eJournal of Tax Research
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
eJournal of Tax Research |
|
Abstract
This paper considers the correct interpretation of double tax agreements in the context of locally resident accumulation trusts established in New Zealand or Australia, that have foreign settlors and foreign-source income. The author explores the relevance and application of double tax agreements intended to minimise double taxation between New Zealand or Australia on the one hand and the source country on the other hand. In the process he identifies a difference in approach in Australia compared to New Zealand as they accord different treatment to accumulation trusts that have foreign settlors and foreign source income. The paper explores the reasons for, and the implications of, the different outcomes highlighting some key issues in tax administration where international tax is concerned.
This paper deals with trusts that are established in New Zealand or Australia, with locally resident trustees, but having foreign settlors and foreign-source income. For purposes of source country tax, it may be relevant whether there is an agreement to minimise double taxation between New Zealand or Australia on one hand and the source country on the other hand. This question can arise in two contexts. First, if by a double tax agreement a source country grants privileges to taxpayers who are “resident” in terms of the treaty in the destination country, will the source country treat a trustee who is resident in the destination country as so “resident” for purposes of the source country’s taxation rules? Secondly, where a country agrees to reduce or eliminate withholding tax on outward-flowing passive income that is received by beneficial owners if they are resident in the jurisdiction of the treaty partner, will the source country treat a trustee for foreign beneficiaries as a “beneficial owner” in terms of the treaty?
These questions relate particularly to accumulation trusts, the subject of this paper. Where trusts do not accumulate, but distribute income as it arises, most jurisdictions that recognize the concept of the trust treat the beneficiary as the taxpayer, not the trustee. That is so in respect of both fixed and discretionary trusts.[1] As a result, in the context of distributing trusts, because the beneficiary is the taxpayer it is the residence of the beneficiary that matters, not the residence of the trustee.
In principle, the questions set out in the first paragraph of this paper raise the same issues and are answered in the same way in both New Zealand and Australia. In practice, although the answers to the questions are the same in both countries there are significant second-order differences. The reason is that New Zealand and Australia accord different treatment to accumulation trusts that have foreign settlors and foreign source income.
Trusts with foreign settlors and foreign source income will often have foreign beneficiaries also. Such trusts are particularly likely to have foreign beneficiaries if they were established pursuant to a plan that is designed to take advantage of the New Zealand tax regime in respect of such trusts, which is described in the next paragraph.
In New Zealand, where New Zealand resident trustees accumulate income in respect of trusts with foreign settlors and foreign source income there is generally speaking no New Zealand tax. The rule is in section HH 4(3B) of the New Zealand Income Tax Act 1994. In contrast, Australia generally, and for practical purposes invariably, taxes accumulations to trustees of such trusts if they are resident in Australia. One result of this difference is that it is fairly common for foreign settlors to appoint New Zealand resident trustees to administer trusts of foreign property that they settle, whereas it is rare for foreign settlors to appoint Australian trustees, at least where there are no Australian beneficiaries or property.
Because there are a good many New Zealand resident trustees who administer trusts that are in other respects foreign, the issue arises as to how other countries that have tax treaties with New Zealand tax outward-flowing income that is bound for New Zealand resident trustees. This general issue gives rise to the specific questions set out in the first paragraph of this paper. As mentioned, readers should bear in mind that the answers to these questions are generally the same in respect of countries that have treaties with Australia as in respect of countries that have treaties with New Zealand, though in Australia the issue is generally less likely to arise in practice.
Positive answers to the questions are not crucial to all plans to employ New Zealand-resident trustees to receive foreign source income, but positive answers can significantly add to the attractiveness of that strategy. This paper addresses the subject from the point of view of New Zealand double tax agreements, of which there are twenty-seven.[2]
Suppose that a trustee of foreign-source income is clearly resident in New Zealand for purposes of the Income Tax Act 1994. Does it follow that New Zealand’s treaty partners must treat that trustee as resident in New Zealand for tax purposes and for purposes of the provisions of their several treaties? In most New Zealand treaties, the residence clause is based on Article 4(1) of the 1977 OECD model convention, which reads:
For the purposes of this Convention, the term ‘resident of a Contracting State’ means any person who, under the laws of that State, is liable to tax therein by reason of his domicile, residence, citizenship, place of management, or any other criterion of a similar nature. But this term does not include any person who is liable to tax in that State in respect only of income from sources in that State or capital situated therein. [It is convenient for purposes of this paper to call the clause beginning “But this term ...” the “proviso” to Article 4(1).]
Most of New Zealand’s treaties follow Article 4(1) of the OECD Model or a simplified version of it: Belgium, China, Denmark, Finland, France, Germany, India, Ireland, Italy, the Netherlands, Norway, the Philippines, and Switzerland. The treaty with Indonesia omits the proviso or any counterpart of it. The Swedish treaty and most treaties with British Commonwealth countries have a much simpler format, but a format that, like the Indonesian treaty, contains no proviso: Australia, Canada, Fiji, Malaysia, Singapore, Sweden, and the United Kingdom. The New Zealand-United Kingdom Article 4(1), for example, reads:
For the purposes of this Convention, the term “resident of a Contracting State” means, as the context requires: .... (b) any person who is resident in New Zealand for the purposes of New Zealand tax.
