Although overshadowed by the January death of David Bowie, which was featured in Tax Notes,1 more relevant for advisers and tax wonks is the death of the Eagles' Glenn Frey eight days later. Although he collected an impressive array of honors and accolades during his lifetime, Frey undoubtedly would have been proud to know that the United States continues to mine his rock band's extensive song catalog to develop its tax policy.
Several years ago, I wrote about Hotel California taxation, the U.S. policy of trying to ensure that taxpayers can ''check out any time you like'' but ''never leave'' the U.S. tax net.2 At the end of 2015, the United States turned its lonely eyes to search for tax policy in ''Lyin' Eyes,'' the Eagles classic about a woman who tries to recapture her lost youth by heading to the ''cheatin' side of town.''
To understand why, it's necessary to take a detour to discuss the OECD's base erosion and profit-shifting project.
The goal of BEPS is to create an OPEC-like cartel for tax ''prices.'' This isn't meant to imply anything sinister or conspiratorial—it's simply an accurate if overlooked description of what's going on. Governments are banding together, or attempting to band together, to reduce and constrain competition among them regarding the price they charge companies for having their headquarters in or doing business in their jurisdictions. The goal is simple if shortsighted — increase government revenue — just as OPEC's goal is to increase government revenue from the sale of oil. Simply put, BEPS seeks to increase tax revenue by curtailing competition among countries to lower prices (tax rates) as a way to increase their ''market share.'' This is precisely the same goal that OPEC or any cartel has. The only difference is that the price is called ''taxes'' and the ''product'' — a place to incorporate or headquarter a business — is something intangible rather than a commodity like oil. The IRS has seemingly acknowledged this cartel-like behavior by referring to U.S. taxpayers as ''customers'' — customers who can and should be prohibited from seeking better prices elsewhere.
It is interesting that despite some aspirational slogans (which we can charitably refrain from calling propaganda), no policy goal has rationally been put forward to support these initiatives, particularly because most of the BEPS's focus is on the taxation of corporations and similar entities rather than individuals. This distinction is important, because it means that the usual arguments in favor of progressive taxation don't apply. Subjecting Apple Corp. or any other large multinational enterprise to higher taxes because it is more profitable than other enterprises serves no underlying premise of progressive taxation. Its only consequence is to tax success precisely because of that success. Taxing Apple is no more progressive than taxing any other business. And allowing a corporation that happens to lose money to pay no taxes likewise serves no progressive principle. Wherever the true burdens of corporate tax lie, the amount and rate of corporate profits have no correlation to whether one corporation's labor, customers, or providers of capital are richer, poorer, or otherwise more deserving of a higher or lesser tax burden than their counterparts in less profitable companies. The only relative effect the corporate tax has is to burden profitable companies more than less profitable companies.
BEPS is therefore not a policy. It serves no purpose other than to try to stifle competition among governments to raise revenue. Regardless of whether that purpose is good or bad, we should at least be honest that this is what is being done and ignore the clever and self-serving language that BEPS is designed to attack ''tax evasion'' or get companies to pay their ''fair share.'' Raising oil prices, or any other cartel behavior, is also designed to prevent price evasion and ensure that users pay their fair share. In the case of both taxes and oil, there is no reason why one price is fairer than another. And as between governments, there is no reason why we should look at a cartel regarding taxes more or less favorably than cartels involving natural resources like oil.
The United States has not only adopted the cartel-like ''customer'' language for its taxpayers, it has also signed on to the anti-competition agenda through several technical initiatives. For example, the recently released revised U.S. model tax treaty seeks to deny treaty benefits for a particular type of ''mobile'' income when that income is subject to preferential tax rates in the recipient's state of residence. In other words, the revised treaty provision seeks to restrain a resident country's ability and incentive to compete by offering lower prices — taxes — to attract that capital. It does so by implementing the penalty of denying treaty benefits to anyone who takes advantage of that lower price. The United States has also acted unilaterally to try to curtail tax price competition through a seemingly endless and amorphous attack on inversions by way of legislation, regulation, and quasiregulation. Here the attempt is designed to prohibit or severely restrict a U.S. company's ability to become a customer of a nation with a lower tax rate.
