By Martin A. Sullivan -- email@example.com
- We'll make sure that companies can't avoid paying tax on income they earn in the U.S. by pretending that they earned it in an overseas tax haven instead.
The Wall Street Journal,
Apr. 7, 2013
Why the change? First, House Ways and Means Committee Chair Dave Camp, R-Mich., released his discussion draft for a U.S. territorial system and an accompanying technical explanation on October 26, 2011. Camp intends to use anti-base-erosion provisions to raise revenue in any plan he will bring before his committee.
Second, Republicans did not win back the White House or the Senate in 2012, dashing hopes of a more business-friendly climate for tax reform. Third, the OECD, spurred by British Prime Minister David Cameron, has adopted a more aggressive tone and promised new efforts to combat base erosion and profit shifting. (Prior coverage: Tax Notes, Apr. 15, 2013, p. 259.)
The question is no longer whether, but how, we should limit base erosion. On the list of possibilities are formulary apportionment, expansion of thin capitalization rules, and strengthening transfer pricing rules. But the approach that seems to be getting the most attention in the United States is the imposition of extra U.S. tax when profits may have been inappropriately shifted.
There are six key anti-base-erosion proposals being discussed. Their primary goal is to serve as a backstop to transfer pricing rules that do not prevent inappropriate profit shifting out of the United States. A secondary goal, not as widely accepted, is that new rules should reduce tax bias in favor of foreign over domestic investment.
In their role as a backstop, any new base erosion rules come into play when indicia suggest inappropriate transfer pricing is likely to occur. The dictionary definition of indicia is "circumstances that point to the existence of a given fact as probable, but not certain." New anti-base-erosion rules will impose U.S. tax when circumstances point to the existence of inappropriate transfer pricing as probable, but not certain.
The indicia that can trigger new taxes generally fall into four categories: (1) low foreign tax; (2) high rate of profit; (3) intangibles-related income; and (4) an absence of business activity.
Part I. Proposals
1. Obama Excess Returns Proposal/Camp Option A. This proposal first appeared as part of Obama's fiscal 2011 budget and has been in all of his budgets since. The basic idea has remained unchanged: Excess income of a controlled foreign corporation that is related to transferred intangibles and is subject to low tax will be subject to immediate U.S. tax under subpart F.
A legislative draft of the proposal surfaced in September 2011. In October 2011 Camp incorporated nearly identical legislative language into his tax reform discussion draft, offering it as option A in his three alternative anti-base-erosion provisions. For purposes of defining intangible income subject to the new tax, the legislative drafts refer to the extensive list of intangibles in section 936(h)(3)(6).
One of the most striking features of option A is that the definition of subpart F income would include not only returns directly generated by transferred intangible assets, but also any income from the sale of a product or the provision of a service that exploits a transferred intangible. That greatly extends the scope of the provision in a somewhat arbitrary way, because tax can be triggered on sales and service income by the use of an intangible asset that may account for only a minute fraction of value added.
Under option A, excess returns from intangibles are calculated as gross income in excess of 150 percent of costs allocable and attributed to that income. And if those returns are subject to foreign tax of less than 10 percent, they are subpart F income. To prevent a sharp increase in overall tax when foreign tax rates rise above 10 percent, the percentage of excess returns subject to tax is linearly reduced from 100 to zero as the foreign effective tax rate increases from 10 percent to 15 percent. Also, option A includes a same-country exception that excludes income from subpart F if it is related to sales or services provided in the jurisdiction where the CFC is incorporated.
2. Camp Option B. Under option B, all foreign profits, except those satisfying a home-country exception, would be included in subpart F if the effective foreign tax rate on those profits exceeds 10 percent. There is no requirement that income subject to this new tax be generated by or related to an intangible asset. And all low-tax foreign profit is taxed under the new subpart F rules, regardless of whether it is considered excessive or otherwise. Subpart F income and the effective foreign tax rate are calculated separately in each country where CFCs are subject to tax.
