Skadden Partner Says Camp Territorial Plan a Good Start on Reform

I. Introduction

Reforming the international tax system in a balanced and sustainable way is an important goal. The key features of the current system -- taxation of domestic corporations' worldwide income subject to a foreign tax credit and the deferral of taxation on certain categories of income of foreign subsidiaries -- represent a legislative compromise forged in 1962, when U.S. companies were far more dominant in the global economy than they are today. Since then, changes in U.S. and foreign tax law, as well as in the global economic environment, have given rise to stresses and challenges that threaten to undermine that regime.

Two fundamental aspects of the current regime are unsustainable over the long run. First, in a world where most foreign countries have substantially reduced their corporate tax rates, the ability of U.S. companies to defer U.S. taxation of certain foreign income has encouraged those companies to leave their lower-taxed foreign earnings offshore, often indefinitely. This "lock-out" effect, which has trapped over $1 trillion of U.S. corporate earnings offshore, breeds economic inefficiency and discourages reinvestment of those funds in the United States. It is a ticking time bomb for U.S. multinationals. Second, with respect to the few major countries (principally, Japan) with rates higher than U.S. rates, U.S. companies' ability to "cross-credit" those high taxes with foreign-source interest and royalties arguably circumvents intended limitations on the creditability of foreign taxes by reducing U.S. tax on items of income that are properly taxable in the U.S. and not in any foreign jurisdiction.

With the Discussion Draft, the Committee has offered a bold proposal that refocuses and advances the international tax reform debate in important ways. In commenting on the Discussion Draft, I will address both the policy goals that I believe should be advanced and specific issues that arise in implementing legislation to further those goals. There are four goals in particular that I believe international tax reform legislation should advance:

By offering the Discussion Draft in the form of legislative language, the Committee has provided a foundation for beginning to think about issues in implementing this kind of legislation. Based on that language, I will offer my initial thoughts on the major provisions of the Discussion Draft.

II. Participation Exemption

The Discussion Draft's 95% dividends-received deduction ("DRD") for foreign-source dividends from CFCs, which is equivalent to a 95% exemption for such dividends, is the cornerstone of the territorial system proposed by the Discussion Draft and is consistent with international practice. With regard to the DRD, I would like to focus in particular on the rationale for taxing 5% of an otherwise eligible dividend. As a policy matter, I have no objection to this minor toll charge if its purpose is to generate revenue needed to implement the overall goals of the Discussion Draft. I do not believe that an effective 1.25% tax (assuming a 25% corporate income tax rate), or even an effective 1.75% tax (assuming today's 35% rate), on foreign-source dividends is significant enough to discourage companies from repatriating foreign earnings in a manner that would perpetuate the lock-out effect of current law. I would like, however, to explore and challenge the rationale offered for this toll charge by the Technical Explanation of the Discussion Draft, namely that the toll charge is intended as a substitute for disallowance of deductions for expenses incurred to generate exempt foreign income. As a tax policy matter, I do not believe any deduction disallowance beyond a thin capitalization-type rule that is included in the Discussion Draft to be appropriate in any case.

I start with the premise that in a properly designed territorial system, every expense that is currently in the nature of a deductible expense should be currently deductible. The relevant question for determining current deductibility of expenses in a territorial system is which jurisdiction should permit the deduction for a particular expense. In addressing this question, tax policy should be guided by the fundamental matching principle that an expense should be deductible against the tax base that will include revenue arising from that expense. Consider in turn the application of this principle to the three major categories of indirect expenses: interest expense, general and administrative ("G&A") expense, and research and development ("R&D") expense.

