FLORIDA TAX REVIEW VOLUME 8 2007 NUMBER 3 WILL U.S. INVESTMENT Go ABROAD IN A TERRITORIAL TAx: A CRITIQUE OF THE PRESIDENT'S ADVISORY PANEL ON TAx REFORM by James R. Repetti I. INTRODUCTIO N ..................................................................................... 304 II. THE EFFICIENCY EFFECTS OF A TERRITORIAL SYSTEM .................. 306 A. Analysis of Claimed Benefits ................................................... 306 B. The Altshuler-Grubert Article ................................................. 311 1. The Effective Tax Rate Determinations ...................... 313 2. Analysis of Historic Data: Foreign Investment in Low-Tax Countries by Companies Subject to a Territorial Tax ................................................................ 318 3. Analysis of Historic Data: Foreign Investment in Low-Tax Countries by U.S. Companies with Excess Credits ................................................................ 320 C. Does Foreign Investment Complement U.S. Investment? ....... 322 III. C O NCLUSION ..................................................................................... 324 Florida Tax Review WILL U.S. INVESTMENT Go ABROAD IN A TERRITORIAL TAX: A CRITIQUE OF THE PRESIDENT'S ADVISORY PANEL ON TAX REFORM by James R. Repetti" I. INTRODUCTION The Report of the President's Advisory Panel on Federal Tax Reform' (the "Report") recommends the U.S. adopt a territorial tax system that would exclude income earned by U.S. taxpayers actively conducting foreign businesses. This exclusion would also apply to dividends received from controlled foreign corporations to the extent such dividends were attributable to the corporation's conduct of active foreign businesses.2 The Report justifies this recommendation by stating that a territorial system is simpler than the current global approach employed by the U.S. and that a territorial tax will improve efficiency. In an accompanying article, Professor McDaniel has demonstrated that simplicity cannot be achieved under a territorial or global system. Under either system, the U.S. will have to address source issues, transfer pricing issues and IRC Section 367 concerns. This article focuses on the Report's efficiency arguments for a territorial tax. The Report asserts that a territorial tax will permit U.S. multinationals to compete more effectively in low-tax jurisdictions and will eliminate the tax bias against repatriating earnings.3 The Report anticipates that its proposal might generate concern about a potentially significant efficiency problem - whether a territorial system would cause U.S. businesses to allocate more jobs and assets overseas to low-tax countries. It says: At first glance, one might assume that exempting active foreign source income from U.S. taxation would lead to a substantial reallocation of U.S. investment and jobs world wide. A careful * Professor of Law and Thomas Carney Scholar, Boston College Law School. This article was presented at the 2006 Annual Symposium on International Taxation at the University of Florida Levin College of Law Graduate Tax Program. The author thanks Yariv Brauner, Mark Brodin, J. Clifton Fleming, Jr., Marjorie Komhauser, Paul McDaniel, Martin McMahon and Diane Ring for helpful comments to earlier drafts. He also thanks Joshua Gutierrez and Kevin Walker for research assistance and Cassandra Desmond for help in preparing this manuscript. 1. Report of the President's Advisory Panel on Federal Tax Reform, Simple Fair and Pro-Growth: Proposals to Fix America's Tax System 134 (Nov. 2005) (hereinafter "Report"). 2. Id. 3. Id. [VoL 8:3 Will U.S. Investment Go Abroad in a Territorial Tax study of how location incentives for U.S. multinational corporations may change under a territorial system similar to the one proposed for the Simplified Income Tax Plan provides different results. Researchers found no definitive evidence that location incentives would be significantly changed, which suggests that the territorial system the Panel has proposed would not drive U.S. jobs and capital abroad relative to the current system. 4 The study referred to by the Report to support its conclusion is Where Will They Go if We Go Territorial? Dividend Exemption and the Location Decisions of U.S. Multinational Corporations, by Rosanne Altshuler and Harry Grubert.5 In their careful study, Altshuler and Grubert say that they cannot "make any firm prediction of how location behavior would change if the U.S. were to adopt a dividend exemption system."6 They further note that "the analysis provides no consistent or definitive evidence that dividend exemption would induce a large outflow of investment to low-tax locations."7 The Report relies heavily on the Altshuler-Grubert article to bolster its assertion that the territorial system would not adversely affect U.S. production. Note, however, that the Altshuler-Grubert article does not conclude that a territorial system would have no effect on the investment of U.S. capital. The Report correctly states that the Altshuler-Grubert article found no evidence that a territorial system "would induce a large outflow.... ." The Report incorrectly uses this lack of evidence to leap to the inference that "the territorial system the Panel has proposed would not drive U.S. jobs and capital abroad...."' This inference is inappropriate. As lawyers, we have been trained that a lack of evidence means that no conclusion can be reached.9 Although the Report uses the absence of conclusive findings to support the assertion that there will be no adverse effect, a careful reading of the Altshuler-Grubert article reveals that Altshuler and Grubert view their results as inconclusive. They are very careful to point out that their analysis produces mixed results about whether an exemption system would lead to more foreign direct investment ("FDI") by U.S. multinationals. Indeed, as this article will discuss, the 4. Id. at 135 (emphasis added). 5. Rosanne Altshuler & Harry Grubert, Where Will They Go if We Go Territorial? Dividend Exemption and the Location Decisions of U.S. Multinational Corporations, 54 Nat'l Tax J. 787 (2001). 6. Id. at 807. 7. Id. 8. Report, supra note 1, at 135. 9. For example, in a criminal prosecution, the defendant is found "not guilty" when the prosecutor fails to produce evidence that shows beyond a reasonable doubt that the defendant committed the crime. The prosecutor's failure to produce evidence does not mean that the defendant is "innocent." 2007] Florida Tax Review Altshuler-Grubert paper leaves many unanswered questions that make it impossible to conclude what effect a territorial tax will have on domestic investment. This article is organized as follows. It first explains that the benefits of a territorial tax suggested by the Report, increased competitiveness of U.S. multinationals in low-tax countries and increased dividend repatriation, cannot be viewed in isolation.'0 Instead, it is necessary to analyze these benefits in the context of the increased excess burden" a territorial tax will impose on domestic investment. The key to determining the increase in excess burden is the extent to which the tax will increase foreign investment in low-tax countries at the expense of domestic investment. 2 Thus, the Report's reliance on the Altshuler-Grubert article as support for its conclusion that a territorial tax will not increase investment in low tax countries is of critical importance. This article describes the analysis used by Altshuler and Grubert in order to illustrate the inconclusive nature of their findings. 3 It also raises additional issues not addressed in their article that will have to be resolved in order to determine the effect of an exemption system on domestic investment. 4 Given the uncertainty surrounding how U.S. multinationals will respond to a territorial tax and Professor McDaniel's persuasive explanation that a territorial tax will not be simpler than a global tax, this article concludes that Report has failed to make a convincing case for a territorial system. II. THE EFFICIENCY EFFECTS OF A TERRITORIAL SYSTEM A. Analysis of Claimed Benefits The Report justifies its recommendation for a territorial tax by asserting that this system will eliminate the tax impediment to repatriating earnings and will make U.S. multinationals more competitive. A complete analysis, however, 10. See infra text accompanying notes 15-27. 