New U.S.-Poland Treaty Signed—The End of an Era?

Trust Bloomberg Tax for the international news and analysis to navigate the complex tax treaty networks and global business regulations.

By Philip D. Morrison, Esq.  

Deloitte Tax LLP, Washington, D.C.

On Ash Wednesday, February 13, Pope Benedict XVI celebrated his last mass as Pope and a new income tax treaty between the United States and Poland was signed. Both events were near-final acts in closing an era, one a Papacy, the other the era of relatively easy treaty shopping into the United States. For the Roman Catholic Church, a single Papacy is but a blip in a two-millennia history.  For U.S. tax treaties, with a history of less than 100 years, the end of treaty shopping is a big deal (though, in the grander scheme of things, obviously not comparable to the end of a Papacy).

Treaty shopping into the United States got its first meaningful roadblock with the entry into force in late 1993 of the 1992 U.S.-Netherlands Income Tax Treaty. The Netherlands' favorable domestic tax regime, together with its large network of tax treaties, allowed the 1948 U.S.-Netherlands Income Tax Treaty to operate as, essentially, the U.S.'s treaty with the world. While the Netherlands remains today a significant source of inward investment into the United States, it was perceived, prior to the entry into force of the 1992 U.S.-Netherlands treaty, that some of the heavy U.S.-bound investment from the Netherlands was due to the ability of non-Netherlands investors to use the Netherlands as a conduit allowing access to favorable U.S. treaty benefits.

The Joint Committee on Taxation described this in its explanation pamphlet prepared for the Senate Foreign Relations Committee's ratification hearing in late 1993:Among existing U.S. income tax treaties, it is believed that the present Dutch treaty has been notable in its susceptibility to abuse. The combination of Dutch internal law and the Dutch treaty network, including the present treaty, makes it possible in some cases for persons including residents of third countries to earn U.S. income relatively free of all tax, U.S., Dutch, or otherwise. Dutch law in many cases exempts Dutch residents from tax on foreign income, including foreign income that bears very little tax; in addition, payments by a Dutch resident to a foreign resident can be structured at times to bear little or no Dutch withholding tax, either due to Dutch internal law which generally exempts interest and royalties from withholding tax, or due to tax treaties that exempt or favor dividends as well. Finally, the present treaty ensures a U.S. dividend withholding rate as low as that available under U.S. treaties with any other country, forbids U.S. withholding tax on U.S. source interest and royalties, and forbids so-called "second-level" U.S. withholding taxes on dividends or interest paid by a Dutch company but attributable to its U.S. income.1

It was considered "abusive" for third-country residents to use the U.S.-Netherlands treaty to obtain benefits in part because U.S. residents could not obtain reciprocal benefits from the third countries. Since those countries' residents could obtain favorable treaty benefits (such as an exemption from U.S. withholding tax on interest and royalties) via the U.S.-Netherlands treaty, there was no need for those countries to offer a similar reduction in withholding tax rates for U.S. investment inbound into such countries.  This was true both for countries with no tax treaty with the United States as well as those, such as Canada at the time, with treaties with the United States that had higher rates on such income than the U.S.-Netherlands treaty did.

The 1992 U.S.-Netherlands treaty prevented most treaty shopping via the Netherlands by the inclusion of a comprehensive "limitation on benefits" (LOB) article modeled on a similar LOB article included in the then-recently-concluded new treaty with Germany and the anti-treaty-shopping rules in the §884 branch tax regulations.  As most readers know, the U.S.-Netherlands LOB article included, like all more modern LOB articles (though many modern ones are more restrictive in certain ways) a list of ways to qualify a resident by virtue of its ownership (and its lack of base eroding in certain cases) or to qualify an item of income by virtue of its connection to an active trade or business in the residence country.