Option B includes a home-country exception that is more stringent than the same-country exception in option A because it adds the requirement that the CFC maintain an office or fixed place of business in its home country. Under option B, income with an effective foreign tax rate (calculated on a per-country basis) of less than 10 percent is subject to full U.S. tax (less a foreign tax credit). Income with an effective foreign tax rate above 10 percent is subject to no immediate U.S. tax.
3. Camp Option C. Under this proposal, low-tax CFC intangible income is subject to tax under subpart F. This new subpart F income is divided into two categories: (1) income attributable to the exploitation of intangibles related to sales and services for customers outside the United States; and (2) income attributable to the exploitation of intangibles-related sales and services for customers in the United States. The first category of intangible income would be reduced by 40 percent before being taxed under subpart F. So, with a corporate rate of 25 percent, that income is effectively subject to tax at a 15 percent rate. To discourage "round tripping," the second category of income would receive the full force of subpart F and be taxed at 25 percent.
Unlike intangible income in option A, intangible income under option C does not include all income from sales or services employing intangible assets. Only income attributable to intangible assets is included in subpart F.
Income with a foreign effective tax rate of 13.5 percent or more is not subject to the new tax. Importantly, the Ways and Means explanation adds that "the relevant foreign taxes are those imposed on the particular items of intangible income" and that it is "intended that taxpayers not be permitted to engage in inappropriate planning, including the use of disregarded entities, to qualify for the high-tax exception."
Option C includes a tax incentive for domestic intangible income related to sales and services for customers outside the United States. From an economic perspective, the provision clearly is an export incentive. As such, there is a good chance that the proposal, if enacted, could be deemed a violation of international trade agreements, just like the extraterritorial income regime enacted in 2000 and repealed by the American Jobs Creation Act of 2004.
4. 'Option D.' So-called option D is not part of the Camp discussion draft but was developed by a group of companies. In the Camp discussion draft, the benefit of territorial taxation is effectuated by a 95 percent deduction for dividends received from a CFC. So, with a 25 percent corporate rate, foreign earnings would be subject to a 1.25 percent U.S. tax on repatriation. Unlike other proposals summarized in this article, option D generally would not require immediate U.S. taxation of some low-tax foreign income. Instead, it would reduce the dividends received deduction, which would apply to a broad category of foreign-source income. It would not be limited to intangible income and would not apply only to excess returns.
Under option D, CFCs with effective foreign tax rates between 7.5 percent and 15 percent would get an 85 percent dividends received deduction (resulting in a 3.75 percent U.S. tax on repatriation). CFCs with effective foreign tax rates below 7.5 percent would get a 75 percent dividends received deduction (resulting in a 6.25 percent U.S. tax on repatriation). A stacking rule would subject repatriated foreign income first to the least generous dividends received deduction that applies to a multinational's foreign-source income irrespective of which CFC actually repatriates. As under the general rules of the Camp plan, no FTCs would be allowed to reduce the repatriation tax.
Also, there is a special "deemed current dividend" rule that would subject CFC income to immediate U.S. tax at 40 percent of the U.S. rate (10 percent under the Camp plan) if the CFC is domiciled in a jurisdiction that is not a possession and that has no treaty or tax information exchange agreement with the United States. But those jurisdictions -- which do not include Ireland, Bermuda, Luxembourg, Switzerland, and the Cayman Islands -- account for a small share of low-taxed CFC income.
Figure 1. Cliff Effects Under Options A and B and Under Option C
(for intangibles supporting U.S. sales)
Note: Calculations exclude 1.25 percent tax due when income is repatriated.
This proposal has been presented to the staffs of the Ways and Means Committee, the Senate Finance Committee, and the Joint Committee on Taxation. Mike Reilly of Johnson & Johnson described it at the Tax Council Policy Institute's Symposium on the Taxation of Intangibles in Washington on February 14. (Prior coverage: Tax Notes, Feb. 18, 2013, p. 809.)
5. Obama Minimum Tax on Overseas Profits. This proposal was included in the president's framework for corporate tax reform, published in February 2012. It has not been included in any of the president's budgets. Described only in the most general terms, it would subject foreign profits to a new U.S. tax to make up the positive difference, if any, between the unspecified new minimum tax rate and the effective rate of foreign tax. The new tax would apply to CFC income whether or not it was associated with intangible assets and whether or not it was considered an excess return. The description does not include any home-country exception.