11. In determining the efficiency of an income tax, economists often refer to the term "excess burden." The excess burden represents the welfare loss created by a tax that exceeds the tax revenue generated by that tax; see e.g. Richard A. Musgrave & Peggy B. Musgrave, Public Finance in Theory and Practice 444 (1973); Harvey S. Rosen, Public Finance 307 (7th ed. 2005). An income tax has an excess burden because it creates a disparity (or "tax wedge") between the income paid to the taxpayer and the after-tax income received by the taxpayer. Id. at 310-12. This difference causes the taxpayer to vary his behavior from the way he would have behaved in a tax-free world. For example, a tax on wages may cause a taxpayer to work more or less hours in response to the tax. Similarly, a tax on savings may cause a taxpayer to save more or less in response to the tax. This behavioral change creates a welfare loss to the taxpayer in addition to the taxes paid. Id. at 319-20. 12. See infra text accompanying notes 42-53 and 58-65. 13. See infra text accompanying notes 40-72. 14. See infra text accompanying notes 53-57 and 66-82.. [Vol. 8:3 Will U.S. Investment Go Abroad in a Territorial Tax requires that such benefits not be viewed in isolation. It is necessary to consider these benefits in the context of other welfare effects that a territorial tax may have. This part will first analyze the claimed benefits of a territorial tax and then shift the analysis to the broader context of the efficiency effects of such a system. The impact of a territorial tax on dividend repatriation will in part depend on whether the New View or Traditional View of dividends applies. Under the New View, a permanent elimination of a dividend tax on repatriation would have no effect on dividend payments. 5 The New View posits that the tax on repatriation is irrelevant to a decision to retain or distribute earnings because the tax will be incurred when the earnings are ultimately transferred to the parent. 6 The only relevant consideration for retention of earnings by a mature subsidiary is whether the after-tax return on investment of the retained earnings exceeds the after-tax return that could be earned by the parent in the event the earnings were repatriated. 7 Thus, where the foreign subsidiary is in a low-tax country, the decision whether earnings should be retained will depend on whether the after-tax return from retention will be higher than the after-tax return available from an identical investment in the U.S. The amount of a permanent tax that will be assessed on the dividend repatriating the earnings will not affect the decision to retain the earnings. In contrast, under the Traditional View, eliminating a tax on dividends should increase dividend payments.' 8 The Traditional View of dividends posits that a tax on dividends affects the amount of dividends paid. It theorizes that corporations pay dividends despite the tax burden because dividends confer benefits to stockholders, such as reduced agency costs, in addition to the actual 15. David Hartman borrowed learning from the New View ofdividends, which had previously been applied to domestic dividends, to argue that a permanent tax on repatriation should not in theory affect a mature foreign subsidiary's decision to retain earnings since the earnings are "trapped," i.e. the earnings will inevitably be subject to a dividend tax when paid. David G. Hartman, Tax Policy and Foreign Direct Investment, 26 J. of Pub. Econ. 107, 115-16 (1984). See 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, 304 n.10 (2001) for an excellent discussion of some ofthe nuances of Hartman's analysis. See also Dept. of the Treasury, Integration of Individual and Corporate Tax Systems: Taxing Business Income Once, Ch. 13 pp. 116-18 (1992) for analysis of the New View in the context of domestic tax policy. For a discussion of the New View's perspective on how dividend taxation affects the decision to transfer new capital to a foreign subsidiary, see infra text accompanying notes 23-26. 16. Hartman, supra note 15, at 115-16. 17. Id. at 116-17. 18. See e.g. George R. Zodrow, On the "Traditional" and "New" Views of Dividend Taxation, 44 Nat'l Tax J. 497, 503 (1991). For a discussion of the Traditional View's perspective on the impact of dividend taxation on capital contributions to a foreign subsidiary, see infra text accompanying notes 23-26. 20071 Florida Tax Review dollar amount paid.'9 Lower taxes should result in higher dividend payouts under the Traditional View because the after-tax value of the dividend will have increased.2" The empirical evidence is mixed as to which theory provides a better explanation of reality, but recent studies suggest that dividend taxes affect distributions by foreign subsidiaries.2 Consequently, it seems at least plausible that a territorial system would encourage more dividend payments by foreign subsidiaries to their U.S. parents. This does not mean, however, that eliminating the tax on dividends would necessarily increase domestic investment. Increased dividends from foreign subsidiaries may subsequently be invested in low-tax countries. Adopting a territorial tax will in theory create an incentive for U.S. corporations to transfer new investment to low-tax countries since the return on such investment would now be subject to lower rates.22 Since the tax rate in the U.S. would exceed the rate in the low-tax country, the pre-tax rate of return on investment in the U.S. would have to exceed the pre-tax return in the low-tax country to keep investment in the U.S. This incentive that a territorial tax will create to transfer new capital to low-tax countries exists regardless of whether the New View or Traditional View of dividends applies. Although the New View posits that a tax on dividends is irrelevant to a decision to retain or distribute earnings, 23 the New 19. Zodrow, supra note 18, at 497, 503. 20. Id. 21. For a summary of the literature supporting the applicability of the Traditional View to dividend payments by foreign subsidiaries, see Harry Grubert, Comment on Desai and Hines, "Old Rules and New Realities: Corporate Tax Policy in a Global Setting", 58 Nat'l Tax J. 263, 267-268 (2005). It should be noted that the empirical evidence that taxation affects dividends does not mean that the New View is incorrect. That evidence is also consistent with the New View because the New View posits that only permanent taxes on dividends are irrelevant. See Rosanne Altshuler, T. ScottNewlon & William C. Randolph, Do Repatriation Taxes Matter? Evidence from the Tax Returns of U.S. Multinationals, in The Effects of Taxation on Multinational Corporations 253, 256 (Martin Feldstein, James R. Hines Jr. & R. Glenn Hubbard eds., 1995). If corporations expect tax rates to change, the New View predicts that tax rates become relevant in deciding when to pay dividends because payments should be timed to take advantage of low rates. Thus, the empirical results may simply reflect the frequency of statutory rate changes in the U.S. and the ability of corporations to time dividend payments so that they will occur in periods when effective tax rates are low as a result of effective tax planning. 22. See e.g. J. Clifton Fleming Jr. & Robert J. Peroni, Exploring the Contours of a Proposed U.S. Exemption (Territorial) System, 109 Tax Notes 1557, 109 Tax Notes 1557, 1570 (Dec. 19, 2005); Staff ofJoint Comm. on Taxation, 106th Cong., 1st Sess., Description and Analysis of Present-Law Rules Relating to International Taxation, at 75 (Comm. Print 1999). 