One curious feature of inversions is noteworthy: If the foreign company is bigger, or big enough that the 60 percent or 80 percent threshold of U.S. ownership of the combined enterprise is not met, the inversion rules don't apply to the enterprise. Only U.S. companies that are more successful—for example, larger relative to their non-U.S. counterparts — are penalized by the inversion rules. Is there any rational policy justification for this feature? Or is it simply a recognition that at some point we need foreign enterprises to be able to invest in U.S. companies without being subject to the onerous price of the restrictions applicable to inversions? In other words, we need to lower the price of foreign capital investing in U.S. companies to a tolerable level, lest that foreign capital invest elsewhere. So we profess to follow the goal of constraining tax competition while cheating on that goal when it conflicts with the hard reality of what we need even more.
A cartel member's commitment to the noncompetition goal is not absolute, and like the protagonist of ''Lyin' Eyes,'' we aren't averse to heading to the cheating side of town when expedience dictates. A more interesting example of this phenomenon is the partial repeal of the 1980 Foreign Investment in Real Property Tax Act for qualified foreign pension plans as part of the Protecting Americans From Tax Hikes Act of 2015.
Section 897(l), enacted at the end of 2015, states that section 897 will not apply to a qualified foreign pension fund (QFPF). The goal of this partial repeal of FIRPTA is to better compete for the money of these pension funds — that is, to lower the price of investing in the United States by lowering the taxes the funds face from these investments. The commitment to BEPS's implicit goal of not competing on price gave way to the realization that, well, we need the foreign funds' money — the same type of parochial interests that always doom a cartel's ability to restrict its members' natural inclination and incentive to compete.3 Indeed, the entire premise of section 897(l)—that the United States is better off by attracting this capital, even by lowering the cost of entry to zero — is complete anathema to the supposed policy of BEPS that every effort must be made to maintain the taxing price at whatever artificial level the cartel decides is the right corporate tax rate. Both policies can't be right, so which one is?
And in a boon to both taxpayers and advisers, even a seemingly innocuous provision like section 897(l) is full of ambiguity. For example, what exactly is a QFPF? The recently released blue book4 provides some helpful guidance. It provides that a separately organized entity that invests funds for the benefit of a pension fund, or for more than one pension fund, should qualify as part of an ''arrangement'' that is a QFPF.5 Similarly, public pension plans, including those that benefit the general working public rather than only employees of a specific employer, should also qualify. However, the adviser community is rife with ad hoc working groups and conferences trying to figure out what and who is covered by the new law and at what comfort level. The language in section 897(l)(1)(B) that extends the benefits of being a QFPF to an entity all of whose interests are held by a QFPF creates another area of ambiguity because it's unclear whether any wholly owned entity, even one that is a commercial enterprise rather than an investment entity, would qualify, and it's unclear whether a second- or lower-tier subsidiary of a QFPF would qualify.
Perhaps more interesting is the extent to which a QFPF can actually take advantage of this partial repeal of FIRPTA. As one esteemed colleague presciently noted, although section 897(l) repeals section 897 for QFPFs, it doesn't repeal the rest of the code.6 As a result, most investments in real estate will still cause the QFPF to be considered to be engaged in a trade or business in the United States and thus subject to tax on the income that is effectively connected with that trade or business. That income could include gain from the disposition of the assets of that business (U.S. real estate), even though those gains are no longer taxable under section 897. What good is being exempt from section 897, which merely deems you to be engaged in a trade or business, if you are considered to actually be engaged in a taxable trade or business anyway?