Full deductibility of interest expense is a crucial issue for many companies, as interest is often the largest indirect expense a company incurs. In the context of both a territorial system and a worldwide system with deferral, concerns have been expressed that borrowing by a U.S. multinational may finance investments that produce income not subject to current U.S. tax. This concern underlies foreign countries' interest deduction limitations and appears to motivate the Obama Administration's proposal to defer interest deductions allocable to deferred foreign income. Indeed, in some circumstances, it may be more appropriate for a portion of domestic interest expense to be deducted in a country where a CFC earns income and pays tax than for that interest expense to reduce U.S. taxable income. As I will discuss later in this testimony, to effect this outcome where local borrowing by the CFC may not be efficient or feasible, any interest deduction disallowance or deferral provision should allow a U.S. parent with third-party debt to lend to its CFCs with the same consequences as if the CFC had borrowed directly from the third party with a parent guarantee. With a well-designed mechanism that acknowledges a CFC's interest expense deductions in its home country, U.S. multinationals can arrange their global capital structure so that no domestic interest expense deductions actually need to be disallowed.¹ The Discussion Draft's thin capitalization proposal provides a framework for such a mechanism. Thus, there is no need to reduce the DRD to 95% as an alternative to interest expense disallowance.

Both a deferral and a territorial system based on sound principles of tax policy also should permit the domestic parent of a multinational company to fully deduct G&A expenses somewhere in the world. Under the arm's length standard as applied both in the U.S. and under OECD transfer pricing guidelines, a domestic corporation may only require its CFCs to reimburse expenses that provide a benefit to the CFC that the CFC would pay for were it an unrelated party. Thus, for example, a domestic corporation generally is not permitted to charge its CFCs for a portion of the salary of the domestic corporation's CEO since any benefits of the CEO's activities to the CFCs likely would be indirect. Like the current "water's edge" interest allocation under section 864(e),² such a rule may be tolerable in today's system,³ where the allocation of expenses to foreign income only reduces broadly-defined foreign-source income that is eligible for a foreign tax credit. But a system of direct expense disallowance (or deferral) requires re-thinking these rules. Clearly, no deduction disallowance is appropriate to the extent U.S. G&A expenses are properly charged out to CFCs. But unlike with interest expense, deductions for G&A expense of a domestic corporation that cannot be charged out can only be moved to foreign jurisdictions by moving actual functions and jobs. To the extent that G&A expenses cannot be charged out because they are treated under arm's length principles as benefiting only the parent corporation, these expenses should be currently deductible in the United States under a worldwide or territorial system.⁴ Failure to allow such a deduction would prevent taxpayers from deducting G&A expense anywhere in the world. Moreover, it would encourage companies to relocate G&A-related headquarters jobs and operations to foreign jurisdictions where they could be fully deductible.

As with interest and G&A expense, there is no tax policy basis for disallowing R&D expense in a territorial tax system. Where a domestic corporation develops intangible property and transfers that property to a foreign affiliate, section 482 (and section 367(d), in the case of a nonrecognition transfer) requires the foreign affiliate to compensate the domestic transferor for the intangible property on an arm's length basis. Assuming proper application of section 482, the domestic transferor should take into account both R&D expense and the related intangible income (in the form of a royalty) in calculating its U.S. taxable income. This analysis applies equally in a deferral or a territorial system, meaning there is no basis for deferring or disallowing any portion of R&D expense deductions in either system. Moreover, perhaps even more than with G&A expense, disallowing R&D expense deductions would encourage companies to move valuable R&D jobs to a jurisdiction where R&D expense would be deductible. This is presumably the reason why the Obama Administration limits its expense deferral proposal to interest expense.

Thus, there is a sound basis in tax policy to permit deductions of all major indirect expenses in a territorial system, taking into account refinements of the thin capitalization proposal. The Discussion Draft appropriately permits deduction of these expenses; the limitation of the DRD to 95% of eligible dividends should not be viewed as a substitute for disallowance of deductions for these expenses.

III. Treatment of Gains and Losses on Disposition of Certain Stock

The Discussion Draft's exemption of 95% of a 10% domestic shareholder's gains on disposition of stock in a qualified foreign corporation is an important element of a consistent territorial system. This exemption is needed to maintain the principle, effectuated first and foremost by the 95% DRD, that active foreign earnings are generally exempt from U.S. tax. Failure to allow the exemption for such gains would create tension within the system by taxing foreign affiliate earnings differently depending on whether a shareholder realizes the benefit of those earnings through a dividend or through a sale of stock.