23. See Hartman, supra note 15, at 115, 116-17. [Vol. 8:3 Will US. Investment Go Abroad in a Territorial Tax View, like the Traditional View of dividends, considers a tax on dividends to be relevant to the decision whether new capital should be invested in a foreign subsidiary.24 This occurs because the value of equity received for the capital transfer is directly related to the tax assessed on repatriation of the subsidiary's earnings.25 The value of the investment in the subsidiary is the discounted present value of the subsidiary's expected after-tax income stream. Thus, the lower the tax assessed on the repatriation is, the greater the value of the investment will be.26 The adoption of an exemption system will increase the expected return from capital transfers to foreign subsidiaries in low-tax countries and as a result would encourage new investment, all other factors remaining the same. Similarly, an exemption system will encourage increased investment in branches located in low-tax countries since the retum on such investment would only be subject to the low tax rate. Clearly, the effects of a territorial tax on investment have to be viewed as part of a larger picture. It is not sufficient to merely observe that a territorial tax will increase dividends from subsidiaries in low-tax countries. Any benefits that are actually generated by a territorial tax may be offset by an increase in the excess burden on domestic production because of the incentive created for overseas investment. Similarly, the goal of increasing the competitiveness of multinationals in low-tax countries by adopting a territorial system is not a useful policy objective. It is always possible to make a business more competitive by reducing its tax burden. The picture is only complete when the impact of the tax preference on other activities and welfare is accounted for. To obtain a complete picture, this article will view the Report's proposal through application of traditional tax policy tools that examine efficiency effects of tax changes in order to assess possible welfare effects. The debate about efficiency in international tax has usually focused on concerns about capital export neutrality and capital import neutrality.27 Capital export neutrality ("CEN") requires that income from domestic and foreign investments be taxed at the same tax rate. Implementing CEN would require that foreign income be taxed immediately (i.e. there would be no deferral)28 and that there be an unlimited foreign tax credit.29 Capital import neutrality ("CIN"), which 24. Hartman, supra note 15, at 117, 119-20; Zodrow, supra note 18, at 497 and 503. 25. Id. 26. Id. 27. See e.g. Michael J. Graetz, The David R. Tillinghast Lecture: Taxing International Income: Inadequate Principles, Outdated Concepts, and Unsatisfactory Policies, 54 Tax L. Rev. 261, 270-71 (2001). 28.Id. at 271; Robert J. Peroni, Back to the Future: A Path to Progressive Reform of U.S. International Income Tax Rules, 51 U. Miami L. Rev. 975, 981(1997) 29. See e.g. Graetz, supra note 27, at 271; Stephen E. Shay, Clifton Fleming, Jr. & Robert J. Peroni, The David R. Tillinghast Lecture:"What's Source Got to Do With It?" Source Rules and U.S. International Taxation, 56 Tax L. Rev. 81, 108 (2002). 2007] Florida Tax Review is what a territorial tax system would seek to accomplish, requires that investments in foreign countries be taxed at the same rate as the rate applied to investments by residents in that country.30 This means that income earned in a foreign country would be exempt in the resident country. It is very difficult to achieve both CIN and CEN because all residence and source countries would have to exempt foreign income and apply the same tax rates to and have the same tax base for their resident income.3 Both CEN and CIN create efficiency distortions. CEN introduces a tax wedge between savings and consumption since savings are taxed in an income tax more heavily than consumption, but does not drive a tax wedge between domestic and foreign investment since both are taxed equally.32 In contrast, CIN drives a wedge between foreign and domestic investment because the income thereon is taxed differently. CIN does not drive a wedge between savings in the foreign country and consumption since the return on savings in the foreign country is not taxed in the resident country.33 The current U.S. system represents a compromise between CEN and CIN. CEN is not achieved because foreign income from active business operations is not taxed currently under Subpart F and the foreign tax credit is limited?4 CIN is also not achieved because dividends from foreign operations are taxed, although deferral can reduce significantly the effective rate of tax.35 A territorial tax will increase the excess burden on domestic investment since it will increase the tax wedge between domestic and foreign investment, i.e. it will increase the tax cost of domestic production in comparison to foreign production.36 At the same time, a territorial tax will decrease the excess burden 30. See e.g. Graetz, supra note 27, at 270-71. 31. See e.g. Graetz, supra note 27, at 272; Joint Comm. On Tax'n, 102d Cong., 1st Sess., Factors Affecting the International Competitiveness of the United States 5 (Comm. Print 1991). 32. Alberto Giovannini, Capital Taxation, Economic Policy 346, 366-67 (Oct. 1989). 33. Id. See Graetz, supra note 27, at 272-73. 34. Robert J. Peroni, supra note 28, at 975, 977-78. 35. Id. 36. The excess burden is a function of the elasticity of the compensated demand curve for the item being taxed and the square of the tax-exclusive tax rate. Jane Gravelle, The Economic Effects of Taxing Capital Income 30 (1994); John Creedy, The Excess Burden of Taxation and Why: It (Approximately) Quadruples When the Tax Rate Doubles, New Zealand Treasury Working Paper 3/29 p. 17 (2003). The elasticity of the demand curve is in turn a function of the willingness of the taxpayer to substitute another item for the item being taxed. The less willing a taxpayer is to substitute the item being taxed with another item, the less elastic the item is. A territorial tax increases the excess burden on domestic investment because it motivates the taxpayer to substitute foreign investment for domestic investment. The magnitude of the increase in excess burden will depend upon the relative substitutability of foreign investment for domestic investment (i.e. the elasticity of domestic investment). [Vol. 8:3 Will U.S. Investment Go Abroad in a Territorial Tax on savings since foreign income will be exempt. Whether there is a net efficiency gain or loss is determined by comparing the size of the decreased excess burden on savings to the increased burden on domestic investment. The difficulty in ascertaining this net benefit or cost is that theory cannot predict whether the decrease in excess burden for savings will be less than the increase of the excess burden for domestic investment.37 The magnitude of the excess burdens will depend upon the relative substitutability of consumption for savings and of foreign investment for domestic investment, both of which are empirical questions.3" The U.S. Treasury Department has suggested that CEN is preferable to CIN because CEN maximizes global39 and national welfare.40 This is based on the view that foreign investment is more readily substituted for domestic investment than consumption for savings. If this is correct, CEN would allocate investment among countries in the most efficient manner since the tax burden borne by such investments would be the same and saving would not be significantly affected. B. The Altshuler-Grubert Article Since theory cannot predict the net effect of a territorial tax on excess burdens, the issue whether the territorial tax will increase efficiency is an empirical question. As discussed above, the magnitude of the excess burden 37. Giovannini, supra note 32, at 367; Thomas Horst, A Note on the Optimal Taxation of International Investment Income, 94 Q.J. Econ. 793, 797 (1980). See Musgrave & Musgrave, supra note 11, at 451 (making this point in the context of comparing a consumption tax to an income tax); Congressional Budget Office, Revisiting the Individual Income Tax 46 (1983) (same); Gravelle, supra note 36, at 31 (same). 38. Giovannini, supra note 32, at 367; Horst, supra note 37, at 797. See supra note 36, for a discussion of how the excess burden is calculated. 39. Treasury Dep't., The Deferral of Income Earned Through U.S. Controlled Foreign Corporations: A Policy Study 26-36 (2000) (hereinafter "Treasury Subpart F Study"); Richard E. Caves, Multinational Enterprise and Economic Analysis 229-31 (1982). 