My name is H. David Rosenbloom. I am a tax attorney and a professor of tax law. My area of specialization is international, or cross-border, taxation. I am a member of Caplin and Drysdale, a U.S. law firm. I am also director of the international tax program at New York University School of Law. In the late 1970s, I was the international tax counsel in the United States treasury department. In that capacity, I was the chief U.S. negotiator of the 1980 income tax convention between Canada and the United States.
I thank the committee for this opportunity to offer observations on Bill S-17, an act to implement conventions, protocols, and agreements between Canada and various countries. All of these agreements are for the avoidance of double taxation and the prevention of fiscal evasion with respect to income taxes.
My comments are necessarily constrained by both my relative unfamiliarity with Canada's tax treaty policies and the extremely brief amount of time I have had to devote to a study of the bill. I am not a Canadian tax expert, and I was unaware of the bill prior to the afternoon of June 14. Furthermore, I have not been informed regarding the specific aspects of the bill on which I have been asked to comment.
My working assumption is that the committee may be most interested, not in the new conventions with Namibia and Serbia, parts 1 and 2 of Bill S-17, but rather in the convention with Poland and the agreement with Hong Kong, parts 3 and 4; with the protocol to the existing convention with Luxembourg, part 5; and the supplementary convention with Switzerland, part 6.
These last two parts deal with the subject of information exchange. Parts 3 and 4, on the other hand, are a convention and an agreement with jurisdictions, Poland and Hong Kong, that have been used by investors from other countries to invest outside those jurisdictions.
I thus confine these initial comments to the newly proposed exchange-of-information provisions with Luxembourg and Switzerland, and the agreements with the intermediary jurisdictions, Poland and Hong Kong.
I begin with information exchange. The protocol to the convention with Luxembourg appears consistent with the current practices of the Organisation for Economic Co-operation and Development and with the pending protocol to the income tax convention between the United States and Luxembourg.
There are some differences among these texts, but they are of a technical nature, and I assume of relatively little interest to the committee. I have some reservations about the efficacy of such provisions for achieving useful information exchange, but I cannot see them doing any harm.
The supplemental convention with Switzerland, on the other hand, relieves a requesting country from the need to provide a specific name to the requested country in order to obtain information about a person and in order to identify the person in possession of that information.
Since the requesting country is often in need of the name—that is the reason for the request in the first place—a requirement that the name be given in order to obtain the requested information might often render the information exchange provision nugatory. Thus, this supplementary convention responds to a real problem, and despite my abiding skepticism about information exchange via tax convention, I can see no substantial objection to it.
The agreement with Hong Kong and the new convention with Poland are a different and much larger and more complicated matter. A major concern with jurisdictions that lend themselves, and their treaty networks, to investors from elsewhere is the possibility that their conventions become in effect agreements with the entire world.
In the United States, we think that most tax conventions are bilateral in nature and that the benefits they confer should be confined to persons with a genuine connection with one of the treaty partners.
One means of implementing this policy is to simply not enter into conventions with jurisdictions that serve as intermediaries, especially if there is no demonstration of a genuine risk of double taxation. Regrettably, the United States has not always followed this route. We have, for example, conventions with Bermuda, Cyprus, and Barbados, to name but three examples of jurisdictions where the need for a U.S. tax convention would not appear compelling.
Apart from the strategy of not negotiating conventions with certain jurisdictions, the United States relies on certain measures both within the text of its conventions and drawn from its general jurisprudence to combat treaty shopping. A limitation on benefits article requiring a genuine nexus between the party claiming benefits and the treaty partner is now standard in all modern U.S. tax conventions. And the economic substance doctrine, recently enacted into statutory law but of lengthy vintage in our courts, has served as a potent weapon against at least some types of treaty shopping.
I note that paragraph 3 of article 26 of both the convention with Poland and the agreement with Hong Kong represent an abbreviated version of what has become, in the United States, the “limitation on benefits” article. The article 26 provision, which also appears in the conventions with Namibia and Serbia, effectively precludes foreign-owned entities from enjoying a more beneficial regime in the treaty partner than do domestically owned entities. This is where early versions of the U.S. limitation on benefits provision began, but the provision has since gone much further. Whether it has always been effective is an open question.
I conclude by citing a provision of this type that actually seems to work. It appears in the U.S. convention with Cyprus, and it contains two substantive rules of general relevance: first, that U.S. benefits are available only to Cypriot entities that are owned to a large extent, both legally and economically, by genuine individual residents of Cyprus or, in some limited circumstances, by citizens of the United States; and second, that such benefits are allowed when it is determined on a discretionary basis that the establishment, acquisition, and maintenance of the entity and the conduct of its operations did not have as a principal purpose obtaining benefits under the convention. The provision is general and, some might say, unacceptably vague. Yet for that very reason it seems to have succeeded in thwarting attempts by third-country investors to use the Cyprus convention to obtain inappropriate treaty benefits in the United States.
I would add two thoughts on the basis of what I have heard thus far in this hearing. I throw them out for further elaboration. One, I do suggest to the committee that you carefully distinguish between concern about corporate-level tax avoidance, the use of tax havens by multinational companies, and transfer pricing, things that concern the multinational company on the one hand, and things that concern individuals taxpayers on the other. For the most part there we're talking about offshore accounts, the use of offshore trusts, etc. I think we're talking about two related but distinct problems, and I think there ought to be different responses.