The exemption for gains on qualified foreign corporation stock is consistent with the Code's policy of preventing corporate-level double taxation in the context of domestic stock sales. Through an election under section 338(h)(10), an acquirer of domestic stock can obtain a stepped-up basis in the acquired corporation's assets and avoid potential gain with respect to those assets where the selling consolidated group has already recognized gain with respect to those assets. Although it is not possible to compel a foreign government to tax stock acquisitions as asset acquisitions and give an acquirer of foreign stock a step-up in the basis of the acquired company's assets (as section 338(h)(10) would provide), the Discussion Draft's exemption for sales of foreign stock similarly prevents a potential double tax.

The Discussion Draft appropriately recognizes that exempting gains on the sale of stock is not appropriate where the foreign corporation's income is currently taxed as subpart F foreign personal holding company income. It differentiates foreign corporations earning such income by limiting the exemption to stock in foreign corporations that have not had more than 30% of their assets generate foreign personal holding company income as of the end of any quarter over the prior 3 years. Such a rule creates a huge cliff effect that can apply in a wide variety of situations. As an alternative, I would suggest consideration of an asset look-through approach, akin to a mandatory section 338(g) election, that would effectively limit the exemption to gain on the sale of stock that reflects underlying gains in business assets.⁵

IV. Transition Tax

The Discussion Draft's 5.25% transition tax on a deemed dividend of accumulated deferred foreign earnings is evidently a revenue-raising measure. As a onetime tax on an existing asset stock, it is a relatively efficient tax in that it should not distort economic decision-making in the way that income and other taxes may. Most importantly, it eliminates the lock-out effect for pre-effective date earnings. Nonetheless, considerations of equity and ease of compliance may argue for refining the design of this tax.

As drafted, the 5.25% transition tax would apply to all earnings and profits not previously subject to U.S. tax, regardless of whether those earnings represent available offshore cash. Thus, the tax would apply with respect to earnings and profits that have been invested in overseas operations or used to finance stock or asset acquisitions. Because the transition tax base is broader than available liquid assets, companies may face liquidity issues in paying the transition tax (although the installment payment option may alleviate this burden in certain cases). In addition, for many companies, especially those with longstanding foreign affiliates, determining historical earnings and profits may be challenging. The challenge may be even greater with respect to so-called 10/50 companies for which a 10% domestic shareholder may not have easy access to available records.

One possible alternative to taxing all deferred foreign earnings would be to simply tax, perhaps at a somewhat higher rate to reduce the revenue impact, earnings and profits to the extent of cash and other liquid assets. Note that this alternative would potentially create three categories of foreign earnings -- previously taxed earnings, previously untaxed earnings subject to the transition tax, and previously untaxed earnings invested in a foreign business -- and query how the last category of earnings should be treated if distributed.

As an alternative to taxing all deferred earnings or deferred earnings corresponding to available liquid assets, the Committee also might consider tying the transition tax to the current-law rules for the amount characterized as a dividend under section 1248. With respect to a particular CFC, the amount includible under section 1248 is essentially the smaller of the U.S. shareholder's gain with respect to its stock and the earnings and profits accumulated during the U.S. shareholder's ownership period for that stock. In many circumstances, since the amount included under section 1248 is limited to stock gain, it may be an easier amount to determine than all deferred earnings and profits. Additionally, applying the transition tax to the amount included under section 1248 may be more equitable, since it would limit application of the transition tax to a shareholder's actual economic gain with respect to its foreign affiliates.

As a simplification, I also would suggest limiting the transition tax to that portion of the amount included under section 1248 that reflects "post-1986 undistributed earnings" as defined in section 902(c)(1). Since the Tax Reform Act of 1986, U.S. multinationals have had to keep records of cumulative earnings and profits for most, if not all, CFCs for foreign tax credit calculation purposes. For prior earnings (likely to be relatively small in amount), record keeping can be a real issue.

I also would highlight an apparent technical glitch with respect to the scope of the transition tax. As currently drafted, the transition tax properly excludes from its scope "subpart F income" previously taxed under section 951(a). Note, however, that the term "subpart F income," as defined in section 952(a), does not include earnings taxed under section 956, and possibly not earnings taxed under section 1248 or section 964(e). I assume that the Committee intends for all previously taxed earnings to be excluded from the transition tax.