Take, for example a QFPF that invests in a project to purchase real estate, develop it, and sell condominiums. Presumably, this is clearly engaging in a U.S. trade or business. If the investor is in a treaty jurisdiction, the physical location of the real estate in the United States should be sufficient to cause the investor to have a permanent establishment here. Or does it? Could a QFPF organized in a treaty jurisdiction engage an independent agent to develop the real estate and handle all the selling activity, allowing the fund to argue that it doesn't have a PE in the United States? Most of the case law involving independent agents arises in situations in which the PE is owned by the independent agent rather than the taxpayer, and the question is whether the independent agent's PE should be attributed to the non-U.S. investor. In our scenario, the only establishment is the real estate being developed, so why would using an independent agent to conduct that development avoid the owner of the real estate being considered to have a PE? Little case law or attention has been paid to this and related matters because these types of questions were moot once section 897 was enacted over 30 years ago. FIRPTA's repeal for QFPFs brings those issues to the surface again.
Many treaties specify that a construction site can be a PE if it lasts longer than one year.7 Does this mean that a foreign investor could avoid having a PE if the condominium development is completed in less than a year? Although this provision was intended to cover situations in which a non-U.S. contractor is necessarily occupying a construction site it doesn't own, nothing in its terms says it can't be applied to a construction site owner that either forms its own contractor to manage the site or uses an independent contractor to do so.
Suppose instead that the foreign investor engages in a different type of real estate activity, such as owning an infrastructure asset. What if the QFPF invests either directly or through a partnership in a transportation facility (such as a bridge or road), an energy asset such as transmission lines, a power plant, or a similar type of asset? Again, presumably if the QFPF operates the facility, it will be treated as engaging in a U.S. trade or business, and unless the analysis above regarding operating through an independent contractor is adopted, the QFPF, as owner of the real estate, will have a PE, and income from the investment will therefore be subject to full U.S. net income tax. What advantage does section 897(l) then provide to a QFPF? If part of the impetus for its enactment was to encourage investment in infrastructure, how does it accomplish that?
For QFPFs that reside in treaty countries or are governmental pensions entitled to the benefits of section 892, one solution, and perhaps the only solution, would be to invest through a noncontrolled real estate investment trust. The REIT could then own the assets and lease them to an operator to produce qualifying REIT income. That income could be distributed through dividends that are exempt under the pension article of the applicable treaty or under section 892. The QFPF could then exit the investment tax-free either by selling REIT shares or having the REIT sell the property and distribute the proceeds. Although that structure was previously available to government pension plans entitled to the benefits of section 892, the QFPF regime opens it up to QFPFs that have the benefit of a treaty exemption for dividends but have not had a way to exempt gains on an exit from the investment since the issuance of Notice 2007-55, 2007-1 C.B. 13. The QFPF regime also allows both categories of investors to exit through property sales followed by a distribution of the proceeds by the REIT rather than requiring a sale of REIT shares.
If this is all section 897(l) does, however, it is essentially nothing more than a complicated means of repealing Notice 2007-55 for a limited class of pension investors.
If, by contrast, the QFPF invests directly or through a partnership in the real estate rather than through a REIT, both the income and the gain would generally be subject to tax as effectively connected income. Or would it?
What if the QFPF or the partnership leases the property to an operator, the same as the REIT would, in a manner that keeps the investor from being considered to be engaged in a trade or business? Presumably, the gain would then be exempt under section 897(l), and no other code provision would change that result. The main problem with this approach, however, is that the rental income would be subject to gross basis withholding, with no deductions for expenses, at a 30 percent rate, which is rarely reduced by treaty and would not be reduced by section 892.
But what if the rental business was active enough to cause the investor to be engaged in a trade or business (or the investor made the election to be so considered under section 882(d))? Would gains still be subject to net basis tax? Rev. Rul. 91-32, 1991-2 C.B. 20, suggests they might not. Although the revenue ruling concludes that the sale of an interest in a partnership that is engaged in a trade or business produces ECI, it states that the ''ECI (United States source) property of a partnership does not include United States real property interests held by the partnership.'' Rev. Rul. 91-32 then says that the treatment of gain from a disposition of real estate owned by a partnership is governed by section 897(g) rather than the rules described in the revenue ruling. Moreover, Rev. Rul. 91-32 indicates that the result is dictated by the ''view of Congress'' that real estate be governed by section 897 rather than the rules otherwise applicable to ECI. Does that mean that gain from a disposition of real estate held by a partnership, or even directly, is subject only to the FIRPTA regime and cannot be treated separately as ECI? If so, section 897(l) would mean that the gain from a disposition of real estate, even when the real estate was used in a trade or business, would be exempt. In our infrastructure example, the foreign investor, while taxable on operating income, could at least exit from the investment without being subject to U.S. tax. Other than these statements in Rev. Rul. 91-32, another potential argument to support this result is the source rule of section 861.