6. Grubert-Altshuler Minimum Tax With Expensing. In a paper to be presented at an April 26 Washington conference sponsored by the American Tax Policy Institute and the James A. Baker III Institute for Public Policy, Harry Grubert of Treasury and Rosanne Altshuler of Rutgers University examine various options for taxing foreign-source income. (See "Fixing The System: An Analysis of Alternative Proposals for the Reform of International Tax." http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2245128) The proposal they consider to have the most advantages relative to current law is a dividend exemption system that includes a minimum tax under which taxable income would be calculated with expensing of tangible investments instead of depreciation.
Economists like to divide the income from capital into normal and excess returns. Normal returns are the minimum necessary to motivate a company to invest. Excess returns (sometimes referred to as rents) do not change investment decisions (because by definition only normal returns are required). Companies may earn excess returns because they have some unique advantages (otherwise, market competition would bid those excess returns to zero). It turns out that the tax advantages of expensing are equivalent to exemption of normal profits from tax. So under the Grubert-Altshuler minimum tax with expensing, the normal returns on foreign-source income would be effectively exempt from U.S. tax. Only excess returns would be subject to U.S. tax.
Under the proposal, the minimum tax rate would be 15 percent. Minimum tax credits would be available for foreign tax up to 15 percent. Income in the denominator of the foreign effective tax rate would be calculated using depreciation rather than expensing. Intercompany dividends received that are taxed in other jurisdictions would be exempt from tax. There is no home-country exception. For tangible capital obtained in an acquisition, a special deduction would be allowed in lieu of expensing.
Table 1. Different Methods of Calculating Minimum Tax
A. Per-Country Basis
Foreign Foreign Min. Net U.S. Total
Country Income Tax Tax Rate Tax Rate Min. Tax Tax
A 100 5 5% 15% 10 15
B 100 10 10% 15% 5 15
C 100 20 20% 15% 0 20
D 100 25 25% 15% 0 25
Total 60 15 75
B. CFC-by-CFC Basis
Foreign Foreign Min. Net U.S. Total
Country Income Tax Tax Rate Tax Rate Min. Tax Tax
A and D 200 30 15% 15% 0 30
B and C 200 30 15% 15% 0 30
Total 60 0 60
C. Overall Basis
Foreign Foreign Min. Net U.S. Total
Country Income Tax Tax Rate Tax Rate Min. Tax Tax
A, B, C, 400 60 15% 15% 0 60
Grubert and Altshuler examine two versions of the minimum tax with expensing. Under the first, minimum tax would be calculated separately in each country where income is earned. That would block tax planning that would mix high-tax income of disregarded entities with income of low-tax CFCs. The second version of the tax would calculate each multinational's minimum tax liability on an aggregate basis across all its CFCs. In both cases, taxes and income would be averaged over five years to determine whether the minimum tax applies.
Part II. Issues
Under options A, B, and C, low foreign tax would trigger a new U.S. tax under subpart F but would also make FTCs available. So income below the low-tax threshold would be subject to (a) foreign tax plus (b) U.S. tax reduced by (c) U.S. FTCs. Items (a) and (c) offset each other, yielding a net overall tax rate equal to whatever U.S. rate subpart F triggers. Income that is not considered low-tax would be subject only to foreign tax.
The net result is that under those options, the peculiar situation arises that by increasing foreign tax, it is often possible to significantly reduce overall tax. When there is such a large decrease in total tax because of a very small increase in foreign tax, it is referred to as the cliff effect.
Figure 1 illustrates how total tax varies with foreign effective tax rates under the three options. Option B would have the largest cliff effect. Under option B, total tax would fall from 25 percent to 10.5 percent by increasing the foreign effective tax rate from 9.5 percent to 10.5 percent.
