Keywords: GLAM 2015-002, tax rate, branch rule

With the promulgation of the check-the-box regulations and the proliferation of branches (for U.S. tax purposes), a number of U.S. international tax provisions gained significantly in importance, e.g., the loss recapture rules of §367(a) and §904(f),1 accounting for foreign currency gains and losses of a branch under §987, and the dual consolidated loss regime of §1503(d). Another provision that gained significantly in importance with the proliferation of check-the-box branches is §954(d)(2), setting forth the so-called ''branch rules'' for purposes of the foreign base company sales income (FBCSI) Subpart F provisions.

In very simple terms, the branch rules (i.e., the ''sales branch rule'' and the ''manufacturing branch rule'') introduced by §954(d)(2) and the regulations thereunder prevent taxpayers from avoiding FBCSI by using branches (instead of controlled foreign corporations (CFCs)) for sales or manufacturing activities. Central to the branch rules is the ''tax rate disparity test.''

Notwithstanding its critical importance, up until recently there was a paucity of published guidance regarding the tax rate disparity test; the Internal Revenue Service (''Service'') several years ago published three private letter rulings (PLR 200942034, PLR 200945036, and PLR 201002024), and that was pretty much it. That said, on February 13, 2015, the Service published GLAM 2015-002 (the ''GLAM''). The GLAM provides a five-step calculation to test for tax rate disparity. Specifically, the five-step calculation uses the sales taxable income (determined as if earned under the laws of the non-sales jurisdiction) as the common denominator (the ''hypothetical tax base'') to calculate the ''actual effective rate of tax'' and the ''hypothetical effective rate of tax.''

Arguably, the GLAM's five-step calculation provides a fair basis for testing for tax rate disparity, but two questions come to mind with respect to its test:

  1. Is the proposed five-step calculation supported by the law? and
  2. Even if supported by the law, just how practical is it to implement?

After reviewing the FBCSI branch rules and the GLAM's five-step calculation, this article provides some thoughts on those two questions.

FBCSI; THE BRANCH RULES

The FBCSI rules were enacted to prevent a U.S. taxpayer or its foreign subsidiary from artificially separating manufacturing and sales to reduce taxation on its sales income.2 Under §954(d)(1), FBCSI generally includes certain income derived in connection with the purchase of personal property from a related person and its sale to any person, the purchase of personal property from any person and its sale to a related person, or the purchase or sale of personal property on behalf of a related person. However, such income is not FBCSI if the personal property is manufactured by the CFC (the ''CFC manufacturing exception''), is manufactured by anyone within the CFC's country of organization (the ''same-country manufacturing exception''), or is sold for use in such country (the ''same-country use exception'').3

The branch rules provide that when a CFC carries on purchasing, selling, or manufacturing activities through a branch located outside of the CFC's country of incorporation, and the use of such branch has substantially the same effect as if such branch were a wholly owned subsidiary of the CFC, the FBCSI rules of §954(d)(1) will apply, to treat such branch as a wholly owned subsidiary of the CFC.4

Absent the ''sales branch rule,'' a CFC that manufactures a product in its country of incorporation and sells that product through a sales branch in a low- or no-tax jurisdiction would not have FBCSI due to the CFC manufacturing exception. Similarly, absent the ''manufacturing branch rule,'' a CFC that manufactures property through a branch or similar establishment and sells that property through the remainder of the CFC located in a low- or no-tax jurisdiction would not have FBCSI due to the CFC manufacturing exception.

In order to determine whether the use of a branch has substantially the same effect as if such branch were a wholly owned subsidiary of the CFC (and, thus, whether it will be treated as such for FBCSI purposes), the regulations under §954(d)(2) employ a tax rate disparity test.

