The New U.S. Model Treaty Is Out!

Bloomberg BNA’s Premier International Tax Library is a comprehensive global tax resource. Trust Bloomberg BNA's Premier International Tax Library for the guidance you need on...

By James J. Tobin, Esq.1

Ernst & Young LLP, New York, NY

I was excited by the release of the new U.S. Model Treaty (the Model). It brings to mind one of my favorite scenes from the old movie "The Jerk" where Steve Martin is running down the street exclaiming "the new phone books are out, the new phone books are out." I was hoping that the new Model would be more impactful than the new phone books back in 1979. Of course, nobody uses phone books any more. But, unfortunately, after reading the Model and recognizing the difficulty the United States has in ratifying treaties or protocols in the first place, I'm not convinced the new Model will have much more effect than the new phone books Steve Martin was so excited about. I'll elaborate a bit below but overall it seems to me that the Model is so cluttered with domestic U.S. policy anti-abuse concepts which should be dealt with by legislation, if at all, that potential treaty partners would be well-advised to generally avoid the process of trying to negotiate for a reasonable treaty outcome and then hope for eventual ratification.

It's certainly not that we don't need more treaties.  U.S. companies are the most global in the world and the United States is the leading destination for global investment. However, the number of our income tax treaties has lagged behind our key trading partners.  The United States has 57 comprehensive income tax treaties in effect covering 66 countries. The United Kingdom, the Netherlands, and Switzerland all have well over 100 and even Canada has over 90. Given the increasing tax controversy environment fueled by BEPS, it's more important than ever to increase our treaty network to ensure access to potentially effective mutual dispute resolution. So a new Model should have been the first step in an open reach-out to the world to increase our treaty network. Seems like quite the opposite tone and impression is created.

The primary changes in the Model include the following:

  •   Denying treaty benefits for certain payments that benefit from a "special tax regime"
  •   Denying treaty benefits in certain cases for payments made by expatriated entities
  •   Changes, generally tightening, to the Limitation on Benefits test
  •   Modifications to the Mutual Agreement Procedure of Article 25 to require mandatory binding arbitration.

I applaud this binding arbitration provision change, which has already been included in recent treaties. I have problems with most of the tightening aspects of the changes above, as discussed below.

Let me start with special tax regimes (STRs).  Article 3(l) defines an STR which is made operative in the Interest (11), Royalty (12), and Other Income (21) articles. The gist is that if the income in question paid from the United States is subject to an STR, the full 30% withholding rate will apply to that payment.  An STR is defined as a foreign tax regime that provides a preferential rate or permanent reduction with respect to interest, royalties, or guarantee fees as compared with income from the sale of goods and services. The special regime could be in the form of a lower tax rate or a base reduction that is expected to result in a rate of taxation that is less than the lesser of either a rate of 15% or 60% of the general statutory corporate tax rate in the treaty country. Note base reductions are viewed broadly and include a deduction for dividends (but not for a REIT or RIC equivalent) or deductions without regard to any corresponding payments or obligation to make a payment. Notional interest deductions (NIDs) sound like they would be covered here but in most countries NIDs can offset all income, including sales profit.  However, Article 11(e) makes clear that interest paid to a related person who benefits from a NID can be taxed at 30%.

Overall seems to me like a lot of uncertainty for our treaty partners to accept in figuring out what could be an unacceptable base reduction and could interfere with their domestic tax policy objectives. Two aspects of modest good news are that: (1) a patent box regime would not constitute an STR as long as the benefits provided are conditioned on the extent of research and development (R&D) activities that take place in the Contracting State; and (2) there is a requirement to notify the other Contracting State before its law or practice can be considered an STR. The latter is good news against rogue IRS agents but there is no requirement for mutual agreement, so it's not that comforting. (The Preamble refers to the need for consultation before notification and it would be nice to add that to the Model language as well.) The former is consistent with the agreed approach on low-tax patent or innovation boxes in Europe to avoid unfair tax competition or EU State Aid risks. But it also encourages the movement of R&D jobs to secure the low-tax incentive which I thought the United States was at least wary of, if not fully against.

The second big carve-out from treaty withholding tax benefits is with respect to payments by an expatriated entity.  This carve-out applies to dividend, interest, royalties, and guarantee fees and applies to such payments by an expatriated entity for 10 years after the date of acquisition of the expatriated entity. There is no definition in the Model of an expatriated entity and the Preamble released with the Model notes that the definition will be based on the §7874 definition as of the date the bilateral treaty is signed.