40. Treasury Subpart F Study, supra note 39, at 36-42. 41. Id. at 30 n. 14,36-42; Graetz, supra note 27, at 272. The low substitutability for savings means that the response of savings to tax is inelastic and, therefore, the excess burden is low. See supra note 36, which discusses the manner in which excess burden is calculated. Not everyone would agree that a low elasticity for savings means that the excess burden is low. See Martin A. Feldstein, The Effect of Taxes on Efficiency and Growth, Tax Notes 679, 683 (May 8, 2006). Feldstein argues that the reduction in savings is not the relevant consideration in measuring the excess burden of a tax on savings, but rather that the relevant consideration is the reduction in future consumption that will occur as a result of the tax. In Feldstein's view, inelasticity in savings would be irrelevant to the excess burden created by a tax on savings. 20071 Florida Tax Review imposed on domestic investment by a territorial tax is a function of the extent to which foreign investment may be readily substituted for domestic investment.42 The Report relies heavily on the Altshuler-Grubert article, which is the only empirical study of the extent to which the U.S. tax system affects U.S. investment in low-tax countries,43 to conclude that a territorial tax will not encourage investment in low-tax countries by U.S. corporations." As discussed below, this reliance is misplaced since the results of Altshuler and Grubert's careful analysis are mixed. Altshuler and Grubert examine the potential effect of adopting an exemption system by analyzing a number of different aspects of foreign investment. Their approach can be divided into three broad categories. First, they calculate the effective tax rate on foreign investment by U.S. firms under the current global system and under the proposed territorial system to determine whether the territorial system would create an incentive to invest in low-tax countries. Their calculations suggest that foreign investment would not increase under a territorial system because the effective tax rates in a territorial system would not differ significantly from the current effective rates.45 Second, they utilize historic data to predict how U.S. firms would respond to a territorial system by examining the investment decisions of foreign firms already subject to a territorial system. This analysis yields mixed results. Third, they use historic data to determine the probability that U.S. multinational firms will increase investment in low-tax countries under a territorial system. Their analysis of U.S. multinational behavior indicates that foreign investment would increase in a territorial tax, although the magnitude of the response would be small. 42. See supra note 36, which discusses the principle that the excess burden is a function of the elasticity of the compensated demand curve for the item being taxed and the square of the tax exclusive tax rate. The elasticity of the demand curve is in turn a function of the willingness of the taxpayer to substitute another item for the item being taxed. 43. For studies that have analyzed the impact of the host country's tax system on the location of foreign investment, see e.g. Altshuler & Grubert, supra note 5, at 801; James R. Hines, Jr., Lessons from Behavioral Responses to International Taxation, 52 Nat'l Tax J. 309-13 (1999). 44. Whether such investment is a substitute for or complement to domestic investment is another important issue in determining the efficiency effects of a territorial tax that is not addressed in the Altshuler-Grubert article. See infra text accompanying notes 73-82, for further discussion. 45. For an earlier analysis that also argued that effective tax rates should not differ significantly between the current U.S. system and a territorial system, see Terrence R. Chorvat, Ending the Taxation of Foreign Business Income, 42 Ariz. L. Rev. 835, 843-44 (2000). [Vol. 8:3 Will U.S. Investment Go Abroad in a Territorial Tax 1. The Effective Tax Rate Determinations In their effective tax rate analysis, Altshuler and Grubert compare the effective tax rates that would apply to U.S. investment in low-tax countries if the U.S. adopted a territorial system to the rates that currently apply. They assume a territorial system similar to that proposed by the Report - dividends from a controlled foreign corporation conducting an active foreign business would not be taxed, but interest and royalties paid by such corporation would. They calculate that under the current U.S. system the effective tax rate on dividends of income earned in low-tax countries for firms that are "excess credit ' is equal to the effective rate that would apply to such companies under a territorial system.47 Thus, they argue a territorial system would not increase the incentive for excess-credit firms to invest in low-tax countries as compared to the current U.S. system. They also find that the effective tax rates on income earned in low-tax jurisdictions would increase in a territorial system for firms that are currently "excess limitation."'4 In reaching this surprising conclusion, they make two important assumptions, which will be described more fully below. Given that a territorial tax would not change the effective tax rate for excess-credit firms and would actually increase the rate for excess-limitation firms, they determine that a territorial system is unlikely to encourage more investment in low-tax jurisdictions as compared to the current system. 46. A firm is excess credit if its foreign tax payments exceed the amount that may be claimed as a foreign tax credit. In that situation, the firm is in the same tax posture as though the U.S. had adopted an exemption system. For example, if a U.S. taxpayer pays 1,000 of foreign tax on 4,000 of foreign income and the U.S. tax on that income is 800, the U.S. taxpayer will not owe any U.S. tax with respect to that income. This is the result that would have occurred had the U.S. adopted an exemption system. Paul R. McDaniel, Hugh J. Ault & James R. Repetti, Introduction to United States International Taxation 89 5th ed. (2005). 47. Altshuler & Grubert, supra note 5, at 798. 48. Id. at 790. A firm is excess limitation when its foreign tax payments are less than the U.S. limit on the foreign tax credit for such income. In this situation, a firm will pay U.S. income tax on the foreign income to the extent the U.S. tax exceeds the foreign tax. For example, assume that a firm pays 1000 of foreign tax on income of 4000, and that the U.S. tax on that income is 1500. After claiming the 1000 foreign tax credit, the taxpayer will still owe 500 in U.S. tax. The rationale for Altshuler and Grubert's surprising result is discussed, infra, at the text accompanying notes 49-52. 2007] Florida Tax Review Altshuler and Grubert's results are reproduced below. ALTSHULER AND GRUBERT TABLE 3 EFFECTIVE TAX RATES FOR INVESTMENT ABROAD IN A Low TAX COUNTRY Investment comprised of: All All 85 % tangible and tangible intangible 15 % intangible assets assets assets Dividend exemption 4.8% 35.0% 9.3% Current system (assuming 25% of firms in excess credit) 1.7 26.3 5.4 Excess limitation firms 0.7 35.0 5.8 Excess credit firms 4.8 0.0 4.1 Assumptions Statutory and effective tax rates: * the U.S. statutory tax rate is 35% * the host country statutory tax rate and effective tax rate is 7% Investment Financing: tangible capital receives economic depreciation allowances and no investment tax credits intangible capital generates royalty income, which is deductible in the host country but taxable in the United States "other" overhead expenses (expenses besides interest and R&D) account for 10% of the pre-tax required rate of retum (net depreciation) on capital marginal tangible investment is funded one-third with debt and two-thirds with equity the required after-tax rate of return on capital equals the real interest rate firms repatriate 7% of net host tax earnings on marginal tangible capital and gross-up dividends for the purpose of the foreign tax credit at 15% the deadweight loss from restricting dividend repatriations for firms in excess limitation is 1.7% of net host tax earnings on marginal