The tax rate disparity test as it applies to a sales branch is set forth in Reg. §1.954-3(b)(1)(i)(b):

The use of the branch or similar establishment for such activities will be considered to have substantially the same tax effect as if it were a wholly owned subsidiary corporation of the controlled foreign corporation if [the income derived by the branch or similar establishment from the purchase or sale of personal property on behalf of a related person] is, by statute, treaty obligation, or otherwise, taxed in the year when earned at an effective rate of tax [the ''actual effective rate of tax''] that is less than 90 percent of, and at least 5 percentage points less than, the effective rate of tax [the ''hypothetical effective rate of tax''] which would apply to such income under the laws of the country in which the controlled foreign corporation is created or organized, if, under the laws of such country, the entire income of the controlled foreign corporation were considered derived by the controlled foreign corporation from sources within such country from doing business through a permanent establishment therein, received in such country, and allocable to such permanent establishment, and the corporation were managed and controlled in such country.

The tax rate disparity test as it applies to a manufacturing branch is set forth in Reg. §1.954- 3(b)(1)(ii)(b):

The use of the branch or similar establishment for such activities will be considered to have substantially the same tax effect as if it were a wholly owned subsidiary corporation of the controlled foreign corporation if [the income derived by the remainder of the controlled foreign corporation from the purchase or sale of personal property on behalf of a related person] is, by statute, treaty obligation, or otherwise, taxed in the year when earned at an effective rate of tax [the ''actual effective rate of tax''] that is less than 90 percent of, and at least 5 percentage points less than, the effective rate of tax [the ''hypothetical effective rate of tax''] which would apply to such income under the laws of the country in which the branch or similar establishment is located, if, under the laws of such country, the entire income of the controlled foreign corporation were considered derived by the controlled foreign corporation from sources within such country from doing business through a permanent establishment therein, received in such country, and allocable to such permanent establishment, and the corporation were created or organized under the laws of, and managed and controlled in, such country.

As such, with respect to sales income derived in connection with property manufactured by the CFC or a branch of the CFC, the regulations under §954(d)(2) determine whether there is tax rate disparity by comparing what the effective rate of tax is on the sales income come in the sales jurisdiction (the ''actual ERT'') with what the effective rate of tax on the sales income would have been if the sales had occurred in the manufacturing jurisdiction (the ''hypothetical ERT'').

THE GLAM AND THE FIVE STEPS OF THE TAX RATE DISPARITY TEST

In GLAM 2015-002, the Service sets out five steps for the calculation of the tax rate disparity test:

Step One: Determine the relevant income upon which the test will be based.

Step Two: Determine the actual tax imposed on that income.

Step Three: Determine the hypothetical tax base.

Step Four: Determine the hypothetical tax which would have been imposed on the income in the manufacturing location.

Step Five: Compare the actual ERT with the hypothetical ERT.

As such, the tax rate disparity test first requires the identification of the relevant income on which the tax would be imposed.5 In the case of income derived by a branch from the sale of property manufactured by the remainder of the CFC, the regulations require an allocation to the branch of the gross income derived in connection with the purchase or sale of property to, from, or on behalf of a related person (the ''TRD Gross Income'').6

Second, the actual tax paid with respect to the TRD Gross Income, if any, must be determined. If the sales branch earned income other than TRD Gross Income and paid or incurred income tax with respect to its entire income, the amount paid or incurred with respect to the TRD Gross Income must be separately determined.

Third, the hypothetical tax base must be determined by calculating the amount of TRD Gross Income that hypothetically would be subject to income tax in the CFC's jurisdiction. In this regard, the Service believes that, for the comparison to be meaningful, a common hypothetical tax base must be used to calculate both the actual and hypothetical ERTs. Further, the Service understands that the relevant tax laws to be taken into account in determining such common hypothetical tax base are those of the manufacturing jurisdiction.7 Thus, to determine the hypothetical tax base, the TRD Gross Income is reduced by exclusions and deductions that would be permitted under the laws of the country in which the property is manufactured (regardless of whether those exclusions and deductions would be allowed for U.S. income tax purposes, or under the laws of the sales jurisdiction), taking into account the source of income and permanent establishment assumptions in Reg. §1.954-3(b)(1)(i)(b). This requirement to use one common hypothetical tax base (i.e., sales taxable income determined under the laws of the manufacturing jurisdiction) is the crux of the GLAM and the point that raises two significant questions that are addressed below.