Section 7874 is a very complex international tax provision. It's also one where there is relatively little practical experience as the statute is relatively recent (2004) and the regulations are very recent, especially the temporary regulations issued in April 2016. Add to this that there are very few companies in the fact pattern that will be impacted by the Model expatriated entity rules (just those having done a 60-80% inversion and not satisfying the significant presence exception). Take all this together and it just doesn't seem to me like a provision that should be included in an income tax treaty where both the foreign country and U.S. Treasury treaty negotiators need to understand well the effects of the treaty bargain they are negotiating. In my experience, most U.S. advisors who focus heavily in this area find aspects to disagree about or at least that they find very uncertain. For a foreign country to sign up to that uncertainty is unrealistic. This is an area which should be left to U.S. legislation (which I guess is considered too difficult to enact).

Despite the fact that this provision is ill-suited for inclusion in a treaty, let's analyze a bit what it attempts to do. As stated above, dividend, interest, royalty, and guarantee fee payments by an expatriated entity would not be eligible for reduced treaty withholding. Section 7874(a)(2)(A) essentially defines an expatriated entity as a U.S. corporation or partnership that was a party to an inversion transaction after March 4, 2003, that resulted in ownership by the former U.S. shareholders/partners of between 60% and 80% in the foreign parent (Surrogate Foreign Corporation). The definition also includes any U.S. person who is related to that U.S. corporation or partnership within the meaning of §267(b) or §707(b)(1).  Thus, it seems that U.S. entities acquired or that otherwise become related parties after the inversion could become tainted expatriated entities.

The Model excludes from this related-party taint a U.S. entity that was not related to the U.S. expatriated entity before the date of the acquisition, but only if that related U.S. entity does not join in a consolidated U.S. tax return with the expatriated entity or another entity that was a connected person with the expatriated entity. Note that a subsequent acquisition of the Surrogate Foreign Corporation/foreign inversion parent does not seem to cure the expatriated entity status of the U.S. subsidiaries, and newly acquired U.S. companies generally become tainted expatriated entities once they become related parties.

So it seems to me this inversion carve-out should be wholly unacceptable to a potential treaty partner:

  •   Policywise, why should a foreign country accept that its multinationals lose U.S. treaty benefits if there was a prior merger/inversion which to my mind was likely heavily influenced by an uncompetitive U.S. tax policy?
  •   The potentially lasting effect of expatriated entity status seemingly could apply when any foreign corporation acquires a former inverted group. For example, a large foreign treaty resident group buying a minnow inverted group could taint treaty access for up to 10 years if it integrates a U.S. group with the "minnow" recently expatriated entity group.
  •   The lack of a treaty definition of an expatriated entity would mean that the foreign country and its local multinationals would need to take detailed advice on the potential impact of the provision. The Model makes clear that the definition of expatriated entity is the law in effect at the time the treaty is signed. Given the tendency to modify §7874 by notice and regulation, that's not going to be easy to keep a fix on.

Obviously, including this provision in the Model is just another example of an attempt to create yet another deterrent to inversions, seemingly with little thought to the practicalities of that action.

Changes to the Limitation on Benefits provision (LOB) in Article 22 have some positive and negative aspects. But overall the LOB becomes an even more complex provision to understand and apply. Given the preference of many foreign countries for a principal purpose test rather than an LOB, as made clear in the OECD BEPS process, I fear the complexity of the Model only pushes them all the more in that direction. Key provisions that caught my eye include:

  •   The 50% ownership/base erosion test of Article 22(3)(f) requires at least 50% ownership (direct or indirect) by certain qualified persons (publicly traded corporations, government entities, individuals, or qualified pension funds) only of the other Contracting State. So ultimate ownership by U.S. qualified persons, such as individual shareholders, would not also qualify. Plus, in the case of an indirect owner, all intermediate owners would need to also be treaty-qualified. Both of these requirements were also present in the 2006 Model but are more restrictive than most pre-2006 treaties.
  •   A derivative benefits provision, as has been included in our recent treaties, is now finally included for the first time in the Model.  I whined about this quite awhile ago in an earlier commentary,2 where I complained that the lack of a consistent Model resulted in inconsistent provisions on derivative benefits in each of our treaties, resulting in lots of traps for the unwary. Naturally, I'm not crazy about all aspects of the Model provision either, so consistency might be overrated.