The relevant terminology in the Japan treaty is to a similar effect but takes a different form. The treaty with the United States, which is influenced by the United States model draft, contains a second proviso, subparagraph (b) in the relevant article:
In the case of income derived or paid by a partnership, estate, or trust, this term applies only to the extent that the income derived by such partnership, estate, or trust is subject to tax in that State as the income of a resident, either in its hands or in the hands of its partners or beneficiaries.
These various versions of Article 4(1) pose the question: where a trustee is resident in New Zealand in the ordinary meaning of the term, but is not taxable on trustee income in New Zealand because the income is foreign and no settlor has any New Zealand connection, can the trustee be considered to be resident in New Zealand for treaty purposes? The United States formula poses the question most acutely. It is arguable that the question must be answered by reference to liability as a trustee alone, because New Zealand treats a trustee’s income independently from the trustee’s personal income.[3]
Robert Venables takes the view that mere residence of a trustee within the jurisdiction of a treaty partner does not give a trust foreign-residence status vis-à-vis the United Kingdom revenue authorities. He says in Non-Resident Trusts[4]
It should be borne in mind that even if the trustee is resident in a jurisdiction which has a DTT[5] with the UK and for the purposes of that treaty is regarded as a resident of the other jurisdiction and not of the UK, the treaty will not deem him to be non-resident in the UK for any purpose other than that of his own taxability. In particular, it will have no bearing on the question of where the trust of which he is trustee is resident.
Venables’s opinion may be the view that would prevail among the United Kingdom Commissioners of Inland Revenue. Moreover, that opinion is probably the better view in general. On the other hand, a study by John Avery Jones and others published in 1989 took the view that at least in the United States, the United Kingdom, Canada, and Australia, “a trustee is reasonably clearly to be treated as a person, who can therefore probably be a resident of a state for treaty purposes.”[6] Avery Jones points out that this conclusion fits awkwardly with treaty rules that apply to resolve dual residence issues that arise for trustees, because the criteria by which treaties determine the residence of individuals (availability of a permanent home, centre of vital interests, an habitual abode, and nationality)[7] are not appropriately applied to persons in their capacity as trustees. Referring initially to Canadian law, Avery Jones explains:[8]
It is arguable that because a trustee in his capacity as such does not have a permanent home, a centre of vital interests, an habitual abode or a nationality (particularly in the case of a corporate trustee), the tests set out in article 4(2) are inappropriate and therefore inapplicable. If this argument prevails, article 4(3) applying the place of effective management to persons other than individuals, would presumably apply. ... Article 4(3) would apply in the United Kingdom, where a trustee is not treated as an individual, and the United States, and Australia.
Avery Jones’s argument appears to be that although, as explained in the previous paragraph, a trustee is a person, and therefore can be resident somewhere for treaty purposes, it is not appropriate to determine the residence of a trustee by tests that apply to individuals. The need to make this argument gives rise to the question as to whether Avery Jones’s basic premise can be correct. If a trustee is a “person” for treaty purposes, but if trustees who are individuals are not to be treated as such, the text of Article 4(2) and (3) does not aptly apply to them. But if this is so, why should we assume that Article 4(1) is meant to apply to trustees as well? That is, one could argue that a state that adopts Article 4(2) and (3) is perhaps assuming that none of Article 4 (in particular, not Article 4(1)) applies to taxpayers in their capacity as trustees. This conclusion leads back to the argument of Robert Venables, discussed above, that the personal residence of individuals does not determine their residence as trustees if they happen to be trustees. Nevertheless, it is worth examining particular treaty drafts to determine whether there are possible contrary views, at least in respect of some versions of Article 4(1). One thesis of this paper is that, all things considered, these contrary views must command considerable respect.
As mentioned in section 2 of this paper, subject to certain exceptions, section HH 4(3B) of the New Zealand Income Tax Act 1994 provides that where a trustee is resident in New Zealand, that New Zealand residence does not render the trustee liable for New Zealand tax on foreign source income that he or she receives as trustee.[9] How does this rule mesh with the several versions of the residence rule in New Zealand’s treaties?
Where it is a question of the taxation of income that is sourced in the jurisdiction of one or other of New Zealand’s treaty partners, the answer to this question is ultimately a matter for the courts of those treaty partners. Nevertheless, because countries try to adopt similar approaches to the interpretation of double tax treaties, one can reasonably make an attempt to identify the relevant questions and sometimes to answer them.
New Zealand resident trustees may be either individuals or companies. In the case of a company the residence of directors does not make any difference to the fiscal residence of the company from the point of view of New Zealand domestic law.[10] Double tax agreement rules on residence of companies are generally to the same effect. Both the OECD model and the incarnations of that model in New Zealand’s tax treaties take a fairly formal approach to the question of the separate personality of companies. As a result, by the terms of those treaties residence of shareholders and directors is not directly relevant to the determination of fiscal residence of a company.
What is the import of Article 4(1) of the OECD model and its counterparts in New Zealand’s treaties? Take first the United States treaty. It is the only agreement that contains text to the effect of its subparagraph (b), which was quoted earlier.[11] Suppose that a New Zealand resident company derives United States source income, income that the company receives beneficially and that arises from United States investments that the company owns beneficially. In those circumstances, and in respect of that income, vis-a-vis the United States Internal Revenue Service the company must be treated as a New Zealand resident that can make use of the protections that are in the treaty. The same applies to an individual taxpayer who is resident in New Zealand.