The fourth step is to calculate the hypothetical tax by multiplying the hypothetical tax base by the applicable marginal tax rate(s) in the manufacturing jurisdiction.

Under the fifth and final step, the hypothetical tax and the actual tax paid on the TRD Gross Income are each divided by the common hypothetical tax base, which results in the ERTs that will be compared for purposes of determining whether there is or is not tax rate disparity.

The GLAM concludes this five-step analysis with a short, but relevant, footnote indicating that once it has been determined that tax rate disparity exists, U.S. tax principles apply to determine the correct amount of net income attributable to the branch and whether any exceptions to FBCSI apply.8

The GLAM includes an example of the application of the five-step analysis to a simple set of facts.

Under these facts, CFC, incorporated in Country B, manufactures Product X in its country of incorporation. DE is the wholly owned subsidiary of CFC and has elected to be treated as a disregarded entity of CFC for U.S. income tax purposes. DE is located in Country A and derives 100x euros of commission income in connection with the sale of Product X by CFC to unrelated customers located outside of Country A and Country B. DE incurs 30x euros of sales expenses related to the sale of Product X. CFC has no other income that would constitute foreign base company income under §954.

Country A and Country B both impose a 20% statutory rate of tax on sales income. Country A allows DE to exclude half of its income from the sale of products manufactured and sold for use outside of Country A. Country B does not tax DE's sales income until it is remitted to CFC as a dividend. Both Country A and Country B would allow a 30x euros deduction for the sales expenses.

First, as noted above, TRD Gross Income must be determined. Because DE derived 100x euros of gross income in connection with the sale of Product X, the TRD Gross Income here is 100x euros.

Second, the actual tax paid or incurred in Country A must be determined. Under these facts, the actual tax paid or incurred in Country A is 4x euros, calculated as follows:

TRD Gross Income: 100x euros
Exclusion: (50x euros)
Sales expenses: (30x euros)
Taxable income: 20x euros
Statutory tax rate: 20%
Country A tax: 4x euros

Third, the hypothetical tax base must be determined. Here, the hypothetical tax base is 70x euros, calculated by taking the 100x euros of TRD Gross Income less the 30x euros of sales expenses that are allocable and apportionable to the TRD Gross Income under Country B's tax laws.

Fourth, the hypothetical tax that would have been incurred if the TRD Gross had been derived in Country B is determined. The hypothetical tax that would apply in Country B is 14x euros, calculated as follows:

TRD Gross Income: 100x euros
Sales expenses: (30x euros)
Hypothetical tax base: 70x euros
Statutory tax rate: 20%
Country B tax: 14x euros

Finally, the actual tax and the hypothetical tax are each divided by the hypothetical tax base (determined in the third step to be 70x euros), to arrive at the ERTs that are compared to determine whether there is tax rate disparity between DE and the remainder of the CFC:

Actual ERT (Country A): 4x euros/70x euros = 5.71%

Hypothetical ERT (Country B): 14x euros/70x euros = 20%

So, on these facts, the actual ERT is 5.71%, and the hypothetical ERT is 20%. Therefore, the actual ERT in Country A (the sales jurisdiction) is less than 90% (28.55%) of, and at least 5 (14.29) percentage points below, the hypothetical ERT in Country B (the manufacturing jurisdiction). Thus, there is tax rate disparity between DE and the remainder of the CFC and, pursuant to Reg. §1.954-3(b)(1)(i)(b), DE is treated as a wholly owned subsidiary of CFC under §954(d)(2).

THE QUESTIONS RAISED BY THE GLAM

Is the Five-Step Calculation Supported by the Law?