The entry into force of the 1992 U.S.-Netherlands treaty with its complex LOB article did not end treaty shopping into the United States, however. It just moved it to other jurisdictions.  First, treaty shopping moved to Ireland, Luxembourg, or Switzerland.  In the mid-to-late 1990s, however, new treaties between the United States and those countries were negotiated. They all contained generally similar LOB articles, like the one in the 1992 U.S.-Netherlands treaty.  Again, however, treaty shopping was not stopped; it just moved once again to other jurisdictions.

The jurisdictions of choice for the past decade have been Iceland, Hungary, and Poland. Each had a treaty with the United States offering favorably low rates on dividends and exemption for interest and royalties. And each had an adequately inviting domestic tax system providing low rates, special base eroding methods, certainty through a rulings system, or a combination thereof. Most importantly, none of the three treaties contained a comprehensive LOB article so a nonresident could set up a resident conduit company through which it could invest in the United States and still obtain treaty benefits.

The Iceland "window" was the first to close. A new U.S.-Iceland treaty with a comprehensive LOB provision entered into force in 2008. In 2009 a new treaty with Hungary was initialed, then signed in 2010, threatening to close the Hungary treaty shopping route. Oddly, however, while the treaty was quickly ratified in Hungary, some have speculated that the U.S. Treasury appeared to acquiesce in "slow walking" the new treaty through the U.S. ratification process. While a hearing before the Senate Foreign Relations Committee was finally held in the summer of 2011 and the treaty was voted out of committee shortly thereafter, the treaty still has not received a vote in the full Senate. Some have commented that it appeared as though an arrangement had been made between the United States and Hungary negotiators that the Hungary window would not be closed until the Polish window was shut simultaneously.

The LOB article in the pending U.S.-Poland Income Tax Treaty (Article 22) resembles LOB articles in newer U.S. tax treaties with European countries. It includes a derivative benefits test, a triangular rule, a headquarters company test, a publicly traded company test, and an active trade or business test.2

Publicly traded companies: The publicly traded test in the pending treaty is identical to the publicly traded test in the pending U.S.-Hungary treaty. For a company to be a qualified person on the basis of public trading of its shares, the company must meet a test similar to the one first seen in the 2004 protocol to the U.S.-Netherlands treaty, aimed chiefly at preventing inverted companies from qualifying for comprehensive treaty benefits.  (Article 26(2)(c)(i) of the U.S.-Netherlands treaty; see also Article 22(2)(c)(i) of the U.S. Model.) The test requires that either: (1) the company's country of residence be its primary place of management and control (where "management and control" is a concept expanded well beyond other countries' concepts); or (2) certain economic activity zones be the location of the exchange on which its principal class of shares is primarily traded. To be a qualified person on the basis of place of primary trading of principal class of shares, a Polish resident company's shares would need to be primarily traded on a recognized stock exchange located within the European Union or in any other European Free Trade Association (EFTA) state,3 and a U.S. resident company's shares would need to be primarily traded on a recognized stock exchange located in a country that is a party to the North American Free Trade Agreement.

Subsidiaries of public companies: As with the pending U.S.-Hungary treaty, to be a qualified person on the basis of direct or indirect ownership of a company's shares by U.S. or Polish publicly traded companies, ownership by such companies must add up to 50% or more of the aggregate vote and value of the shares and at least 50% of any disproportionate class of shares.  In the case of indirect ownership, each intermediate owner must be a resident of either Poland or the United States.4

Ownership/base erosion test: The ownership prong of this Polish treaty test is identical to Article 22(2)(e)(i) of the U.S. Model, and for this reason is more stringent than most recent U.S. tax treaties that have not adopted the U.S. Model provision.5 Thus, it includes a requirement that the income recipient be owned by persons in the same Contracting State.  In the case of indirect ownership, each intermediate owner must also be a resident of the same Contracting State. The base erosion prong of the test is met if less than 50% of the company's gross income for the taxable year (as determined in the company's state of residence) is paid or accrued, directly or indirectly, to persons who are not qualifying residents of either Contracting State in the form of payments deductible in the company's country of residence. As under Article 22(2)(e)(ii) of the U.S. Model, arm's-length payments in the ordinary course of business for services and tangible property are excluded from this calculation. This provision differs from the same in the pending U.S.-Hungary treaty and some other recent treaties since payments on obligations to banks are not also excluded.

