the widely applied principle that the source of services income for taxing purposes is the place where the services are performed. See I.R.C. §§ 861(a)(3), 862(a)(3) (2012); ALI, International Tax Proposals, supra note 81, at 57; Avi-Yonah, International Tax as International Law, supra note 53, at 45.

     [186].   See supra Part II.B.1.

     [187].   See supra Part II.C.1.

     [188].   The UKCo retained earnings that were the source of the funds loaned to USSub in Example 1 were themselves a return on some investment by UKCo. The proper tax treatment of that investment return, however, was previously effected by the relevant taxing authority. At that point, the retained earnings were simply cash or its financial equivalent. When those after-tax retained earnings were loaned to USSub, the interest thereon became an entirely new and different investment return and under the facts of Example 1, it was earned in the United States. This conclusion is consistent with the widely applied principle that the source of interest income is generally the debtor’s residence country. See I.R.C. § 861(a)(1); ALI, International Tax Proposals, supra note 81, at 67; Avi-Yonah, International Tax as International Law, supra note 53, at 43.

     [189].   See also Wolfgang Schön, International Taxation for a Second-Best World (Part III), World Tax J., Oct. 2010, at 227, 241 (2010) (interest payments on intra-group loans that do not involve outside financing “should be subject to the same treatment as dividends to controlling shareholders”).

Treas. Reg. § 1.482-2(a)(1) provides that

[w]here one member of a group of controlled entities makes a loan or advance directly or indirectly to . . .  another member of such group and either charges no interest, or charges interest at a rate which is not equal to an arm’s length rate of interest . . .  the district director may make appropriate allocations to reflect an arm’s length rate of interest for the use of such loan or advance.

Treas. Reg. § 1.482-2(a)(1) (2011). If this regulation is intended to reflect a U.S. decision that the USSub-UKCo MNE is entitled to have the United States recognize a deductible arm’s-length interest payment from USSub to UKCo in Example 1, we conclude that the decision is wrong as a matter of sound policy. However, we believe that the words “may” and “appropriate” in the preceding quotation mean that this regulation is not intended to create such an entitlement. We believe that these words should be read to permit disallowance of a USSub deduction for the interest paid to UKCo in Example 1. This view is supported by the following language from Treas. Reg. § 1.482-1(a)(1) (2013): “The purpose of section 482 is to ensure that taxpayers clearly reflect income . . . and to prevent the avoidance of taxes,” and by the following language from Treas. Reg. § 1.482-1(a)(2): “The district director may make allocations between or among the members of a controlled group if a controlled taxpayer has not reported its true taxable income.” We maintain that in Example 1, USCo’s true taxable income would not be reported if it were allowed to deduct its interest payments because those payments do not represent a real cost for reasons outlined in Part II of this Article.

     [190].   See supra Part II.B.1.

     [191].   See supra Part II.C.1.

     [192].   See supra Part II.C.3.

     [193].   OECD, Model Convention, supra note 151, at art. 9.

     [194].   See supra text accompanying notes 143–44, 146, 158–59.

     [195].   Compare supra Part II.B.1, with supra Part II.C.1.

     [196].   See ALI, International Tax Proposals, supra note 81, at 7; Avi-Yonah, International Tax as International Law, supra note 53, at 5–6, 89–90.

     [197].   The general international income tax rule is that interest has its source in the debtor’s residence country. See, e.g., I.R.C. § 861(a)(1) (2012); ALI, International Tax Proposals, supra note 81, at 66–67; Schön, Debt-Equity, supra note 183, at 500. The allowance of the interest deduction would be subject to the otherwise applicable interest expense allocation and allowance rules. See, e.g., I.R.C. §§ 163, 861(b).

     [198].   See supra Part II.B. USSub’s deduction is also required by the United States-U.K. income tax treaty. See U.S.-U.K. Treaty, supra note 20, at art. 25, para. 3.

     [199].   See supra Part II.B.

     [200].   Cf. supra text accompanying notes 139–44.

     [201].   See Fleming, Peroni & Shay, Designing, supra note 9, at 401–02.

     [202].   See id. at 402–03.

     [203].   See supra text accompanying notes 148–53.

     [204].   See, e.g., sources cited supra note 93.

     [205].   These problems would not be made easier by U.S. adoption of a territorial tax regime that limited the U.S. corporate tax base to U.S.-source income. The difficult allocations described in the text would still be required in order to make theoretically correct distinctions between U.S.-source income and exempt foreign-source income. See Wells, Whack-a-Mole, supra note 11, at 1433.