From a policy perspective, it could be argued that the requirement to use a common hypothetical tax base is driven by a valid concern. As the Service noted in the GLAM, computing the actual and hypothetical ERTs with respect to dissimilar tax bases could be interpreted as contrary to the legislative purposes of §954(d). In that regard, the use of the tax base in the sales jurisdiction to calculate the actual ERT (while the hypothetical tax base in the manufacturing jurisdiction is used to determine the hypothetical ERT) would provide an incentive to shift income to a sales jurisdiction that, even with a high statutory tax rate, grants exclusions and deductions to derive a smaller tax base.9

Even if the Service's concern is understandable, query whether the use of a common denominator in the tax rate disparity test is supported by the statute or the regulations.

The Service cites to Reg. §1.954-3(b)(2)(i)(e) as support for the proposition that the tax rate disparity test requires the use of a common tax base, and that such tax base must be determined under the laws of the manufacturing jurisdiction. Specifically, Reg. §1.954-3(b)(2)(i)(i) provides that ''[t]ax determinations shall be made by taking into account only the income, war profits, excess profits, or similar tax laws (or the absence of such laws) of the countries involved'' (emphasis added).

It is unclear how the Service concludes from this provision that both the actual and the hypothetical ERTs shall be determined in accordance with the laws of one same country (i.e., the manufacturing jurisdiction). If anything, Reg. §1.954-3(b)(2)(i)(e) appears to suggest the opposite — that the actual ERT is calculated under the laws of the sales jurisdiction and the hypothetical ERT is calculated under the laws of the manufacturing jurisdiction.

Further, as noted above, Reg. §1.954-3(b)(1)(i)(b) provides that:

The use of the branch or similar establishment for such activities will be considered to have substantially the same tax effect as if it were a wholly owned subsidiary corporation of the controlled foreign corporation if the income allocated to the branch or similar establishment under the immediately preceding sentence is, by statute, treaty obligation, or otherwise, taxed in the year when earned at an effective tax rate that is less than 90 percent of, and at least 5 percentage points less than. . . . [Emphasis added.]

This regulation refers to the effective tax rate at which the sales income of the branch is taxed by statute, treaty obligation, or otherwise in the sales jurisdiction (as opposed to the effective tax rate at which the sales income of the branch would be taxed if the tax base were determined in accordance with the statutes or treaties of the manufacturing jurisdiction).

It should also be noted that the Service's position on this point contradicts certain assertions from its prior branch rule guidance. In PLR 200942034, the Service held that the expenses of the sales branch had to be allocated and apportioned in accordance with the laws of the sales jurisdiction for purposes of the tax rate disparity test.10 Similarly, in PLR 200945036 the Service held that ''with regard to DE 2 [the sales branch]. . . for purposes of applying the tax rate disparity test in Reg. §1.954-3(b)(1)(ii)(b) to DE 2 with respect to DE 2's sales of products, the determination of FBCSI and the effective rates of tax applied thereto are determined solely under Country 2 [the sales jurisdiction] tax law principles as and to the extent modified by the Country 2 advance pricing agreement'' (emphasis added).

Is the Five-Step Calculation Practical To Implement?

Assuming, arguendo, that the use of that common hypothetical tax base is supported by the law, the question then arises as to its practicability.

In the GLAM's example, it is simple to calculate the hypothetical tax base in accordance with the laws of the manufacturing jurisdiction. The TRD Gross Income of the sales branch was 100x euros and we are told that the manufacturing jurisdiction would allow 30x euros as a deduction for sales expenses (but would not exclude half of the sales income, as the sales jurisdiction did). Under this limited set of facts, the calculation is straightforward. The hypothetical tax base is 70x euros. Unfortunately, usual scenarios may not be straightforward, and the differences between the tax laws of two different jurisdictions may be many and significant.

Taken to the extreme, the taxpayer would need to undertake the complex exercise of recalculating taxable income taking into account the thin capitalization and transfer pricing rules of the manufacturing country. Presumably, the taxpayer would also need to reassess the deductibility of interest on any hybrid instruments that, while treated as debt in the sales jurisdiction, may be characterized as equity in the manufacturing country.