Suppose, however, that the company receives the income as a trustee. Then, as far as that income is concerned, it would seem that the company would not be a New Zealand resident for the purposes of the United States treaty, because the income would not be “subject to tax in [New Zealand] as the income of a resident.” Accordingly, the income could not claim treaty protection from United States taxation. This company, receiving income as a trustee, may be taken as a generic example of a New Zealand resident, either a company or an individual, who is trustee of a trust where settlor and income are foreign.
Secondly, take a residence definition from a treaty that includes a counterpart to the proviso to Article 4(1) but no counterpart of the United States subparagraph (b): say the treaty with France. The relevant text of the proviso is: “[resident] does not include any person who is liable to tax in [New Zealand] in respect only of income from sources in [New Zealand].” In respect of French-source income, does this text disqualify the company from treaty protection?
When the treaty is strictly interpreted, the answer seems to be no, the proviso does not disqualify the company from protection. The important factor is that the fulcrum of Article 4(1)(a) is the taxpayer (that is, the resident), not the tax. Unlike the New Zealand-United States paragraph (b) that is considered above, Article 4(1)(a) does not focus on particular income streams, treating a taxpayer as resident in respect of one income stream and not in respect of another. In contrast, Article 4(1)(a) simply postulates a test whereby the taxpayer either is or is not resident for treaty purposes.
As a resident of New Zealand, the company is liable to New Zealand tax on both domestic income and foreign income if the company should happen to derive income of either type. This is the position whether the company derives the income beneficially or as a trustee. In respect of trustee income, the rule is subject to the exception that is vital to the theme of this paper. That is, section HH 4(3B) of the New Zealand Income Tax Act 1994 simply relieves the company from liability for tax on foreign source income that it derives as a trustee where, to generalize, there is no settlor resident in New Zealand.
Article 4(1)(a) of the New Zealand–France treaty is drafted on an all-or-nothing basis. For treaty purposes, the taxpayer either is, or is not, a resident of the relevant contracting state. Looking first at the “all” basis of this dichotomy, it follows that the correct interpretation of Article 4(1)(a) seems to be that if a taxpayer who is resident in one of the countries is liable to tax on at least some types of foreign source income if the taxpayer were to derive those types of income, then the taxpayer is to be treated as a resident of that country for all treaty purposes. It follows, for instance, that if a New Zealand-resident taxpayer is protected by the treaty in respect of any income that has its source in France, then the treaty protects the New Zealand resident in respect of all French-source income (assuming that the income is income of one of the kinds that enjoys treaty protection).
A consideration of the possible alternative (that is, the “nothing” alternative in the all-or-nothing dichotomy) reinforces the interpretation offered here. The “nothing” alternative would say that if someone who appears to be a New Zealand resident does not qualify for treaty protection in respect of one kind of income (in this case, income derived as a trustee that New Zealand chooses not to tax) then that apparent resident would appear not to be a resident for any purposes of the relief of double taxation as far as the treaty is concerned. This argument can hardly be sustained. First, it has no support from the text of the treaty. Secondly, it would mean that few New Zealanders could benefit under a treaty that is clearly meant to protect them from double taxation.
The problem is that in principle, many, perhaps most, New Zealand resident taxpayers are potential trustees (in that some day someone may appoint them as trustees) and, as trustees, they may one day receive trustee income from France, income that would not be assessable in New Zealand (assuming that the settlors of the trusts in question were foreign). If the New Zealand-France treaty fails to protect the New Zealand taxpayers in these circumstances it would appear not to protect them in any circumstances. That is an absurd result. The New Zealand-France treaty cannot intend that the possibility that New Zealand resident taxpayers might one day derive foreign-source income as trustees (being income that New Zealand does not tax) should disqualify these same New Zealand resident taxpayers from the protection of the treaty in respect of other income that the New Zealand residents may derive. One is forced to reject the “nothing” conclusion and to adopt the “all” conclusion.
A response to the argument just made is that the treaty requires taxpayers to be treated as two notional persons. In the case of companies, the first is an artificial person that is taxed as an ordinary corporate taxpayer. The second is the company in its capacity as a trustee. Qualification for or disqualification from treaty protection of someone in the company’s second (trustee) capacity has no effect on the company’s treaty rights as an ordinary corporate taxpayer.
This response is attractive from a policy point of view. No doubt, that is how treaties should work.[12] The problem is that the text of a double tax convention that is drafted in terms of the New Zealand-France treaty offers no support for this interpretation. One can argue that a taxpayer who is a trustee cannot be resident in respect of trust income, but the text contains no basis for saying that a taxpayer can be resident for some fiscal purposes and not for others. More so, if a taxpayer that is a trustee is excluded from treaty benefit, why stop at that point? What about taxpayers who might one day be trustees? Such a result seems to be compelled by the logic of the “nothing” alternative, but it would be absurd to exclude taxpayers from treaty benefits on the basis that they might one day become trustees. As argued above, it seems to follow that trustees may take treaty benefits, both as trustees and in their own capacities as taxpayers.