tangible capital [Vol. 8:3 Will U.S. Investment Go Abroad in a Territorial Tax Interest and "other" overhead deductions: * Under the current system, firms in excess limitation deduct 50% of interest expense and 75% of "other" overhead expenses against the U.S. or other high-tax income. Firms in excess credit deduct 100% of interest expense at the 7% rate and lose the advantage of deducting overhead at the 35% rate. Under exemption, allocation rules require that all expenses be allocated against exempt income. Firms deduct 100% of interest expense at the 7% rate and lose the advantage of deducting overhead at the 35% rate. There are several interesting features in their results that merit further attention. As shown in column one, the aggregate current effective tax rate for a firm with excess credits investing in a low-tax country is calculated to be 4.8%. Altshuler and Grubert determine that the effective tax rate in a territorial system for such firms would be the same. This is expected since firms with excess credits are essentially exempt from U.S. tax on dividend repatriations. But, surprisingly, Altshuler and Grubert calculate that the current effective tax rate for firms that are excess limitation is lower than the rate that would exist in a territorial system. This is surprising because one would expect firms that are excess limitation to incur currently a higher tax liability than they would in a territorial system since excess limitation means that they will pay a U.S. tax on foreign income.49 Altshuler and Grubert make two significant assumptions that account for this unexpected result. First, they assume that firms in our current system are subject to an effective U.S. tax rate of only 3.3% on dividend repatriations because they can time the repatriation of their foreign subsidiaries' income to minimize tax.5" Second, they assume that excess-limitation firms currently are able to allocate 75% of their overhead expenses (other than interest and research and development expenses) to high-tax jurisdictions such as the U.S. and, as a result, generate significant tax savings from such deductions.5' In contrast, they assume that in a territorial system such firms would be required to allocate their overhead expenses to the exempt income, thereby generating no U.S. tax benefit.52 Altshuler and Grubert also predict that a territorial system will increase the tax burden on royalty income from licensing intangibles. Royalty income is taxable in our current global system, but since a territorial system would lack foreign tax credits for income not subject to U.S. tax, the credit would not be available in a territorial system to shelter royalty income. Their calculations in column two show that excess credit firms, which own a subsidiary whose assets consist entirely of intangibles and which receive 49. See supra note 48. 50. Altshuler & Grubert, supra note 5 at 797-98. 51. Id. at 795, 801. Firms that are excess credit are assumed not to deduct the overhead expenses against high-tax income because this would reduce the foreign tax credit. 52. Id. at 795. 2007] Florida Tax Review distributions from that subsidiary solely in the form of royalties, would move from a situation where none of their royalty income is currently taxed to one where all royalty income is taxed in a territorial system. Similarly, in column three, when they calculate the effective rates for firms that have an asset mix consisting of 85% tangible and 15% intangible (with the return on the intangibles being received in the form of royalties), they again determine that the effective tax rates in low-tax jurisdictions under the current system are less than for an exemption system. As a result, they conclude that moving to a territorial system is unlikely to encourage more investment in low-tax jurisdictions, as compared to the current system, since a territorial system will not reduce the tax burden.53 Altshuler and Grubert's effective tax rate calculations are very interesting and impressive. The calculations raise two questions, however, that need to be resolved. The first is the reasonableness of the assumption that firms that are excess limitation can deduct 75% of overhead expenses from high-tax income. Altshuler and Grubert do not explain the rationale for this assumption. Moreover, their article provides no sensitivity analysis of what would happen if that assumption were changed. It would be very helpful to see what the effective tax rate would be for firms that are excess limitation if they could only allocate a smaller amount of their overhead to high-tax income. The second and more important issue pertains to Altshuler and Grubert's use of an effective U.S. tax rate of only 3.3% on dividend repatriations in our current system. They calculate this low effective rate based of the assumption that firms can currently time the repatriation of their foreign subsidiaries' income to minimize tax.54 This is certainly true for existing multinational firms that have experience in managing foreign earnings and can cross credit or use credit carryovers to minimize tax. Firms lacking experience in managing dividend repatriations may believe, however, that they face an effective U.S. tax rate in our current system that is much higher than the 3.3% rate that Altshuler and Grubert calculate for experienced firms. As a result, the current U.S. tax scheme may present a much greater deterrent to new firms making foreign investments for the first time than for experienced firms. Stated another way, it is possible that the adoption of an exemption system will eliminate uncertainty about the tax burden, and as a result decrease the risk associated with foreign capital transfers. This reduction in tax risk may encourage firms that previously had little or no FDI to increase it. To see this, consider that firms experienced in FDI that are analyzing a new equity investment have a history of managing foreign earnings that enables them to anticipate fairly accurately the extent that they will be able to cross credit taxes in high and low jurisdictions and generate tax carry forwards. This history enables the experienced firms to calculate what the tax burden will 53. Id. at 790. 54. Id. at 797-98. [Vol. 8:3 Will US. Investment Go Abroad in a Territorial Tax be on the repatriation of foreign investments. In contrast, firms lacking experience in FDI are not able to predict as accurately what their foreign tax credit position will be when they make foreign investments and when such investments mature. Moreover, such firms would lack a sufficiently large pool of FDI that would enable them to cross credit and use other strategies to reduce U.S. tax. This uncertainty and inability to use tax-reducing devices increases the risk of foreign investments for the inexperienced firm in relation to domestic investments. This may on the margin deter foreign investment as compared to domestic investment since the firm will select investments that present the best risk-adjusted after-tax return.5 In comparing a foreign investment to a domestic investment that have equal after-tax returns that are not adjusted for tax risk, the inexperienced firm will select the domestic investment because its expected after-tax return will be higher than foreign investment with the less certain tax treatment. The adoption of an exemption system will eliminate this tax risk and, therefore, make foreign investment more attractive to firms that had previously avoided foreign investment, since tax planning will no longer be required to calculate the after-tax return of the foreign investment. The role that tax risk plays in making foreign investments is consistent with literature that has shown that large firms make more foreign investments than small firms because they can better handle the risk of over- seas investments.5 6 This view is also consistent with the observations that one of the deterrents to domestic companies in making foreign investments is the fixed costs of acquiring knowledge about "learning how things are done abroad."57 It is reasonable to expect that the lack of experience needed to predict the availability of credits to shelter repatriation on the part of new entrants into foreign markets would also act as a deterrent. 