V. Foreign Tax Credit Modifications

In permitting a 95% DRD for foreign-source dividends and eliminating the section 902 foreign tax credit with respect to income eligible for the DRD, the Discussion Draft would fundamentally change the role of the foreign tax credit in the U.S. system. This achieves an important goal given the incredible complexity of our current system and the cross-crediting planning it permits. Although a foreign tax credit no longer would be necessary to relieve double taxation with respect to DRD-eligible dividends, the credit would remain essential for relieving double taxation with respect to foreign taxes imposed on income subject to current U.S. tax. Thus, as the Discussion Draft provides, a foreign tax credit should be available with respect to foreign taxes imposed on amounts included under subpart F and with respect to foreign withholding tax on payments (other than DRD-eligible dividends) from foreign parties to a U.S. company.

Given the limited situations where foreign tax credits would be available, changes to the section 904 limitation such as those in the Discussion Draft are appropriate. Aspects of the section 904 limitation such as basketing of income and foreign tax credits and allocation of expenses, as well as the new anti-splitter rules at section 909, add considerable complexity to the international tax regime. These aspects may be eliminated where other elements of the Discussion Draft are sufficient to ensure that the overall purposes of the section 904 limitation are satisfied. Given the importance of eliminating cross-crediting and the separation of credit from income as a goal of international tax reform, however, changes to sections 904 and 909 should be done in a way that prevents these activities going forward even if the result causes some residual complexity in the system.

VI. Base Erosion Options

In putting forth the three so-called "base erosion" options for taxing certain income subject to low foreign tax, the Discussion Draft begins an important discussion about the proper scope of subpart F. That discussion should be expanded to include other aspects of subpart F as well. In particular, the scope of any active finance exception, the role of the related party look through rules, and the policies behind subpart F services and sales income should all be re-examined.

Subpart F is the product of decades of legislative compromise that has as one of its lessons the need for great care when expanding current taxation to income from active businesses, especially where foreign countries do not impose similar rules on their multinational companies. Drifting outside of the mainstream can have serious consequences for competitiveness of U.S. companies and industries, as for example where the effective deferral regime for shipping income was repealed as part of the 1986 Act. The historical treatment of active financing income under subpart F is also instructive on this point. With these lessons in mind, the Discussion Draft's options for expanding subpart F should be evaluated in light of international norms.

Before exploring each of the Discussion Draft's options individually, I want to address a notion that seems to form part of the motivation for broadening subpart F. No doubt, the Committee has often heard warnings that under a territorial system incentives for U.S. taxpayers to transfer intangibles and other income-producing assets abroad would increase dramatically. Notwithstanding these warnings, I simply do not see the incentives under a territorial system as substantially different than under current law. In general, U.S.-based companies already effectively enjoy exemption with respect to the large share of their foreign earnings that they treat as permanently reinvested abroad. Moreover, the reduction of foreign tax rates over the past 5 years has resulted in most U.S. multinationals paying some substantial U.S. tax on their royalty income, thereby providing, on the margin, a strong incentive to minimize royalty income consistent with U.S. transfer pricing rules. In practice, a territorial system may only alter transfer pricing incentives for smaller companies that have relatively greater domestic cash needs and fewer borrowing options. For larger taxpayers and the bulk of income, incentives to shift functions, activities, and intangible property offshore likely would not differ appreciably under a territorial system as compared to the current deferral regime. Nonetheless, in the context of international tax reform and the adoption of a territorial system, it remains important to consider the proper scope of subpart F.

The first fundamental problem with Option A is that, in capping a CFC's return at a percentage of its current-year costs for purposes of determining excess intangible income, it does not provide the CFC with a return related to the risk it undertook in funding the R&D in prior years. As an example, think of a CFC that develops and brings to market a pharmaceutical compound to which it acquired rights during an early-stage development period. If the product is successful, Option A would subject to current U.S. tax as "excess" income most of the CFC's return to risk-taking undertaken in prior years. In effect, the U.S. would be taxing income earned from successful intangible development without even granting a deduction for prior intangible development costs.