A more substantive interpretation might suggest that the focus should be neither on the taxpayer, nor on whether the taxpayer is potentially liable to tax on some foreign source income, but on items of foreign source income, on a case-by-case basis. Is the New Zealand taxpayer assessable to New Zealand tax on this particular income from France, being trustee income? If not, the argument runs, the treaty should not protect the income. The answer to this argument is that the proviso deals with persons, not items of income. A person either is, or is not, entitled to treaty protection. Article 4(1)(b) of the New Zealand-United States treaty, on the other hand, considers income on an item-by-item basis, as already explained. When New Zealand’s treaties were drafted, income on an item-by-item basis could have been added to the treaty text to the effect of the New Zealand-United States Article 4(1)(b). The omission arguably shows an intention to retain the taxpayer-by-taxpayer approach that is mandated by a literal interpretation of the proviso. The better view, therefore, is that the French treaty, and others like it, protect income derived by New Zealand resident trustees even where the receipts are trustee income and not subject to New Zealand tax.
It follows with greater force that where the corresponding article takes the simplified form that is found in the United Kingdom treaty,[13] foreign source trustee income derived by New Zealand resident trustees enjoys treaty protection in the same manner. This conclusion, and the conclusion in the previous paragraph, may be modified by consideration of the second question raised by double tax agreements that is relevant in the present context. This question is whether a trustee can be said to derive income as a “beneficial owner”, which is commonly a pre-requisite for treaty protection, at least in respect of passive income. That question is considered next.
Apart from the treaty with Japan, all New Zealand’s double tax agreements provide for a reduction in withholding tax on interest, dividends, and royalties that flow between parties resident in New Zealand and the relevant treaty partner, provided that the recipient is the beneficial owner of the income in question. In this respect, like other countries, New Zealand follows the OECD treaty model. The 1963 OECD model omitted the “beneficial owner” requirement, but it has been present since 1977. The OECD model royalty articles are reproduced here by way of example. Articles 12 and 13 of the New Zealand-United Kingdom treaty, which apply to interest and to royalties, follow, as examples of adaptation of the OECD model in practice.
OECD Article 12(1) Royalties 1963. Royalties arising in a Contracting State and paid to a resident of the other Contracting State shall be taxable only in that other state.
OECD Article 12(1) Royalties 1977. Royalties arising in a Contracting State and paid to a resident of the other Contracting State shall be taxable only in that other state if such resident is the beneficial owner of the royalties.
OECD Article 12(1) Royalties 1997 and 2003. Royalties arising in a Contracting State and beneficially owned by a resident of the other Contracting State shall be taxable only in that other state. [Amendment to the model made in 1995.[14]]
New Zealand-United Kingdom treaty signed 1983.
Article 12(1). Interest arising in a Contracting State which is derived by a resident of the other Contracting State may be taxed in that other State. However, such interest may also be taxed in the Contracting State in which it arises and according to the law of that State but where the beneficial owner of such interest is a resident of the other Contracting State the tax so charged shall not exceed 10 per cent of the gross amount of the interest.
Article 13(1). Royalties arising in a contracting state which are derived by a resident of the other contracting state may be taxed in that other state.
Article 13(2). However, such royalties may also be taxed in the Contracting State in which they arise and according to the law of that State, but where the beneficial owner of such royalties is a resident of the other Contracting State the tax so charged shall not exceed 10 per cent of the gross amount of the royalties.
A number of New Zealand’s treaties use “beneficial ownership” or “beneficially entitled”[15] in their passive income provisions, rather than “beneficial owner”, but there appears to be no difference in meaning. For simplicity, the present paper uses “beneficial owner” as an appropriate generic term.
The Japan treaty has no limitation for tax on interest or royalties, but caps tax on outward-flowing dividends at fifteen per cent. Uniquely in New Zealand’s treaties, although this limit is available to recipients of dividends who are “resident in the other Contracting State”, there is no requirement that such residents must own the dividends beneficially.
If a New Zealand resident trustee were to receive, say, royalties, from someone in the United Kingdom, would the trust be able to take advantage of the ten per cent limitation on withholding tax on royalties that is imposed in the United Kingdom-New Zealand Treaty? The answer to this question is a matter for the law of the United Kingdom, rather than for New Zealand law; the answer depends on how United Kingdom courts would interpret the double tax agreement. Similarly, in respect of interest, dividends, or royalties received by the trustee from any other jurisdiction with which New Zealand has a double tax agreement, the effect of the beneficial ownership requirement is a question for the courts of that jurisdiction.
Having said that, one should note that in a number of countries the beneficial ownership condition is thought to have relatively little effect in practice. That is, taxpayers pay out dividends, interest, and royalties to people or companies that are residents of treaty partner countries, and, reputedly more often than not, deduct withholding tax at only the reduced treaty rate without questioning whether the recipient is the beneficial owner of the income. This is said to happen without adverse reaction by the revenue authorities in the source country, though some fiscal authorities make regular inquiries of people who pay out passive income to alleged “beneficial owners” in other jurisdictions, and other fiscs are becoming more interested in the question.[16]
The approach that is reputedly a fairly general practice has some justification in law, in that the OECD draft model on which the New Zealand treaties are based does not clearly intend that the beneficial ownership limitation should exclude trustees from treaty benefits. For instance, the official Commentary to Article 12, Royalties, of the 1977 draft (in force when most of the New Zealand treaties were signed) states:
Under paragraph 2 [of the draft convention] the limitation of tax in the State of source is not available when an intermediary, such as an agent or nominee, is interposed between the beneficiary and the payer, unless the beneficial owner is a resident of the Contracting State (the text of the Model was amended in 1995 to clarify this point, which has been the consistent position of all member countries). States who wish to make this more explicit are free to do so during bilateral negotiations.