Under option C, there would be only a small cliff effect for foreign income from sales into markets outside the United States. But for intangible income from sales back into the United States, total tax would fall from 25 percent to 14 percent by increasing the foreign effective tax rate from 13 percent to 14 percent.
Option A would apply the subpart F tax on a sliding scale as the foreign effective tax rate declines from 15 percent to 10 percent. The cliff effect would not be so steep, but total tax could be cut by raising foreign tax. Income with a foreign effective tax rate of 9.5 percent would be subject to 25 percent total tax, while income with a foreign effective tax rate of 15.5 percent would be subject only to that 15.5 percent rate. (All those calculations exclude the 1.25 percent U.S. tax generally due on repatriation of any foreign income under Camp's 95 percent dividend exemption system.)
Those cliff effects would create a large incentive for foreign governments and U.S. multinationals to devise ways to increase foreign tax payments. If the rules allow, corporations will try to mix high-tax foreign income from one jurisdiction with low-tax foreign income from another jurisdiction. That could occur, for example, if effective foreign tax rates for purposes of the new tax are calculated for each CFC and check-the-box rules allow high-tax foreign income of a disregarded entity to be added to that of an otherwise low-tax CFC. Alternatively, a U.S. multinational could simply volunteer to pay more foreign tax to avoid an even larger addition to its U.S. tax bill.
Figure 2. Distribution of Effective Foreign Tax Rates
Of U.S. CFCs
Source: Harry Grubert and Rosanne Altshuler, "Fixing the System: An Analysis of Alternative Proposals for the Reform of International Tax," Table 3 (2013).
Fortunately, cliff effects can be easily eliminated by setting the rate of (total) tax equal to the foreign tax rate that triggers that tax -- in other words, by using a minimum tax mechanism. In that situation, an adjustable U.S. tax -- or a fixed U.S. tax less FTCs -- would make up the difference between the effective foreign tax and the minimum U.S. tax rate.
Per-Country, CFC, or Overall Tax Rate
A minimum tax can be calculated for each country in which a multinational generates profits, for each CFC one at a time, or for all of a multinational's CFC profits on an aggregate basis. To begin, let's just compare a per-country tax with an overall minimum tax.
If all profits in each country are taxed above the minimum rate, or if all profits in each country are taxed below the minimum rate, there is no difference between the two. (In the former case, there is no minimum tax, and in the latter case, the minimum tax applies to all income.) But when a multinational has profits in countries with foreign effective rates both above and below the minimum rate, minimum tax calculated on an overall basis results in less U.S. tax than a minimum tax calculated on a per-country basis. A simple example of that is shown in Table 1.
With the availability of check the box, a minimum tax calculated on a CFC-by-CFC basis could readily become the equivalent of an overall minimum tax. To minimize the minimum tax, multinationals could check the box on high-tax affiliates and place them under low-tax affiliates. In the Table 1 example of merging an affiliate in country A with an affiliate in country D and an affiliate in country B with an affiliate in country C, the same tax result is produced as an overall minimum tax. Given the relative cost and complexity of a minimum tax on a CFC-by-CFC basis, there is no apparent reason to ever favor it over an overall minimum tax.
How does a per-country minimum tax compare with an overall minimum tax as an anti-base-erosion measure? For a given rate of minimum tax, the per-country limitation is clearly superior in reducing the incentive for profit shifting out of the United States. Under a per-country minimum tax, the incentive to shift profits from the United States to a tax haven is reduced because the effective rate of tax on those profits will be 15 percent. Under an overall minimum tax, a multinational with an overall foreign tax rate above the minimum tax rate will still have a large incentive to shift income into a tax haven, as under current law. (And that advantage will continue until the multinational has shifted enough former U.S. profit to the haven that the overall average foreign rate falls to the minimum rate.)
Figure 3. Estimated Marginal Effective Tax Rate on
Investment in Jurisdiction With a 5 Percent Tax Rate
Source: Grubert and Altshuler, Table 1 (2013).
However, the per-country calculation has its own problems. It is more complex than an overall minimum tax. And in addition to reducing the incentive to shift profits out of the United States, it also reduces the incentive to shift profits from high-tax foreign countries to tax havens -- a feature that many consider to be an unwarranted impediment to multinational competitiveness.