Moreover, if the manufacturing jurisdiction provides certain tax holidays or benefits for companies located in certain regions of the country, should the taxpayer assume that the hypothetical sales company would have been located in such beneficial regions? If the manufacturing jurisdiction has the practice of granting rulings or Advance Pricing Agreements, should the taxpayer assume that the hypothetical sales company would have obtained such a ruling or agreement on terms consistent with those granted to similar businesses? How would the taxpayer determine an applicable deduction for state and local taxes when calculating the hypothetical tax base if different state and local taxes would have applied depending on where the sales company would have been located within the manufacturing country? These are just a handful of the imaginable issues that arise when calculating the hypothetical tax base under the laws of the manufacturing jurisdiction.

Aware of these complexities, the GLAM states that minor differences in the deductions allowed in the two jurisdictions (or in the timing of the deductions) that do not materially affect the analysis may be ignored in appropriate cases.11 Query as to which differences would be immaterial enough to be ignored.

All that said, there is one lingering practical concern: How do U.S. tax professionals prepare the calculation? Presumably, the sales income calculation will need to be sent to tax professionals in the manufacturing jurisdiction in order for the sales income to be recalculated under the manufacturing jurisdiction tax laws.

CONCLUSION

As noted above, the GLAM provides a five-step calculation to test for tax rate disparity. Specifically, the five-step calculation uses the sales taxable income (determined as if earned under the laws of the nonsales jurisdiction) as the common denominator (the ''hypothetical tax base'') to calculate the ''actual effective rate of tax'' and the ''hypothetical effective rate of tax.''

Arguably, the GLAM's five-step calculation provides a fair basis for testing for tax rate disparity, but two questions come to mind with respect to its test:

  1. Is the proposed five-step calculation supported by the law? and
  2. Even if supported by the law, just how practical is it to implement?

Although from a policy perspective it could be argued that the requirement to use a common hypothetical tax base is driven by a valid concern, it is unclear whether the use of a common denominator is actually supported by the statute or the regulations. Further, the Service's position on this point appears to contradict certain assertions from its prior published branch rule guidance.

Finally, even if the use of a common hypothetical tax base were supported by the law, the question then arises as to the practicability of the calculation proposed by the Service. The fiction of calculating the hypothetical tax base under the laws of the manufacturing jurisdiction opens the door to a number of puzzling questions. In addition, it raises a practical concern: how do U.S. tax professionals prepare the calculation? Presumably, it will be necessary in numerous instances to seek assistance from foreign tax professionals in the manufacturing jurisdiction, adding a layer of complexity to an already complex analysis.

Originally published by Bloomberg BNA, Tax Management International Journal.

Footnotes

1. Unless otherwise indicated, all ''§'' references are to the U.S. Internal Revenue Code of 1986 (''the Code'') and all ''Reg. §'' references are to the regulations issued thereunder (and set forth in 26 C.F.R.).

2. §954(d)(1) and Reg. §1.954-3. The Senate Report to the Revenue Act of 1962 stated:

The foreign base company sales income referred to here means income from the purchase and sale of property, without any appreciable value being added to the product by the selling corporation. The sales income with which your committee is primarily concerned is income of a selling subsidiary (whether acting as principal or agent) which has been separated from manufacturing activities of a related corporation merely to obtain a lower rate of tax for the sales income. This accounts for the fact that this provision is restricted to sales of property to a related person, or to purchases of property from a related person.

S. Rep. No. 1881, 87th Cong. 2d Sess., at 84 (1962), 1962-3 C.B. 703, at 790 and 949.

3. §954(d)(1) and Reg. §1.954-3(a)(2)–§1.954-3(a)(4).

4. §954(d)(2).

5. This analysis assumes that the manufacturing takes place in the remainder of the CFC (i.e., by the CFC's employees in the CFC's country of incorporation) and sales income is earned by a wholly owned disregarded entity. Thus, the analysis focuses on the determination of the ERTs when there is a sales branch. If, instead, the manufacturing takes place in a branch, the five-step process would be applied to compare the hypothetical ERT in the manufacturing branch's jurisdiction with the actual ERT in the country of incorporation of the remainder of the CFC.