The Commentary on Article 12 was slightly modified in 1995, and the words in brackets were inserted into the ambulatory OECD model and Commentary that was extant from 1997 to 2003. There were similar amendments to the Model in respect of Articles 10, Dividends, and 11, Interest.
What the Commentary stated is true: that is, the Model Convention text was amended, and the new versions of the articles do make a little clearer “this point”, being the point that treaty benefits are intended for beneficial owners only; but there has never been serious doubt on that score. Where doubt exists is, what is meant by “beneficial owner”? The 1995 amendments to the Model Convention and the 1997 amendments to the Commentary shed no light on this question. The Commentary changed again in the 2003 edition of the Model, to read in respect of Article 12:[17]
The requirement of beneficial ownership was introduced in paragraph 1 of Article 12 to clarify how the Article applies in relation to payments made to intermediaries. It makes plain that the state of source is not obliged to give up taxing rights over royalty income merely because that income was immediately received by a resident of a State with which the State of source has concluded a convention. The term “beneficial owner” is not used in a narrow, technical sense, rather, it should be understood in its context and in light of the object and purpose of the Convention, including avoiding double taxation and the prevention of fiscal evasion and avoidance.
It is helpful to focus first on the Commentary as it read before the amendments of 2003. As explained, the Commentary said that it was “an intermediary such as an agent or nominee” who was intended to be excluded from treaty benefits. It is doubtful whether a trustee can sensibly be called an “intermediary” at all, but a trustee is certainly not “an intermediary such as an agent or nominee”.
In common law jurisdictions there is considerable difference between a trustee, on the one hand, and an agent or nominee on the other hand. As recipients of income, the latter have no relevant status independent from their principal. Use of an agent or a nominee to receive income is in substance not much different from using an address of convenience. Trustees, in contrast, have functions that are independent of their beneficiaries and settlors (apart from bare trustees, who should be treated as nominees). On the other hand, and again in common law jurisdictions, the very essence of ownership as a trustee is that it is different from beneficial ownership. From a common law point of view the whole point of being a trustee is that one is not a beneficial owner.
Nevertheless, some commentators argued that, at least where trustees accumulate income, and are not obliged to distribute it in the same year to beneficiaries, the trustees do have beneficial ownership of the income in the sense that is required by the standard formulation in double tax conventions.[18]
The 2003 amendments both erode and reinforce this argument. The erosion is by virtue of the deletion of the words “such as an agent or nominee”, which tended to limit the scope of the meaning “intermediary” to someone who, unlike a trustee, has no powers independent of the principal. The deletion of these words potentially allows greater scope to “intermediary”, scope that might include “beneficiary” in its technical sense as used by common law jurisdictions in the law of trusts. On the other hand, the 2003 version of the Commentary immediately dispels this argument by saying that “the term ‘beneficial owner’ is not used in a narrow, technical sense”. The only “narrow, technical sense” of “beneficial owner” comes from the common law of trusts. The implication is that in the view of the OECD Committee on Fiscal Affairs, who wrote the Commentary, at least some trustees may be “beneficial owners” for purposes of the relevant treaty articles. That is, at lease some trustees are not disqualified from treaty benefits by a narrow, technical interpretation of “beneficial owner”.
A number of New Zealand’s treaties address the question of the interpretation of “beneficial owner” or “beneficial entitlement” by provisions in the interpretation article, which is ordinarily Article 3. There is a typical example in the agreement with Canada:
In determining, for the purposes of Articles 10, 11, or 12, whether dividends, interest, or royalties are beneficially owned by a resident of a contracting state, dividends, interest, or royalties in respect of which a trustee is subject to tax in that Contracting State shall be treated as being beneficially owned by that trustee.
Such provisions in treaties to which New Zealand is a party have their origin in a New Zealand observation to Article 3 (General Definitions) of the OECD Model:
For the purposes of Articles 10, 11, and 12, New Zealand would wish to treat dividends, interest, and royalties in respect of which a trustee is subject to tax in the State of which he is a resident as being beneficially owned by that trustee.
The New Zealand observation and provisions in treaties that reflect it are relevant to cases where trustees are subject to tax in the state of residence. They do not directly address the question that is at issue in this paper: whether trustees who derive passive income can be described as “beneficial owners” of that income even thought they are not subject to tax on the income in their state of residence. Indirectly, however, the New Zealand-Canadian provision and others like it suggest that trustees are not “beneficial owners” as the expression is used in treaties, (otherwise there would be no need for the provision) and, if they are to be treated as beneficial owners when they bear tax on the income in question, there must be a special rule to enable that to happen.