To help decide between the two versions, Grubert and Altshuler calculated the distribution of foreign effective tax rates using 2006 Treasury data. They found that 46 percent of foreign-source income was subject to foreign tax of less than 10 percent, as shown in Figure 2. When they applied an overall minimum tax with a 15 percent rate to that distribution, they found it reduced the amount of CFC income subject to tax of 10 percent from 46 percent to 17 percent. That led them to conclude that while a per-country minimum tax is "more thorough" than the overall minimum tax, the latter "appears to be successful in targeting the companies that have the greatest opportunities for shifting income" and "deserves consideration." Deciding at what level minimum tax should be calculated may be one of the more difficult choices to be made in the design of anti-base-erosion rules.
Marginal Rate vs. Average Rate
It is unclear, especially given the complexity involved, why all the proposals focus exclusively on calculating an average effective foreign tax rate as a trigger for U.S. tax. The incentive for profit shifting is motivated by a difference in marginal rates. It is common in empirical economics to use an average tax rate as a proxy for marginal tax rates because it is easier to calculate. But for anti-base-erosion rules, that does not appear to be the case. On both conceptual and administrative grounds, it may make more sense to use a marginal rather than an average foreign effective tax rate to determine when U.S. tax is triggered.
So, for example, under a per-country minimum tax, minimum tax liability could be calculated in each jurisdiction by multiplying a calibrated minimum tax rate by foreign income from that jurisdiction. The calibrated minimum tax rate would be the excess (if any) of the U.S. minimum tax rate over that jurisdiction's effective marginal rate.
In most cases, the effective marginal foreign tax rate would be the statutory rate. In some cases in which broad relief like the section 199 deduction reduces the marginal tax rate broadly, adjustments to the statutory rate can be made. To promote certainty and reduce complexity, the IRS could calculate a safe harbor effective foreign marginal tax rate for each jurisdiction.
Camp's options A and C would limit application of their new subpart F tax to intangible income. Many commentators have expressed concern about the complexity of that requirement. In congressional testimony, Paul W. Oosterhuis of Skadden, Arps, Slate, Meagher & Flom LLP said options A and C would "require a problematic determination of what income is attributable to the exploitation of intangibles." At the same hearing, T. Timothy Tuerff of Deloitte Tax LLP put it this way:
This measurement of intangible income requires taxpayers to "unscramble the economic egg" by identifying the amount of revenue and expenses attributable to intangible property as compared to income of the CFC derived from a return on capital, services, manufacturing or marketing activities. Requiring segregation of the return from intellectual property will result in significant controversy during the examination process as taxpayers and the IRS attempt to subdivide the returns on transactions. Such a theoretical subdivision of income from a single transaction is considerably more complex than adjusting the transfer prices for actual transactions based on other, actual transactions among uncontrolled taxpayers.
Former Treasury International Tax Counsel Philip D. Morrison has written that options A and C "would create a level of complexity that will make compliance and administration nearly impossible." (See Philip D. Morrison, "Chairman Camp's Territorial Proposal and the Potential Expansion of Subpart F," 40 Tax Mgmt. Int'l J. 90 (Feb. 10, 2012).) The complexity associated with measuring intangible income is one of the main reasons the New York State Bar Association Tax Section favors Camp's option B over his options A and C.
A rule in which tax depends on defining intangible income will create a new dynamic. As much as possible, multinationals will want to shift (within the category of low-tax income) from intangible to non-intangible income. And given the elasticity of U.S. transfer pricing rules, they will have plenty of opportunity. Probably the largest area of potential controversy and uncertainty will be the dividing line between returns attributable to intangibles and returns for the assumption of risk that is assigned to principals in low-tax jurisdictions. Both of those sources of income are typically measured as residuals from total income after returns to more tangible activities (for example, distribution and manufacturing) are subtracted.