     [206].   See U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 7, 21.

     [207].   As stated by Jasper Cummings:

It seems that foreign acquirers have become adept at working around section 163(j). For example, the prospectus filed for the Chinese company purchase of Smithfield Foods states that the cash will be provided by the Bank of China and Morgan Stanley. The prospectus says the capitalization of the parent will be 40% equity. That strongly suggests that the parties are intending to fit in the 1.5 to 1 debt-to-equity safe harbor for section 163(j). Alternatively, the 50% of adjusted taxable income limitation can be easily accommodated in many cases.

Regulations under section 163(j) were proposed in 1991 and have never been finalized. The proposed regulations are rarely cited, and section 163(j) itself is seldom cited. This supports the view that while the section must be observed, even the IRS considers it to be an ineffective preventer of deductions for interest paid to foreign affiliates . . . .

Jasper L. Cummings, Jr., Income Stripping by Interest Deductions, 141 Tax Notes 971, 978 (2013).

     [208].   I.R.C. § 163(j)(1)(A), (2)(A) (2012). For a detailed discussion of the earnings stripping rules in § 163(j) and the proposed Treasury regulations issued under that provision, see 2 Kuntz & Peroni, supra note 23, ¶ C3.04[5][d].

     [209].   I.R.C. § 163(j)(3), (5)(B).

     [210].   Id. § 163(j)(1)(A).

     [211].   Id. § 163(j)(2)(B), (6)(B).

     [212].   Id. § 163(j)(6)(A).

     [213].   See H.R. Rep. No. 101-386, at 567, 569–70 (1989) (Conf. Rep.); see also Hey, supra note 77, at 342–43.

     [214].   See H.R. Rep. No. 101-386, at 564–65, 567–68.

     [215].   See id. at 567–68, 569–70; see also Shay, Mr. Secretary, supra note 14, at 474 (discussing Barclays Bank research analysts’ report that assumes a foreign-owned Walgreens could add intercompany debt up to the § 163(j) limit).

     [216].   See supra text accompanying notes 68–73, 206–12.

     [217].   See I.R.C. § 163(j)(1)(B) (2012).

     [218].   See id. § 163(j)(1)(A).

     [219].   See id.

     [220].   See Hey, supra note 77, at 344 (“The key issue in respect of any legislation limited to ‘excessive’ intra-group interest is the lack of a normative basis for the definition of an adequate level of debt.”). Shareholder loans that arguably exceed the debtor corporation’s reasonable expected repayment capacity raise the question of whether the shareholder-creditors have an unqualified intention to enforce repayment or instead are making a disguised equity investment. This issue does not usually arise when the creditor is an unrelated party dealing at arm’s length with the debtor corporation.

     [221].   See I.R.C. § 163(j)(2)(B).

     [222].   See id. § 163(j)(2)(A).

     [223].   See generally H.R. Rep. No. 101-386, at 567, 569–70 (1989) (finding the 1.5 to 1 debt-equity ratio to excuse many corporations with typical capital structures from any potential disallowances and most net interest payments to be well under this threshold); Cummings, supra note 86 (discussing the benefits of consolidating affiliated foreign corporations owned by one common parent for tax filing purposes).

     [224].   For purposes of § 163(j), a subsidiary’s debts that are owed to its parent are generally taken into account in full for purposes of determining the subsidiary’s debt-to-equity ratio even if the parent obtained the loan funds by borrowing at arm’s length from an unrelated lender. See Prop. Treas. Reg. § 1.163(j)-3(b), CCH, Standard Fed. Tax Rep. 23355–56 (2013). This conflicts with our conclusion that when the unrelated-party debt produces a real interest cost for the parent and the subsidiary’s interest payments are a reimbursement of that cost, the disallowance scheme of § 163(j) should not apply to the subsidiary’s interest expense. See supra text accompanying note 78. More broadly, this indicates that the 1.5 to 1 debt-to-equity ratio is not a useful proxy for distinguishing between real interest expenses and interest payments that are costless foreign related party expenses.

     [225].   The safe harbor ceiling in the German interest deduction limitation rule is 30%. See Kleinbard, Lessons, supra note 12, at 141; Lee A. Sheppard, Interest Barriers Rise in Europe, 141 Tax Notes 251, 252 (2013).