It follows that, to put it at its lowest, if trustees are to argue that they are entitled to double tax agreement relief that depends on beneficial ownership status, then they start with a handicap if the agreement that they rely on contains a provision of the kind just discussed. A further consideration is that the words in the New Zealand observation, “treated ... as beneficially owned”, suggest that the ordinary meaning of “beneficially owned” would not embrace income derived by a trustee.[19] The New Zealand agreements that contain provisions that reflect the observation are those with Australia, Belgium, Canada, Denmark, Fiji, Finland, Korea, Malaysia, Norway, Singapore, and Sweden. This paper passes now to consider whether, even in the absence of such a provision, in general principle trustees can be thought of as “beneficial owners” for double tax agreement purposes.
Working at the International Bureau of Fiscal Documentation, Amsterdam, Mr C.P. du Toit completed in 1999 a doctoral thesis on Beneficial Ownership of Royalties in Bilateral Tax Treaties, under the supervision of Professor J.W. Zwemmer of the University of Amsterdam. While the thesis is directed to royalties only, du Toit’s work is equally relevant in respect of dividends and interest. Addressing the meaning of “beneficial ownership” du Toit adopts the following analytical framework, which appears to have been correct at the time when he wrote (the 2003 amendments to the commentary on the OECD Model may necessitate reconsideration):[20]
The OECD Model does not define “beneficial owner”. Similarly, most, if not all, treaties contain no definition. There are therefore two possibilities. First, Article 3(2) of the Model Convention or its counterpart in individual treaties requires interpretation to be according to the domestic law of the treaty partner whose courts are seized of the question in any particular case, unless the context otherwise requires. Secondly, despite the apparently imperative terms of Article 3(2) that domestic law should apply, it may be that “beneficial owner” should be classified as part of an “international tax language” and interpreted in a consistent manner across all tax conventions that follow the OECD Model. That is so even though there is nothing in standard treaty articles on dividends, interests, or royalties that suggests that “beneficial owner” should not be subject to the ordinary Article 3(2) rule of interpretation according to domestic law.
Du Toit’s conclusion is that in most cases where “beneficial owner” falls to be interpreted the practical result of the two lines of reasoning is the same. If one resorts to domestic law interpretations, at least in common law countries it is hard to depart from the ordinary meaning that the term bears in the context of the law of trusts, that is, “equitable owner” as opposed to owner as trustee.[21]
On the other hand, if the second line of reasoning is correct, there is still the question of deciding what “beneficial owner” means as a term of an international tax language. In pursuing this inquiry, it is logical to start with legal systems that recognize the term, being common law systems. This route leads again to the conclusion that “beneficial owner” bears the meaning that it enjoys in the common law.
Contrary to de Toit’s opinion, there are several arguments that the expression “beneficial owner” can include trustees. First, it was only in the 1977 draft that the OECD Model Convention inserted the beneficial ownership requirement into the dividends, interest, and royalties articles. The former, 1963, draft had no such reference.[22] Also in 1977, the OECD decided to refer in the Commentary to nominees and agents only, and not to trustees, as examples of people who are not beneficial owners. Nevertheless, by 1977 the question of whether trustees could be beneficial owners for treaty purposes had been a live issue for some years. Considering that people were well aware of the issue at the time, reference to trustees cannot have been omitted by accident. Had the OECD intended to exclude trustees in the Model as possible beneficial owners it could have done so. Moreover, there was no change in the ambulatory draft of the Model that was promulgated in 1995.
The 2003 amendments to the Commentary that have been discussed[23] reinforce the argument. If the question of whether “beneficial owner” can include some trustees was a live issue in 1977, by 2003 the OECD Committee on Fiscal Affairs can have been in no doubt that this question was a major, perhaps the major, issue in the interpretation of relevant articles in the model. The fact that, in these circumstances, the Committee said that “the term ‘beneficial owner’ is not used in a narrow, technical sense” is most significant. Until publication of the 2003 edition of the Model, du Toit’s arguments that “beneficial owner” should be interpreted according to trustee law in common law countries had a good deal of traction. From 2003, the prohibition on interpreting the term in a narrow, technical sense can refer only to its sense in the law of trusts. If so, the contention that trustees can be “beneficial owners” for treaty purposes becomes increasingly telling.
Secondly, at least some civil law jurists, whose legal systems do not know the trust, appear to be comfortable with categorizing trustees as beneficial owners. Vogel asserts that:
The “substance” of the right to receive certain yields has a dual aspect. The first is the right to decide whether or not a yield should be realised – ie, whether the capital or other assets should be used or made available for use – the second is the right to dispose of the yield. ... recourse to the treaty is justified – ie is not improper – if he who is entitled under the private law is free to wield at least one of the powers referred to. Hence, the “beneficial owner” is he who is free to decide (1) whether or not the capital or other assets should be used or made available for use by others or (2) on how the yields therefrom should be used or (3) both.[24]
Vogel’s criteria certainly embrace those who the common law calls trustees. Recognising that the element of control is typically crucial to the trust, Vogel submits that trustees may correctly be called “beneficial owners” for tax treaty purposes.[25]
Romyn, a Dutch commentator, appears to take the view that because of the terms of the OECD Commentary, the only recipients who are not to be regarded as beneficial owners are nominees and agents. He says:
In my opinion neither text nor the commentary of the OECD Model Convention allows leeway for applying the provision at issue [beneficial ownership] outside formal “agency” and “nominee” relationships.[26]
Romyn is not alone in taking this formalistic view. Commentators have expressed similar opinions in both Belgium and France.[27] Since 2003, when the reference to nominees and agents was removed, Romyn’s argument has lost some of its force, though he would no doubt take comfort from the Commentary’s eschewing of the “narrow, technical sense” of “beneficial owner” that has been discussed.[28]
Even in common law countries, there are arguments for the proposition that in referring to agents and nominees the OECD meant to focus on agents and nominees alone, and did not intend the commentary to extend to the more difficult case of trustees.