Moreover, notwithstanding the requirement that some U.S. intangible transfer or risk-sharing agreement occur, Option A does not differentiate situations where if income is shifted at all, it is largely, if not completely, shifted from other foreign countries, not the U.S. As discussed above, tax planning to reduce foreign taxes should be encouraged, not punished.

Option A also provides a huge incentive for U.S. multinationals to push routine sales and marketing-type costs into CFCs earning excess returns in order to minimize the impact of the provision. That can mean moving functions (and jobs) from other jurisdictions, including the U.S.

Finally, in expanding subpart F to active business income from intangibles developed by a CFC, Option A would put the United States outside the international norm for territorial systems. Doing so could have significant competitive implications for U.S. companies, since Option A could subject a large share of U.S. companies' intangible income to current U.S. tax while foreign-based multinationals generally may enjoy exemption with respect to their foreign intangible income.

Notwithstanding its simplicity and broad coverage, Option B overreaches relative to the CFC regimes of other countries with a territorial system. Japan's CFC rules, on which Option B appears to be modeled, include a broad exception for a CFC's income from operations in its home country if the CFC has a substantial physical presence as well as local management and operations in that country. The comparable home-country exception in Option B includes an additional requirement that limits the exception to income from sales that ultimately serve the CFC's home-country market. Most U.S. multinationals operate globally. Manufacturing and, in many cases, the provision of services are done in a few locations or even a single location to serve a regional market or even global markets. The requirement that a manufacturer, for example, only sell into its local-country market is simply unrealistic. As a result under Option B, a very substantial amount of foreign earnings would be subject to current U.S. tax, unless a relatively low foreign tax rate threshold were included. To stay within the mainstream of practice among major countries with territorial regimes, the additional local-country requirement should be omitted if Option B is pursued.

It also should be noted that if Option B does apply to a broad category of business income, setting a foreign tax rate threshold for categorization of income as subpart F income could encourage certain smaller jurisdictions with a substantial U.S. multinational presence to raise their tax rates. Doing so could benefit both the U.S. company, whose income would avoid current taxation under subpart F, and the foreign jurisdiction itself, which would enjoy greater revenue at the expense of the U.S. fisc.⁶ Narrowing the scope of Option B by omitting the local market restriction as described above would reduce the incentive for any foreign jurisdiction to raise tax rates.

A significant feature of Option C is the availability of a reduced tax rate for all foreign intangible income, including income earned directly by a domestic corporation and CFC income included in a U.S. shareholder's income under subpart F. By treating subpart F and other income consistently, Option C theoretically mitigates any incentive to transfer or develop intangibles offshore versus maintaining beneficial ownership of intangibles in the United States. As the Discussion Draft is written, CFC intangible income subject to a 13.5% foreign effective tax rate would be exempt from U.S. tax and so would enjoy a modest 1.5% benefit compared to intangible income earned directly by the CFC's U.S. shareholder. However, the U.S. shareholder could deduct its R&D expenses against income taxed at the 25% (or higher) corporate rate, thus providing a large incentive for development funding in the U.S. This feature helps assure that the U.S. tax system would not create an incentive to move R&D offshore.

Because Option C currently taxes all intangible income under subpart F, albeit at a reduced rate, care should be taken to ensure that the overall burden on intangible income is consistent with international norms. Option C's viability and ultimate effect on competitiveness would depend on the exact tax rates and thresholds chosen and how those compare to the tax burden of foreign-based multinationals with respect to their intangible income.

VII. Interest Expense Deduction Disallowance

The Discussion Draft would create new thin capitalization rules to deny interest deductions to the extent a domestic corporation is considered to have excess domestic indebtedness. As I discussed earlier, all interest expense represents a real cost of earning income and should be deductible against taxable income in some jurisdiction. A properly designed thin capitalization provision should provide a path for taxpayers to generate interest deductions in foreign jurisdictions as a way of avoiding an interest deduction disallowance in the United States. The Discussion Draft applies its debt-equity thin capitalization regime to net interest in excess of some percentage of adjusted taxable income. Such a rule makes sense on the theory that at modest levels of interest expense the complexity of a debt-equity analysis is not worth the effort.