The basic concept behind derivative treaty benefits is the same as in our recent treaties:

  •  At least 95% of the aggregate vote and value of the shares of a treaty resident company are owned (directly or indirectly) by seven or fewer equivalent beneficiaries; and
  •  Less than 50% of the companies' gross income is used to make base-eroding payments to non-equivalent beneficiaries.

Some changes from what we have seen in prior treaties:

  •  Most existing derivative benefit provisions limited qualified equivalent beneficiaries to treaty residents within a regional trading block – for example, the U.K. treaty requires equivalent beneficiaries be NAFTA, EU, or EEA members. The Model contains no such geographic limitations.
  •  To qualify as an equivalent beneficiary, the owner must be resident of a country with a comprehensive treaty with the United States, be a qualified resident under that treaty under the equivalent of Article 22(2)(a), (b), (c), or (e), i.e., an individual, a government entity, a publically traded company, or a qualified pension, and be entitled to an equivalent rate of withholding tax as would apply to the treaty in question.
  •  The equivalent withholding rate is interesting. In our existing treaties the qualification requirement acts as a cliff. If the rate, say on interest, is higher in the shareholder's treaty with the United States, the derivative test would be failed with a default to 30%. In the Model, if the shareholder rate is higher, there would still be a potential benefit allowed up to that alternative reduced rate. For example, assume a Mexican publically traded company qualifying for the U.S.-Mexico treaty owned a Canadian subsidiary which loaned funds to a U.S. affiliate.  The Canadian treaty interest withholding tax is zero, whereas the U.S.-Mexico treaty rate is 15%. Under the existing U.S.–Canada treaty, there would be no potential for derivative benefit, resulting in a 30% U.S. withholding rate. But under the Model, a 15% rate would apply.
  •  Interestingly, individual treaty residents who are taxed either on the basis of remittance of income to the country or on a deemed or "forfeit" basis are not considered equivalent beneficiaries.
  •   An interesting change is in the active trade or business LOB qualification test. The test still requires a substantial active business in the treaty country.  But the prior Model and our current treaties require that to be granted treaty benefits with respect to an item of income the income be derived "in connection with" the business of the treaty resident or be incidental to that business. The new Model replaces the term "in connection with" with the term "emanates from." The Preamble to the Model makes clear that this is intended to be a tougher standard.

The new Model, like the 2006 Model, requires for a publically traded treaty corporation that either its shares trade primarily on its local foreign stock exchange or it has its primary place of management in the treaty country. Primary place of management is defined as the place where executive officers and senior management employees and their staff conduct more of the day-to-day financial and operating policy decision-making for the group. While seemingly focused primarily on inverted companies, I have seen this provision cause some concerns in large foreign multinationals with a U.S. stock exchange listing and a significant U.S. presence. It can discourage the relocation of senior management jobs to the United States or away from the headquarters country and could encourage a move of jobs from the United States overseas in a cross-border merger – inversion or not. Is that good policy for the United States?

A positive change to the Model is in Article 25, Mutual Agreement Procedure, which calls for binding arbitration using the last best offer method. This is the approach used in recent U.S. treaties and the approach sought by the OECD, but not a feature of the 2006 Model. To me, it's the most attractive feature of the Model and will be critically important in anticipation of increasing controversy around the globe.

Space constraints prevent a full analysis of all the changes to the Model and from our existing treaties. But as indicated at the outset, I think our Treasury staff is attempting to stuff too much U.S. tax policy-driven detailed rules into the Model – no doubt driven by a frustration at the inability to include some of these provisions in legislation – but, nonetheless, not a good strategy. And it will create long-term harm for our U.S. multinationals who would benefit from a much more comprehensive treaty network, which I highly doubt will be accomplished if we start negotiations constrained by this new Model.

This commentary also appears in the May 2016 issue of the  Tax Management International Journal. For more information, in the Tax Management Portfolios, see Isenbergh, 900 T.M., Foundations of U.S. International Taxation, 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.


  The views express herein are those of the author and do not necessarily reflect those of Ernst & Young LLP.

  Tricky Derivatives, 38 Tax Mgmt. Int'l J. 721 (Nov. 13, 2009).