The first argument is historical. It is based on MacMillan Bloedel Ltd v Minister of National Revenue,[29] where the Canadian court held in effect that even a mere nominee was entitled to benefit from the reduced tax on dividends afforded by the Canada-United States treaty so long as the nominee was resident in the United States and was registered as owner of the Canadian shares in question. (The then treaty had no beneficial ownership rule.) That is, the MacMillan Bloedel case can perhaps be cited for the view that at the time of the 1977 draft the OECD was indeed concerned with such formalistic recipients as nominees and agents, and was not trying to get to grips with trustees, whose ownership rights are much more substantial.
The second argument is based on assumed policy of the drafters of the OECD Model. The argument is that if for treaty purposes some trustees fail to qualify as beneficial owners, it follows that no trustees so qualify.[30] That is, no trustee can ever take advantage of treaty provisions to minimise double taxation on income that the trustee accumulates. That cannot be the policy of the Model, it is argued.
The premises of this argument are robust. The argument notes that the contention that trustees cannot be “beneficial owners” starts from the clear common law distinction between trustee and beneficiary. Because at common law a trustee is clearly not a beneficiary it follows according to the contention that a trustee can never be a beneficial owner of income for treaty purposes.
Logically, it is possible that the drafters of the OECD Model intended that trustees should never qualify for treaty benefits; but that conclusion is implausible. Bear in mind that the issue was certainly before the minds of drafters, even if only by virtue of the New Zealand observation discussed in section 12 of this paper.
This second argument draws our attention to a fundamental characteristic of the overall problem of whether trustees can qualify for treaty benefits: this characteristic is that the answer must inexorably be all or none in paradigm cases. Consider treaties where there are no relevant anti-avoidance rules, no rules that limit trustee residence to cases where the trustee is taxed domestically on the income in question, and no other rules that provide a definite answer to the question of whether trustees can be classified as beneficial owners. Such treaties are the paradigm cases for this paper. An example is the New Zealand-United Kingdom treaty. To quote again from Article 12:
Interest arising in a Contracting state which is derived by a resident of the other Contracting State may be taxed in that other State may ... be taxed in the Contracting State in which it arises ... but where the beneficial owner of such interest is a resident of the other Contracting State the tax so charged shall not exceed 10 per cent of the gross amount of the interest.
Consider first a New Zealand resident trustee who derives and accumulates interest from the United Kingdom, but who is not taxable on the interest because there is no New Zealand resident settlor. One argument is that the trustee should not be entitled to treaty benefits because New Zealand will not tax the income.[31] But suppose that the same trustee distributes the income to a New Zealand resident beneficiary in a later year. New Zealand will tax this beneficiary. Indeed, New Zealand will impose tax at a penal rate in response to the deferral that the income has enjoyed since the trustee derived it.[32] In these circumstances, it would not seem unreasonable for the trustee to enjoy a treaty benefit. Whichever set of facts obtains the New Zealand-United Kingdom treaty must be interpreted in the same way: either “beneficial owner” includes a trustee who accumulates or it does not. Those who advance the argument now under discussion say that only a positive answer is consistent with the general policy of double tax conventions.
A third argument is that the French text of the OECD Model, supports the proposition that “beneficial ownership” included ownership by a trustee. The French expression is “bénéficiare éffectif”. Because French law does not recognise trusteeship “bénéficiare éffectif” includes both full owners and trustees. Since the French and English texts of the Model are equally authoritative it follows that the English text must have the same meaning. Admittedly, this meaning does not necessarily travel the long route from the French version of the Model to a New Zealand treaty. However, the broad adherence to the OECD Model that is apparent in all bilateral tax conventions indicates a strong international commitment to consistency of interpretation.
The matters discussed in the foregoing paragraphs may offer some comfort to trustees who receive passive income from foreign countries that are treaty partners of the trustees’ jurisdictions of residence. There is much to be said for the point of view that “beneficial owner” should be interpreted in the same manner in all treaties, and for the argument that Vogel’s civil law meaning of the expression should be considered as some kind of lowest common denominator. Although it would be rash to claim that this lowest denominator is established as the law, the 2003 amendments to the Commentary[33] add a good deal of force to Professor Vogel’s arguments.