A division of profit between intangible and other income would require a tedious facts and circumstances determination. All the administrative and policy difficulties of transfer pricing analysis would be right back on our doorstep: the uncertainty, the cost, the controversy, and the likely prevalence of favorable outcomes for taxpayers because of their ability to devote more resources to the determination. Although the nature of the controversy is familiar, isolating intangible income would be unfamiliar territory for most taxpayers.
Excess Returns and Expensing
Another indicium of inappropriate profit shifting is a high rate of profitability. Like any indicium, even in concept, it is far from perfect. It is possible that a CFC that is earning high returns is lucky or that it has its own valuable intangibles. However, the data strongly suggest that in general, aggressive transfer pricing is a major cause of high profitability in low-tax affiliates. For example, the profitability of U.S. multinationals' Irish subsidiaries is consistently at levels two or three times those of their European counterparts. It is hard to believe any significant portion of that is because of the luck of the Irish or Irish research and development.
As mentioned earlier, economists practically equate the existence of persistently high profits with intangible assets. In their models, high rates of profit should be bid away by competition. If high rates of return on observable physical assets persist, that must be caused by assets that are unseen. (And that's why intangible income is usually measured as a residual rather than directly. It is the profit that is unexplained by tangible assets.)
How should excess returns be measured? The fiscal 2011 budget documents that introduced Obama's original proposal stated merely that an "excessive return" from a low-taxed transferred intangible would be subpart F income, but it did not define the term. In a budget briefing in 2010, a Treasury official said that for purposes of the budget's revenue estimate, excessive return was assumed to be any excess over a 30 percent rate of return on the transferred intangible.
In subsequent budgets, Treasury modified the definition of excess intangible income. Excess income subject to subpart F was gross intangible income minus 150 percent of costs allocated and apportioned to that income, including R&D expense and excluding taxes and interest. In congressional testimony, Oosterhuis, Tuerff, and David G. Noren of McDermott Will & Emery each expressed concern about the significant incentive effect the proposal would create to shift real business activities out of the United States.
Under the proposal, each dollar of costs incurred as the result of the migration of real business activity into low-tax jurisdictions would shelter 50 cents of additional income from the new subpart F rules. Because most costs will generate considerably less than a 50 percent return, the migration of those costs would exempt large amounts of low-tax income unrelated to costs used to compute excess returns.
The bad incentive effect alluded to in the congressional testimony can be illustrated with an example. Suppose a U.S. multinational generates $100 of profits in a jurisdiction where it pays tax at a 5 percent rate. With a new 15 percent minimum tax, there would be an additional $10 of U.S. tax. Now suppose the multinational puts new activity in that country that generates $50 of cost and $10 of additional profits (a 20 percent cost plus markup). With the option A excess return feature in place, the multinational pays an additional 50 cents of foreign tax on the new $10 of income. But the excess return calculation would also shield $25 of income from U.S. tax. Therefore, the new $10 of income generates no U.S. tax, and $15 of the original $100 is shielded from the minimum tax. The net result is that the action produces an additional $10 of income that also reduces the total tax bill from $15 ($5 foreign, $10 U.S.) to $14 ($5.50 foreign, $8.50 U.S.) -- a marginal effective tax rate of -10 percent.
Of our six proposals, the only other one to adopt an excess return approach is the Grubert-Altshuler minimum tax with expensing. As noted above, expensing in lieu of depreciation is the equivalent of a deduction from taxable profits of the normal return on capital. By simply expensing purchases of tangible capital, the proposal targets excess returns without all the complexity of option A.
But the advantage of a minimum tax with expensing over option A is not just simplification. It also seems to avoid the undesired incentive effects that got so much attention in congressional testimony. At first, that may seem a little too good to be true. After all, under the Grubert-Altshuler approach, the more real investment is shifted into low-tax countries, the more that foreign income will be exempt from their minimum tax (similar to what occurs in option A when expenses are pushed into low-tax locations).
The difference in incentive effects under the two proposals lies in the level of the threshold dividing excess from normal returns. The level of profit needed to be considered excess in option A will generally be much higher than that needed to be considered excess in the minimum tax with expensing. As in the example above, under option A, new investment would often shelter a lot more income from tax than the income it generates. In contrast, under the Grubert-Altshuler approach, new investment would usually shelter only a portion of new income (given that excess returns are usually positive).