     [226].   See I.R.C. § 163(j)(3)(A), (5)(B) (2012).

     [227].   See Kleinbard, Lessons, supra note 12, at 142–44.

     [228].   See Wells & Lowell, Tax Base Erosion, supra note 174, at 542, 604–08; see also Hey, supra note 77, at 343–44.

     [229].   See Wells & Lowell, Tax Base Erosion, supra note 174, at 605.

     [230].   See id. at 605 n.414, 608.

     [231].   For background on earnings stripping, see J. Comm., Cross-Border Income, supra note 5, at 54, 58–59; U.S. Treas. Dep’t, General Explanations, supra note 14, at 10.

     [232].   See generally Lodin II, supra note 67 (explaining inverted credit mechanism); Sven-Olof Lodin, Intragroup Lending in Sweden—A Vehicle for International Tax Arbitrage, 63 Tax Notes Int’l 177 (2011) [hereinafter Lodin I] (describing a proposal relating to interest deductions).

     [233].   See Lodin II, supra note 67, at 1318.

     [234].   See Lodin I, supra note 232, at 178; Lodin II, supra note 67, at 1317.

     [235].   See Lodin I, supra note 232, at 178; Lodin II, supra note 67, at 1317. Unlike our proposed analytical framework in this Article, Professor Lodin’s proposal does not deal with services fees payments between related parties.

     [236].   See Lodin I, supra note 232, at 178.

     [237].   See id.

     [238].   See Lodin II, supra note 67, at 1318–19.

     [239].   See id. at 1318.

     [240].   See id.

     [241].   See id.

     [242].   See id.

     [243].   See, e.g., 1 Kuntz & Peroni, supra note 23, at ¶ B3.05[8].

     [244].   See supra Part III.B.1.

     [245].   See U.S. Treas. Dep’t, General Explanations, supra note 14, at 10–12.

     [246].   See id.

     [247].   See id. at 11.

     [248].   See id. at 10.

     [249].   See id. at 11.

     [250].   See Cummings, supra note 207, at 978.

     [251].   See U.S. Treas. Dep’t, General Explanations, supra note 14, at 10–11; supra Part II.E.

     [252].   See U.S. Treas. Dep’t, General Explanations, supra note 14, at 11.

     [253].   See id. at 10–12.

     [254].   53 C.F.R. § 1.861–9T(a) (2014).

     [255].   Id. § 1.861–9T(g), (h), (j).

     [256].   See, e.g., 26 C.F.R. § 1.881–3 (2014) (the so-called anti-conduit regulations).

     [257].   See 2006 U.S. Model, supra note 22, at art. 11, para. 1.

     [258].   For the 2010 tax year, U.S. corporations that claimed a foreign tax credit received approximately $57 billion in interest from all foreign sources and paid approximately $55 billion of deductible interest to foreign creditors, both related and unrelated. See IRS Statistics of Income, Table 1-U.S. Corporation Returns with a Foreign Tax Credit, 2010, available at www.irs.gov/uac/SOI-Tax-Stats-Corporate-Foreign-Tax-Credit-Table-1. See generally Nuria E. McGrath, Corporate Foreign Tax Credit, 2009, IRS Statistics of Income Bull., Summer 2013, at 125 (providing a detailed analysis of foreign tax credit claims by U.S. corporations for the 2009 tax year). For 2006, the last pre-recession tax year for which pertinent IRS data is available, U.S. corporations with $500 million or more of total receipts and with 25% or greater ownership by a single foreign shareholder paid $53.4 billion of interest to foreign related parties and received $22.9 billion of interest from such parties. See Isaac J. Goodwin, Transactions Between Large Foreign-Owned Domestic Corporations and Related Foreign Persons, 2008, IRS Statistics of Income Bull., Fall 2012, at 182, 183.

     [259].   See U.S. Treas. Dep’t, Earnings Stripping, supra note 15, at 23, 26; OECD, Action Plan, supra note 31, at 8.

DOWNLOAD PDF | 93 N.C. L. Rev. 673 (2015)

Warning: Unknown: open(/home/content/83/7305283/tmp/sess_3m74l5k3df37f850p46hvm3nh3, O_RDWR) failed: No such file or directory (2) in Unknown on line 0

Warning: Unknown: Failed to write session data (files). Please verify that the current setting of session.save_path is correct () in Unknown on line 0