55. It might be argued that this uncertainty only exists if the company plans on repatriating the earnings. However, it seems likely that companies would plan to repatriate at least part of the earnings for investment in the United States since the United States is a strong market. See e.g. Hartman, supra note 15, at 115 (noting that 40% of foreign subsidiary earnings were paid to U.S. parents in 1980). Empirical evidence from 1992 suggests that the extent to which a U.S. parent repatriates its earnings from foreign subsidiaries is related to the tax rate of the foreign country. Grubert & Mutti found that foreign subsidiaries of U.S. parents in 1992 repatriated only 6.1% of their earnings when they operated in a country with a tax rate of less than 10%. In contrast, subsidiaries operating in countries with tax rates of more than 30% repatriated 53.9% of their earnings that year. Harry Grubert & John Mutti, Taxing International Business Income: Dividend Exemption versus the Current System 30 Table 2 (2001). 56. See, e.g., Thomas Horst, Firm and Industry Determinants of the Decision to Invest Abroad: An Empirical Study, 54 The Rev. of Econ. and Stat. 258,259 (1972); Bernard M. Wolf, Industrial Diversification and Internationalization: Some Empirical Evidence, 26 J. of Indus. Econ. 177, 179 (1977). 57. Neil M. Kay, Penrose and the Growth of Multinational Firms, 26 Managerial and Decision Econ. 99, 102 (2005); Caves, supra note 39, at 13. 2007] Florida Tax Review The result is that predicting the impact of a territorial system by determining effective tax rates is very complex. We need to consider firms that previously had no FDI. Even if a territorial system did not encourage mature firms to increase foreign investment, it might encourage firms, which had previously not made foreign investments, to do so for the first time. 2. Analysis of Historic Data: Foreign Investment in Low-Tax Countries by Companies Subject to a Territorial Tax As discussed, above, the effective tax rate calculations are not helpful in predicting the response of domestic investment to the adoption of a territorial tax because the effective tax rate is only determined for firms that have experience in making foreign investments. Moreover, the effective tax rate calculations failed to explain the rationale for the assumption that firms that are excess credit limitation can allocate 75% of their overhead expenses to high-tax income. In addition to the effective tax rate calculations, Altshuler and Grubert used historical data to determine whether a territorial system would increase FDI by looking to see whether FDI by companies resident in countries that already have an exemption system (Canada and Germany) differed from FDI by U.S. companies. Specifically, Altshuler and Grubert looked to see whether investments in low-tax jurisdictions in Asia (Singapore and Malaysia) and in Europe (Ireland) by Canadian and German companies differed from investments made by U.S. companies in those regions. If Canadian and German companies had more investment in low-tax countries than U.S. companies, that would suggest that U.S. companies would similarly increase investment in low- tax countries under a territorial system. The results, which are mixed, are reproduced below. ALTSHULER AND GRUBERT TABLE 1 U.S., German, and Canadian Foreign Direct Investment In Manufacturing in 1998 U.S. Germany Canada Asia Singapore and Malaysia as a share of total Asia 0.269 0.153 0.066 Europe Ireland as a share of European Union (except Germany) 0.067 0.016 0.170 Ratio of Ireland to U.K. 0.181 0.095 0.278 Sources: Survey of Current Business (Sept. 2000), Deutsche Bundesbank: Kapitalverflechtung mit dem Ausland (May 2000), and data released by request from Statistics Canada, Balance of Payments Division. [Vol. 8:3 Will US. Investment Go Abroad in a Territorial Tax Altshuler and Grubert conclude that the "cross-country comparison gives a mixed picture of how location incentives may change under dividend exemption."" They observed that in Asia, U.S. affiliates held a larger share of investment in low-tax countries than Germany and Canada. Almost 27% of manufacturing FDI of U.S. firms in Asia was located in Singapore and Malaysia in 1998. In contrast, the percentage for Germany was only 15% and for Canada it was under 7%. They interpret this as suggesting that exempting dividends from U.S. taxation may not induce a significant reallocation of investment across low-tax jurisdictions in Asia. 9 Altshuler and Grubert further note, however, that the "evidence from Europe . . . presents a more guarded prediction."6 The data for Germany suggests that a territorial system will not encourage increased U.S investment in low-tax countries since German affiliates hold a substantially smaller share of FDI in Ireland (as a share of their investment in the European Union) than U.S. affiliates: 1.6% versus 6.7%. The Canadian experience suggests the opposite, however. Canadian firms have significantly more FDI in Ireland than U.S. firms. Canadian investment in Ireland accounts for 17% of the stock of Canadian FDI in the European Union. In contrast, U.S. firms located only 6.7% of their European investment in Ireland. Further, the ratio of Canadian investment in Ireland relative to Canadian investment in Great Britain is 28% while it is only 18% for U.S. companies. Altshuler and Grubert conclude: Thus, the Canadian experience in Europe hints that dividend exemption may have some effect on the location decisions of the U.S. MNCs [multinational corporations]. Taken as a whole, however, the evidence from the FDI data presents a mixed picture.6' In summary, the comparison of FDI by countries with an exemption system (Canada and Germany) to FDI by the U.S. gives mixed results. There appears to be no difference between the two systems for investment in Asia. In Europe, however, there is a significant difference between Canada and the U.S. for FDI in low-tax countries. This suggests that at least in the case of Canada, a territorial tax encourages more investment in low-tax foreign countries. 58. Altshuler & Grubert, supra note 5, at 792. 59. Id. 60. Id. 61. Id. 20071 Florida Tax Review 3. Analysis of Historic Data: Foreign Investment in Low-Tax Countries by US. Companies with Excess Credits In the last part of their article, Altshuler and Grubert used 1996 tax return data for U.S. multinational corporations to predict the effect of adopting an exemption system. They sought to do this by asking whether corporations that did not expect to pay taxes on repatriations because they had significant foreign tax credit carryovers were more likely to invest in low-tax jurisdictions than in high-tax jurisdictions. A tendency by such corporations to invest in low- tax jurisdictions would suggest that the adoption of an exemption system would similarly encourage multinationals to invest in low-tax jurisdictions since having foreign tax carryforwards is somewhat equivalent to being exempt from tax. To test this they used regression analysis to determine whether companies with significant foreign tax credit carryovers had a higher probability of investing in low-tax countries than in high-tax countries. The dependent variable in their regression model was simply assigned the number one if a multinational has at least one subsidiary in a country or zero if it had no subsidiary in that country.62 The independent variables included various measures of the extent to which a multinational had significant excess foreign tax credits and the effective tax rates of the various countries in the data sample.63 They found that corporations not expecting to pay U.S. tax were more likely to have a subsidiary in low-tax jurisdictions than in high-tax countries, although the magnitude of the response was small.' They concluded that "[i]f firms without foreign tax credit carryforwards ...behave similarly under dividend exemption, there may be some reallocation of foreign direct investment to low-tax jurisdictions." ' The third part of Altshuler and Grubert's analysis raises several questions. First, as Altshuler and Grubert point out, there is the question of how firms without foreign tax credit carryovers would behave under an exemption system. Importantly, firms that do not have excess credits include many domestic businesses that have not yet made significant foreign investment. As discussed earlier,' both the New View and Traditional View posit that taxes on dividends affect decisions to transfer new capital to a foreign subsidiary. Thus, it is possible that firms have been discouraged from making foreign investment in the form of new capital transfers because of the tax due on repatriation. Elimination of that tax could induce firms to increase significantly their foreign investment in low-tax jurisdictions in the form of capital contributions or acquisitions. For example, the Task Force on International Tax Reform has 62. Altshuler & Grubert, supra note 5, at 803. 