At a more technical level, there are two features that I believe an interest disallowance provision should incorporate. First, an interest disallowance provision should operate based on net interest, as the Discussion Draft does for both its adjusted taxable income and its debt-equity tests. Second, to ensure taxpayers can obtain an interest expense deduction in some jurisdiction, parent borrowing and on-lending to a CFC should be treated in the same way as if the CFC had directly borrowed from a third party under the debt-equity test.

Taking into account net interest expense, as the Discussion Draft and current section 163(j) earnings-stripping rules do (but the section 861 income source rules do not), appropriately recognizes the close relationship between interest income and expense. In the financial sector, where the core business activity involves earning a spread by lending at a higher interest rate than the interest rate paid on borrowing, the relationship between interest expense and income is most evident. A similarly close relationship exists where a company accumulates liquid interest-bearing securities to pay off term debt. In these cases and others, allocating net interest expense acknowledges the close factual relationship between interest income and expense. Disallowing interest expense deductions on a gross basis would have a disproportionate negative impact on industries (like the financial service industry) with greater interest income and could harm U.S. competitiveness in those industries.⁷

To ensure taxpayers can deduct interest expense in the appropriate jurisdiction, the Discussion Draft's debt-equity interest disallowance provision should be refined. For a variety of reasons, it is frequently easier, more efficient, and more affordable for a U.S. company to borrow on behalf of its entire worldwide group than for its CFC to borrow on its own behalf. Accordingly, an interest disallowance provision should allow domestic corporations that borrow from unrelated parties to lend to their foreign affiliates with the same consequences as if the CFC borrowed directly from a third party with a parent guarantee. Allowing this flexibility brings no harm to the U.S. tax base (assuming arm'slength interest rates), since interest income paid to the parent by the CFC offsets the parent's interest deductions with respect to the on-lent portion of third-party borrowing. Because the interest disallowance provision in the Discussion Draft does not net a domestic corporation's loans to its CFCs with its third-party indebtedness in determining excess domestic indebtedness, a domestic corporation that incurs third-party debt and also lends to its CFCs could suffer interest disallowance that would not occur if its CFC had borrowed directly from a third party. Thus, under the Discussion Draft, a U.S. taxpayer may be forced to choose between non-deductible interest expense and costly direct borrowing from third parties by its CFCs. There are a number of possible technical refinements that would remedy this consequence of the Discussion Draft's interest disallowance provision; one of them should be adopted.

VIII. Conclusion

The Discussion Draft makes great strides toward remedying the unsustainable aspects of the current U.S. international tax regime -- namely, the lock-out effect inherent in deferral and the ability to cross-credit -- and establishing a system that is consistent with the international tax regimes of other major countries. Many significant design issues, including most importantly the scope of subpart F, must be discussed and addressed. Doing so will require a balancing of important tax policy goals. Nonetheless, the Discussion Draft represents a very good start toward meeting the difficult challenge of reforming the international tax system in a balanced and sustainable way.

² Unless otherwise indicated, all section references are to the Internal Revenue Code of 1986, as amended (the "Code").

³ The section 861 source-of-income rules require allocation of a portion of G&A expense to foreign-source income even though the arm's length principle would not permit an actual charge-out of the expense.

⁴ A way to expand the universe of G&A expenses that can be charged out would be to obtain international agreement to permit cost-sharing for G&A expenses, as is currently done for R&D and marketing expenses.

⁵ Depending on the outcome of subpart F reforms, discussed later, one might consider whether gain attributable to assets that give rise to other forms of subpart F income should be taxable rather than eligible for the exemption.

⁶ Options A and C present a similar incentive, although perhaps to a lesser degree than Option B because of the broad reach of Option B to CFC business income in a world of global markets.

⁷ Indeed, consideration should be given to netting not just interest income but other forms of passive income, such as leasing income, against interest expense.