The unwelcome incentive in option A for shifting real business activity could be reduced by lowering the markup rate. In his version of the excess return proposal, introduced July 11, 2011, as H.R. 2495, Rep. John F. Tierney, D-Mass., defined excess income as gross income that exceeds 115 percent of allocable costs.
For Grubert and Altshuler, an expensing feature is more than just a means to help identify situations in which inappropriate profit shifting may exist. They view it as a way of providing a tax advantage to foreign investment that is more likely to be threatened by international competition. They use reasoning that is well understood by economists but is probably unfamiliar to most tax practitioners. To economists, excess returns are indicators of intangible assets, and intangible assets give companies advantages that make them less susceptible to competition. Businesses generating no more than a normal return have no intangible assets and are highly vulnerable to competition.
Therefore, under a minimum tax with expensing, tax relief is provided to investment that is most likely to be adversely affected by the presence of new taxes. Although high-profit companies will not be pleased with new taxes on their excess returns, economic theory says they will not reduce their investment (because they are still earning at least a normal return at a low tax rate).
Business Activity Exceptions
Another possible indicium of profit shifting is the absence of business activity in a location where profits are booked. That is why anti-base-erosion proposals often include an exception for income generated in locations where substantial business activity is evident. In Camp's option B, a CFC must satisfy three tests to qualify for a home-country exception: (1) the income must arise from the active business conducted in the jurisdiction where the CFC is organized; (2) the CFC must maintain a fixed place of business in that country; and (3) the income must be derived from sales or services in that country.
While it is true that physical business activity indicates that income generated in a low-tax country has some business purpose and is not simply the byproduct of good tax planning, there are three reasons Congress should be cautious in granting active business exceptions.
First, they require complex facts and circumstances determinations.
Second, they tend to increase rather than reduce the overall amount of profit shifting. While meeting the requirements of a business activity test signals that not all profit has been inappropriately shifted, any tax relief for active business activity actually promotes rather than deters profit shifting. In practice, the presence of active business in a low-tax jurisdiction provides skilled transfer pricing specialists with lots of opportunity to shift profits there. An active business in a low-tax location is a profit magnet. Low tax is the incentive. The active business provides the facts and circumstances justification.
A third concern with any business activity test is that it can create a powerful tax incentive for foreign investment. The effects are similar to those under the excess return proposal: Any investment in a CFC that helps satisfy the requirement is likely to shelter a disproportionate amount of income from subpart F tax because shifted income will follow the real business activity. Thus, investment that qualifies for the exception can have negative effective tax rates.
One of the few things in international tax reform that everybody seems to agree on is that waiting to impose tax until foreign profits are repatriated creates a highly problematic lockout effect. In fact, Grubert and Altshuler report in their paper the results of new research showing that the economic cost of lockout is significantly larger than previous estimates suggest.
Lockout is the chief cause of U.S. multinationals keeping nearly $2 trillion of accumulated foreign profits outside the United States. What U.S. tax should be due on those previously earned profits is a wide-open question. (The Camp discussion draft taxes all of it over 10 years at 5.25 percent, whether or not it is repatriated.) Any major overhaul of U.S. international tax rules should eliminate the lockout effect either by taxing foreign profits as earned or not at all.
The Camp discussion draft would retain a small lockout effect because 5 percent of foreign profits would be subject to U.S. tax when repatriated. (That 5 percent "haircut" is a crude substitute for interest allocation rules -- not present in the discussion draft -- that would deny a U.S. deduction for interest costs allocable to exempt foreign profits.) A similar feature is part of the territorial systems in Japan, Germany, and France. Nobody believes a 1.25 percent tax will keep multinationals from bringing profits home, but why not just tax the 5 percent currently and get rid of all the vestiges of the lockout effect? In his congressional testimony, Noren suggested that Congress consider that the 5 percent of nonexempt income be taxed on a current basis rather than allowing the tax to be deferred.