63. Id. at 804. 64. Altshuler & Grubert, supra note 5, at 807. 65. Id. 66. See supra text accompanying notes 23-26. [Vol 8:3 Will US. Investment Go Abroad in a Territorial Tax suggested that a territorial tax would encourage a U.S. firm that manufactures and sells all its products in the U.S. to relocate its manufacturing plant to Ireland and sell its product back to the U.S. so that it could benefit from the low Irish corporate tax and repatriate its earnings free of U.S. tax.67 The second problem with the regression results is that the regression model Altshuler and Grubert use does not take into account the magnitude of investment in each country. As a result, the regression results really provide no help in predicting how multinationals will respond to an exemption system. The dependent variable that Altshuler and Grubert employed (a value of one if the multinational had at least one subsidiary in a country and zero if not) does not reflect the amount of investment the multinational made in each country but only reflects that the multinational had a subsidiary there. Testing to see whether it is more probable that a multinational not expecting to pay U.S. tax will locate a subsidiary in a low-tax jurisdiction than in a high-tax jurisdiction tells us nothing about the relative amounts of investment in the various countries since the subsidiary might represent an investment of one billion dollars or one thousand dollars. Moreover, even if all subsidiaries were the same size, the test would still not be helpful because the dependent variable is assigned a value of one regardless of the number of subsidiaries in that country. For example, a U.S. multinational might have one subsidiary in Country A, 100 subsidiaries in Country B and no subsidiary in Country C, and the dependent variable would be one for Country A and also one for Country B and zero for Country C. The small response that Altshuler and Grubert found for the probability that multinational firms with excess credit will invest in low-tax jurisdictions is irrelevant since the magnitude of investment was not considered. Even if the regression had used the amount of investment as a dependent variable, the results still might be suspect. The data that Altshuler and Grubert examined showed the tax posture of U.S. multinationals in 1996 and the location of their subsidiaries in 1996. Presumably most of the investment in the subsidiaries occurred in years other than 1996. Accordingly, the tax posture of the multinationals in 1996 is not nearly as helpful as the tax position of the U.S. multinationals in the years in which the investment occurred. But even data about the tax posture of U.S. multinationals for the years in which FDI occurred may not be helpful. It is possible that a large portion of investment by multinationals in their foreign subsidiaries occurred in the form of retained eamings.68 In an earlier study, Grubert and Mutti observed that profitable controlled foreign corporations repatriated on average 67. Report of the Task Force on International Tax Reform, 59 Tax Law. 649, 723 (2006). 68. Hartman, supra note 15, at 115 (noting that in 1984 earnings retained by foreign subsidiaries accounted for approximately three quarters of direct foreign investment). 20071 Florida Tax Review only 6.1% of their after-tax earnings in low-tax countries in 1992.69 Under the New View of dividends, the tax on dividends to the parent is irrelevant to the decision whether the subsidiary's earnings should be retained because the tax will be incurred regardless of when the earnings are transferred to the parent.70 The only relevant consideration for retention of earnings by a mature subsidiary is whether the after-tax return on investment of the retained earnings would exceed the after-tax return that could be earned by the parent in the event the earnings were repatriated. 7' Thus, the tax posture of the parent may not have played a significant role in foreign investment to the extent it was funded through retained earnings. If the New Theory is an accurate description of dividend behavior,72 only new capital transfers would have been affected by the U.S. tax system. In summary, the regression analysis of the 1996 data does not provide very useful information about the likely response of U.S. firms to a territorial tax because the analysis fails to account for the magnitude of foreign investment in low-tax countries. The dependent variable used by Altshuler and Grubert merely registered that a U.S. firm had at least one subsidiary in each foreign country without measuring the amount of that investment. In addition, the analysis did not account for the response of transferring new capital overseas by U.S. firms that currently do not have any foreign investment. C. Does Foreign Investment Complement U.S. Investment? As shown above, the efficiency effects of a territorial tax are unclear because of the lack of conclusive evidence about the effect of such a tax system on foreign investment. Another uncertainty in the analysis is whether it is possible to view foreign investment as complementary to domestic investment. That is, if a territorial tax encourages FDI, is it nevertheless possible that an increase in FDI will increase U.S. domestic investment because it expands the market for U.S. exports and, as a result, the need for production in the U.S.?73 If this were the case, FDI would not be readily substitutable for domestic 69. Harry Grubert & John Mutti, supra note 55, at 30. 70. Hartman, supra note 15, at 115-17. 71. Id. 72. Empirical evidence supporting the applicability of the New View to foreign subsidiaries is mixed. Recent studies suggest that divident taxes affect the dividend behavior of foreign subsidiaries. See supra note 21. 73. See e.g., Martin A. Feldstein, The Effects of Outbound Foreign Direct Investment on the Domestic Capital Stock, in The Effects of Taxation on Multinational Corporations 43, 49 (Martin Feldstein, James R. Hines Jr., & R. Glenn Hubbard, eds.) (1995). [Vol. 8:3 Will US. Investment Go Abroad in a Territorial Tax investment and, as a result, the territorial tax would not significantly increase the excess burden on domestic investment.74 Unfortunately, the empirical results and the analysis of the empirical results are mixed. A report recently prepared by the Staff of the Joint Committee on Taxation 75 that addresses whether FDI is a substitute for U.S. investment, observes that overseas production increases U.S. production. The report stated:76 Generally, empirical studies find either no effect or a positive effect of overseas production in a host-country market on home-country exports to that country. One survey of the empirical literature reports that, on average, studies find one dollar of overseas production by U.S. affiliates generates $0.16 of exports from the United States. The evidence suggests that overseas production does displace certain types of domestic production as the parent firm shifts to more capital intensive and skill intensive domestic production. Note that this quote, however, does not focus on the impact of foreign investment on U.S. investment, but rather on the impact of foreign production on U.S. production. It is not clear how strong the connection is between U.S. production and U.S. investment. For example, an increase in U.S. production will not correspond to an increase in U.S. investment to the extent that there was excess capacity in U.S. production facilities. Empirical studies of the effect of FDI on domestic investment have been mixed. Analysis of investment data for several OECD countries from the 1970's and 1980's suggests that outbound FDI reduces domestic investment on approximately a dollar-for-dollar basis.77 The same results were obtained in analyzing data for OECD countries from the 74. See supra note 36, and the text accompanying supra note 38, for a discussion of the relationship between the substitutability of foreign investment for domestic investment and excess burden. 