6. Reg. §1.954-3(b)(2)(i)(a) and §1.954-3(b)(2)(i)(b)(1).

7. The GLAM notes that the use of taxable income computed under the laws of the sales jurisdiction as the common denominator would more likely result in a branch being treated as a separate subsidiary corporation under the tax rate disparity test. This is because the taxable income in the sales jurisdiction is often less than the hypothetical taxable income in the manufacturing jurisdiction, and when used as the common tax base, will produce in both jurisdictions ERTs that are higher, but not proportionally higher, such that it is more likely that the ERTs will differ by more than 5 percentage points. Thus, if the sales jurisdiction permits the same or additional deductions as the manufacturing jurisdiction, and there is no tax rate disparity calculated using the taxable income in the sales jurisdiction as the common tax base, there is no need to recalculate using the hypothetical tax base as the common denominator.

8. GLAM, n. 19, citing Reg. §1.954-3(b)(2)(ii) and §1.954- 3(b)(2)(b)(3). The use of U.S. tax principles at this stage of the analysis is a reasonable choice because, had the sales branch been a CFC, its FBCSI would have been determined under U.S. tax principles (including the applicable exceptions to FBCSI). That said, it is worth noting that this adds yet another level of complexity to the overall tax rate disparity analysis. After having recalculated the sales income under the laws of the manufacturing jurisdiction to obtain the common denominator for the tax rate disparity test, the taxpayer needs to recalculate such income once again, but now under U.S. tax principles.

9. The GLAM illustrates this point in n. 8: Assume that Country X and Country Y each have a statutory tax rate of 25% on taxable income, but Country Y excludes from taxable income one-half of income derived from the sale of property that is neither produced in Country Y nor sold to customers in Country Y. This special exclusion produces a smaller taxable base and an incentive to shift sales income from high-tax Country X to low-tax Country Y. However, if the actual ERT for the tax disparity test were determined with respect to Country Y's tax base, and the hypothetical ERT were determined with respect to Country X's tax base, the test generally would have the effect of comparing statutory rates and ignoring the incentive to shift income.

10. In PLR 200942034, Corporation A (a CFC organized in Country 3) owns Corporation B (a disregarded entity located in Country 2). The taxpayer requested the ruling to determine how to calculate the ERTs for Corporation B's sales income for purposes of the tax rate disparity test. In this respect, the taxpayer asked whether it should take into account the interest expense incurred by Corporation B on a promissory note payable to Corporation A. The ruling answered that the interest deduction should be taken into account according to the laws of Country 2, where Corporation B was located.

11. GLAM, n. 17. For example, the GLAM says that if the sales jurisdiction provides a 19-year useful life for a particular asset for depreciation purposes, and the manufacturing jurisdiction provides a 20-year useful life for that asset, the resulting difference in tax bases likely would not be material.

Originally published September 22, 2015

Learn more about our Tax Transactions & Consulting and Tax Controversy practices.

Visit us at mayerbrown.com

Mayer Brown is a global legal services provider comprising legal practices that are separate entities (the "Mayer Brown Practices"). The Mayer Brown Practices are: Mayer Brown LLP and Mayer Brown Europe – Brussels LLP, both limited liability partnerships established in Illinois USA; Mayer Brown International LLP, a limited liability partnership incorporated in England and Wales (authorized and regulated by the Solicitors Regulation Authority and registered in England and Wales number OC 303359); Mayer Brown, a SELAS established in France; Mayer Brown JSM, a Hong Kong partnership and its associated entities in Asia; and Tauil & Chequer Advogados, a Brazilian law partnership with which Mayer Brown is associated. "Mayer Brown" and the Mayer Brown logo are the trademarks of the Mayer Brown Practices in their respective jurisdictions.

© Copyright 2015. The Mayer Brown Practices. All rights reserved.

This Mayer Brown article provides information and comments on legal issues and developments of interest. The foregoing is not a comprehensive treatment of the subject matter covered and is not intended to provide legal advice. Readers should seek specific legal advice before taking any action with respect to the matters discussed herein.