Option D leans in the other direction. Instead of using subpart F or imposing a new minimum tax, it would penalize low-tax foreign income with a reduced dividends received deduction. Option D would increase the effective repatriation tax rate from 1.25 percent to 3.75 percent for CFCs with effective foreign tax rates below 7.5 percent, and to 6.25 percent for CFCs with effective foreign tax rates between 7.5 and 15 percent.
And so although it would still be much less than under current law, with option D the lockout effect is larger than it would otherwise be under the Camp plan. Advocates of the plan say that flexibility is needed to deploy capital outside the United States where their greatest growth potential lies and that their foreign competitors are headquartered in countries with territorial systems that likewise are not subject to anti-base-erosion taxes until they repatriate income. They also cite estimates, based on data from the 2004 tax holiday and analyses by the JCT, that suggest repatriations would not be significantly impeded by low levels of repatriation tax.
Part III. Conclusion
Many features of current proposals to prevent base erosion are problematic.
Limiting the application of anti-base-erosion rules to intangible income would spawn a whole new class of troublesome transfer pricing disputes, with taxpayers and the IRS battling on a case-by-case basis over the fuzzy dividing line between intangible income subject to the new tax and other income that is exempt.
Unless exceptions for active business income and normal profits are constructed narrowly, those provisions would have the unintended effect of creating large marginal incentives for multinationals to shift investment abroad.
Cliff effects provide a large incentive for taxpayers to increase their foreign taxes.
Allowing some tax to be deferred until repatriated means U.S. international tax rules would retain a small lockout effect.
Given those administrative and economic problems, any anti-base-erosion rules should avoid cliff effects, lockout effects, intangible income definitions, high thresholds for taxable excess returns, and home-country exceptions under which business activity is required. Of the six proposals evaluated here, the one that best fits the bill is the Grubert-Altshuler minimum tax with expensing. That proposal is also relatively easy to administer, and despite the absence of some indicia found in other proposals, it makes clever use of economic reasoning to keep the minimum tax targeted on intangible income and on situations in which inappropriate transfer pricing is probable.
Nevertheless, some will likely argue that their approach, like Obama's minimum tax proposal, is poorly targeted to transfer pricing problems and is instead primarily geared toward raising taxes on foreign investment. In other words, the proposal is really about moving the United States back in the direction of capital export neutrality.
With any anti-base-erosion rule, the move in that direction is unavoidable for the simple reason that taxes on foreign investment will increase. Yes, the old trade-off between promoting multinational competitiveness and promoting neutrality between foreign and domestic investment must come into play. And as usual with trade-offs, the absence of a clearly superior approach means there are legitimate differences of opinion about how far to lean in either direction.
One fact, however, that is too often ignored in the debate about the direction of international tax policy is the starting point. Right now, effective marginal tax rates on foreign investment are often not just low, they are negative. That is primarily because of the significant profit shifting that follows investment into a low-tax country. For example, the tax incentive for investing in Ireland results not only from the 12.5 percent rate of tax on income from investment there, but also from a large reduction of taxes on profits formerly assigned to high-tax locations, including the United States. When the reduction in non-Irish taxes exceeds the 12.5 percent Irish tax, the marginal effective rate of tax is negative. Tax may be further reduced by shifting profits out of Ireland to a zero-tax jurisdiction.
In their paper, Grubert and Altshuler use a simulation model to calculate the marginal effective tax rates under current law and under various reform options on investment in a jurisdiction with a 5 percent tax rate. Their results are shown in Figure 3. Under current law, the marginal effective tax rate on foreign investment is -23.6 percent. In other words, an investment in a country with a 5 percent statutory rate that generates $100 of properly measured profits results in an overall reduction in multinational tax of $23.60. Grubert and Altshuler's proposal for a per-country minimum tax with expensing would raise the effective tax rate to 5.6 percent for normal returns and to -4.4 percent for excess returns.
At a time when revenue raisers -- whether to pay for a territorial system, cut the corporate rate, or reduce the federal deficit -- are in short supply, negative effective tax rates for foreign investment are an extravagance we can't afford.