Derivative benefits test: This provision is essentially identical to Article 22(4) of the pending U.S.-Hungary treaty except that it also requires the resident to satisfy "any other specified conditions for the obtaining of such benefits" with respect to an item of income. The derivative benefits test is similar to that in recent U.S. tax treaties with European countries (e.g., Article 17(3) of the U.S.-Sweden treaty, Article 23(3) of the U.S.-U.K. treaty, and Article 21(3) of the U.S.-Belgium treaty).  It contains a base erosion test similar to that discussed above in the context of the ownership/base erosion test, although payments to equivalent beneficiaries are not base-eroding payments. Like other derivative benefits provisions (with the exception of Article XXIXA(4) of the U.S.-Canada treaty), the ownership prong of this test requires direct or indirect ownership of 95% of the aggregate vote and value (and 50% of any disproportionate class of shares) of the treaty-resident company claiming benefits by seven or fewer persons that are equivalent beneficiaries.

Headquarters company test: The headquarters company test in the pending treaty is identical to the headquarters company test in Article 22(5) of the pending U.S.-Hungary treaty.6 The headquarters company test provides that a headquarters company for a multinational corporate group that is resident in either Poland or the United States may qualify for benefits under the pending treaty if it and its corporate group meet certain specified requirements.

Discretionary rulings: Finally, the pending Polish treaty also contains a typical provision allowing a resident that does not otherwise qualify under the LOB provision to seek a discretionary ruling from the source State's Competent Authority.  As with virtually all other such provisions, the Competent Authority should grant benefits under such a provision - if it determines that the establishment, acquisition or maintenance of such person and the conduct of its operations did not have as one of its principal purposes the obtaining of benefits under the Convention.7

Unfortunately, the current rulings practice of the U.S. Competent Authority is to grant such rulings only where one (or more) of the objective LOB tests is narrowly missed, so the apparent promise of this provision may be larger than it delivers in practice.

So, what do treaty shoppers do now that the last real "windows" into the United States are likely to close within a year? For taxpayers resident in an EU, EFTA, or NAFTA jurisdiction, it appears that derivative benefits provisions in a variety of treaties are still the most useful tool for shopping into the United States.8 For persons resident outside those jurisdictions, it appears treaty shopping may be, for all practical purposes, impossible. We may actually be witness to the end of an era, a successful conclusion of a 25-year effort to implement a tax treaty policy that, when it was embarked upon, few would have predicted would come to a successful conclusion in so "short" a time.

This commentary also will appear in the April 2013 issue of the  Tax Management International Journal.  For more information, in the Tax Management Portfolios, see Levine and Miller, 936 T.M., U.S. Income Tax Treaties -- The Limitation on Benefits Article, and in Tax Practice Series, see ¶7160, U.S. Income Tax Treaties

  1 Joint Committee on Taxation, Explanation of Proposed Income Tax Treaty (and Proposed Protocol) Between the United States and the Kingdom of the Netherlands (JCS 15-93) (10/26/93).

  2 While there is no substantiality safe harbor in the active trade or business test, it is possible that one could be added in a memorandum of understanding or diplomatic notes yet to be agreed upon or released.

  3 Iceland, Norway, Switzerland, and Liechtenstein are the current EFTA members.

  4 See the identical provision in Article 22(2)(c)(ii) of the U.S. Model.

  5 Butsee Article 22(2)(e)(i) of the pending U.S.-Hungary treaty and Article 30(2)(e)(i) of the U.S.-France treaty (as amended by the 2009 protocol), which, like the pending U.S.-Poland treaty, also adopt the U.S. Model provision.

  6 Both provisions are substantially identical to the headquarters company test in Article 26(5) of the U.S.-Netherlands treaty.

  7 Pending treaty, Article 22(7).

  8 Though "shopping" is hardly the right word, given explicit approval of such use in those treaties.

Request International Tax