75. Staff of the Joint Committee on Taxation, 109th Cong., 2d Sess., The Impact of International Tax Reform: Background and Selected Issues Relating to U.S. International Tax Rules and the Competitiveness of U.S. Businesses 61-62 (Comm. Print, 2006). 76. Id. at 62 (citations omitted). The Joint Committee Staff cited as authority, Robert E. Lipsey, Home and Host Country Effects on FDI, 7-19 (Nat'l Bureau of Econ. Research, Working Paper No. 9293, 2002), and Robert E. Lipsey, Outward Direct Investment and the U.S. Economy, in The Effects of Taxation on Multinational Corporations (Martin Feldstein, James R. Hines, Jr. & R. Glenn Hubbard eds. 1995) 77. See e.g., Feldstein, supra note 73, at 57. 2007] Florida Tax Review 1980's and 1990's.78 Other studies, however, find no relationship,79 while others suggest that FDI and domestic investment may complement each other.80 Professors Desai, Foley and Hines found that when they analyzed data consisting of domestic and foreign investments by U.S. multinationals instead of aggregate national data for OECD countries, U.S. domestic investment by U.S. multinationals increased at the same time that FDI increased. They argue that this suggests that, at least in the case of U.S. multinationals, FDI complements U.S. domestic investment.8 The authors caution, however, that their results may be seriously biased by the omission of other important variables, which means there may in fact be no causal relationship between the observed simultaneous increases in FDI and domestic investment." The result is that the answer to the question whether FDI is a substitute for U.S. investment or a complement to it is unclear. More research is needed to determine what the impact of increased FDI will be on U.S. investment. m. CONCLUSION The Report justifies its recommendation for a territorial tax by asserting that a territorial system will eliminate the tax impediment to repatriating earnings and will make U.S. multinationals more competitive. A complete analysis requires that such benefits not be viewed in isolation. To determine whether a territorial tax scheme makes sense, it is necessary to determine whether other harmful impacts of a territorial system will outweigh these benefits. A territorial tax will increase the excess burden on domestic investment since it will increase the tax wedge between domestic and foreign investment, i.e. it will increase the tax cost of domestic in comparison to foreign investment. At the same time, the excess burden on savings will decrease since foreign income will be exempt. The issue for a territorial tax is whether the decreased excess burden on savings is less than the increased burden on domestic investment. Unfortunately, theory cannot predict whether the decrease in excess burden for savings will be less than the increase of the excess burden for domestic investment. The magnitude of the excess burden will depend upon 78. Mihir A. Desai, C. Fritz Foley & James R. Hines Jr., Foreign Direct Investment and the Domestic Capital Stock, 6-7 (Nat'l Bureau of Econ. Research, Wroking Paper No. 11075, 2005). 79. See e.g., David G. Harris, The Impact of U.S. Tax Law Revision on Multinational Corporations' Capital Location and Income-Shifting Decisions, 31 J. of Acct. Res. 111, 132-36 (1993). 80. Desai, Foley & Hines, supra note 78, at 7. 81. Id. 82. Id. at 9. [Vol. 8:3 Will U.S. Investment Go Abroad in a Territorial Tax the relative substitutability of consumption for savings and foreign investment for domestic investment, which are empirical questions. The Report's reliance on the Altshuler-Grubert article as support for its conclusion that a territorial tax will not increase investment in low-tax countries is of critical importance since it goes to the heart of the issue about the substitutability of foreign investment for domestic investment. This article argues, however, that the Report's reliance is misplaced because Altshuler and Grubert's empirical results are mixed. Moreover, this article suggests that to determine the effects of a territorial tax, it will be necessary to expand the empirical inquiry. Altshuler and Grubert's assumption that U.S. multinationals can allocate 75% of their overhead expenses to high-tax jurisdictions, such as the U.S., should be justified and subjected to sensitivity analysis. In addition, the effective tax rate calculations should also determine the effective tax rates under our current global system that apply to corporations making foreign investments for the first time since it is likely that such corporations face a much higher effective rate than corporations that already have significant FDI. Lastly, the dependent variable that Altshuler and Grubert use in their regression analysis to predict whether it is more likely that multinationals will invest in low-tax counties than in high-tax countries in a territorial system does not provide useful information because it does not account for the magnitude of FDI by U.S. multinationals. Altshuler and Grubert's dependent variable is assigned a value of one if the multinational has at least one subsidiary in a particular country and a value of zero if the multinational has none in that country. This does not reflect the amount of investment the multinational has made in those jurisdictions, but only reflects that the multinational had at least one subsidiary there. Testing to see whether it is more probable that a multinational will invest in low-tax jurisdictions than in high-tax jurisdictions by observing whether the multinational has at least one subsidiary in each such jurisdiction does not inform about the relative amounts of investment since the subsidiaries might represent an investment of one billion dollars or one thousand dollars. The result is that we face many uncertainties in regard to the effect of a territorial system. We do not have sufficient information about the impact of a territorial tax on foreign investment. Contrary to the Report's conclusion, the Altshuler-Grubert article does not support the conclusion that a territorial tax will not increase foreign investment. Indeed, theory predicts that a territorial tax will increase the incentive for U.S. multinationals and for firms with no prior foreign investment to increase capital transfers to foreign subsidiaries in low-tax countries. We also lack information about the impact of FDI on domestic investment. All this missing information seriously compromises the Report's conclusion that a territorial tax will not harm U.S. welfare because it prevents us from determining the increase in excess burden that such a tax will impose on domestic investment. 2007] Florida Tax Review The territorial tax proposal represents a significant variance from the norm governing the U.S. tax system that all income should be taxed.83 Such a variance should be justified only where it can be shown that the benefits outweigh the costs. Given Professor McDaniel's findings that a territorial tax would not be simpler than a global tax, and this article's analysis of the efficiency effects of the proposed tax, the Report has not made a case for a territorial tax. 83. See Michael J. McIntyre, Guidelines For Taxing International Capital Flows: The Legal Perspective, 46 Nat'l Tax J. 315, 321 (1993) (an exemption system is in effect a spending provision); Fleming, Peroni & Shay, supra note 15 at 344-46 (arguing that deferral of foreign income or an exemption of foreign income is a tax expenditure because it is a departure from the norm of taxing income and that such departure needs to be justified). [Vol. 8:3