Gain from a disposition of noninventory real estate, even if used in a trade or business, is generally treated as capital gain under section 1231. Capital gain from U.S. real estate is considered U.S.-source under section 861(a)(5). However, section 861(a)(5) refers to a ''United States real property interest'' as defined in section 897(c), and section 897(l) says quite clearly that section 897, which would include section 897(c), does not apply to a QFPF. For a QFPF, it is as if section 897 simply doesn't exist. Therefore, for a QFPF, the reference in section 861 doesn't actually refer to anything — as far as QFPFs are concerned, there is no such thing as a U.S. real property interest as defined in section 897 because section 897 doesn't apply to them. The gain therefore can't be U.S.-source and isn't ECI, because the limited circumstances in which non- U.S.-source income can be treated as ECI would not apply. Although a QFPF would still be subject to net basis tax on operating income as ECI, it could then take advantage of section 897(l) to avoid taxes on the gain that accrues upon exit. This benefit may then provide a sufficient incentive to invest in the United States in infrastructure and other real estate ventures that the QFPF would otherwise have avoided.
If the policy of section 897(l) is in fact to encourage investment in the United States by offering a lower price — for example, a lower tax burden — not only are these arguments not abusive or aggressive, they are the only way to interpret the statute that is consistent with that policy.
Of course, the danger of looking to policy is that doing so implies there is a commitment to some rational policy rather than to whatever Eagles song happens to be playing on the radio.
How these ambiguities are resolved will largely determine whether section 897(l) yields the desired incremental capital investment or whether the cheating side of town proves to be less satisfying than it seemed at first blush. As the Eagles sang, ''city girls just seem to find out early how to open doors with just a smile.'' In a world in which capital really is mobile, however, and taxes are just another price, a smile is no longer enough — despite what the proponents of BEPS want to believe. Indeed, it is the need to resort to Hotel California taxation and BEPS that shows that the world has changed. Business and capital can and will locate anywhere and can be owned, financed, and deployed by anyone. If countries are going to insist on continuing to tax fictional entities, they're going to find that they ultimately have a problem that even OPEC hasn't yet confronted — they don't own the ''oil'' whose price they are trying to maintain.
1 See Lee A. Sheppard, ''Escape From U.S. Tax Jurisdiction,'' Tax Notes, Feb. 1, 2016, p. 483.
2 Scott L. Semer, ''Hotel California Taxation,'' DTR, Sept. 4, 2014.
3 For a prediction of this phenomenon, see Semer, ''The Road to Nowhere: How Two New U.S.Withholding Initiatives Reveal the Deeper Problem With Current U.S. Tax Policy,'' 58 Can. Tax J. 1047 (2010).
4 Joint Committee on Taxation, ''General Explanation of Tax Legislation Enacted in 2015,'' JCS-1-16 (Mar. 2016) (the blue book).
5 Footnotes 967 and 968 of the blue book, id., suggest that such an entity, if created as part of the arrangement that governs the pensions funds it invests for, would itself qualify as a QFPF. The question whether an entity owned by more than one QFPF is covered by the language in section 897(l)(1)(B) would thus be irrelevant.
6 For example, section 291 treats some real estate investment trust distributions as ''gain which is ordinary income'' that is ''recognized notwithstanding any other provision of this title.'' Does this rule control over section 897(l)? What exactly does it mean to treat something as ''gain'' that is ordinary income? Is it still gain that is only taxable to a non-U.S. investor if section 897 applies?
7 See, e.g., Treasury, ''Technical Explanation of the United States Model Income Tax Convention,'' art. 5, para. 3 (2016).
Previously published in Tax Notes, May 9, 2016, p. 797
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.