New Zealand has no general capital gains tax, but some gains that are capital according to ordinary concepts are taxed as income. The primary example is gains in respect of financial arrangements. For example, discounts on bonds and foreign exchange gains or losses are treated as revenue items.[34] A second example is gains that result from profit-making schemes, or gains from the sale of property that was acquired with the intention of sale.[35] The tax bite is more comprehensive when the property in question is land.[36] Thirdly, the principle in Californian Copper Syndicate Ltd & Reduced v Harris[37] applies in New Zealand, with effect that, inter alia, financial institutions that as part of their business have to turn their investments over now and then must treat those investments as being on revenue account. However, these considerations are not likely to be relevant to trustees of foreign-source income where beneficiaries and settlors are also foreign.
A further consideration is that in New Zealand the same rules apply to capital gains that are assessable as income as apply to income in general. That is, if a New Zealand resident trustee derives a foreign-source gain that is potentially taxable as income, the gain, like an income receipt, will not be assessable if there is no connection to the trust by way of a New Zealand resident settlor or beneficiary.
Most of New Zealand’s double tax agreements contain a clause like Article 13(7) of the United States treaty, which reads:
Income or gains from the alienation of any property other than property referred to in the previous paragraphs of this Article shall be taxable only in the Contracting State of which the alienator is resident.
Article 13(7) can perhaps be referred to as a “residual property exemption provision”. The gains referred to in the “previous paragraphs” of the article are, in brief, gains on the disposal of real property; on shares in companies that own real property principally or wholly; on property that forms part of a permanent or fixed establishment of a business; or on ships or aircraft. Important exceptions include corporate shares, bonds, and debentures that are held as ordinary investments. That is, if the words are given their plain meaning, a sale by a New Zealand resident trustee of “residual property”, such as shares in a United States company, is not assessable for capital gains tax in the United States. Nor is the sale assessable in New Zealand because (a) there is no capital gains tax and (b) the gain is in any event foreign source income.[38]
The fact that the gain accrues to a trust rather than to a beneficial owner does not appear to be significant. The benefits that Article 13 of the United States-New Zealand agreement accords to residents of the two countries are not confined to gains that people derive as beneficial owners. That position is the same in respect of other New Zealand treaties that contain a similar article, viz. New Zealand’s treaties with Belgium, Denmark, Finland, France, Germany, Indonesia, Korea, and with the Philippines.
There is no alienation of property article in New Zealand’s treaties with Fiji, Japan, Malaysia, or Singapore. The relevant articles in the treaties with Australia, Sweden and Canada do not contain residual property exemption provisions like the New Zealand-United States rule. New Zealand’s treaty with Norway has a residual property exemption provision, but there are certain limitations in respect of property that comprises substantial participatory shareholdings.
The New Zealand-Ireland treaty has a residual property exemption provision, but it is followed by a proviso of uncertain scope:
Provided that where under the law of that Contracting State [the state of residence of the alienator] an individual, in respect of such gains, is subject to tax thereon by reference only to the amount thereof which is received in that Contracting State, the foregoing provisions of this paragraph shall not operate in relation to so much of such gains as is not received in that Contracting State.
In spirit, this proviso is presumably meant to withhold the benefits of the New Zealand-Ireland residual property exemption provision from taxpayers who are not assessable on gains derived from alienating property located in the other jurisdiction. If so, a New Zealand-resident trustee could not use the residual property exemption provision to escape tax on profits derived from alienating property located in Ireland. However, literally the proviso does not have this effect. It withdraws benefits only where gains would be taxable if received in that alienator’s jurisdiction, but are not in fact received. Where the alienator’s gains are not taxable at all, whether received in the jurisdiction or not, the proviso has no application if it is interpreted literally. The literal interpretation is probably correct, since the terms of the proviso are tolerably clear.
Oddly, the treaty with China seems intentionally to give jurisdiction to tax to the country of situs of movable property, just as it and others do in respect of immovable property and property that is associated with permanent establishments or fixed bases. The net result is that the article as it relates to gains from property located in the jurisdiction of a contracting state is essentially in the following form: the country of situs may tax where the property is type X; it may also tax where the property is type Y; it may also tax when the property is of any other type, except ships or aircraft operated in international traffic: altogether an elaborate way of saying something rather simple. In any event, New Zealand resident trustees get no comfort from the New Zealand-China treaty.
Oddly, the relevant clause in the United Kingdom treaty, Article 14(4), omits the word “only”. Like the corresponding provisions in New Zealand’s other double tax agreements, the clause fairly closely follows Article 13(5) of the 1977 OECD Model Double Taxation Convention, which contains the word “only”. However, the versions that are printed in both the New Zealand Statutory Regulations and in the relevant United Kingdom Statutory Instrument[39] each omit “only”. The author’s checking shows that the word was omitted also from the original signed versions of the treaty. As a result, Article 14 (4) reads:
Income or gains from the alienation of any property other than that referred to in paragraphs (1), (2), and (3) of this Article, shall be taxable [“only” omitted] in the Contracting state of which the alienator is resident.
It is probable that the omission from the New Zealand-United Kingdom convention is inadvertent, and the treaty should be read as if it included “only”. Perhaps the main argument is that otherwise Article 14(4) is pointless, because the country of residence does not need the authority of the treaty to tax the gains in question.
Whether this argument should be accepted is a matter for the United Kingdom courts. If it is accepted, the result appears to be that a New Zealand trustee who derives a gain from the sale of movable property in the United Kingdom (apart, mainly, from property that forms part of a permanent establishment of the taxpayer) is not taxable on that gain.
[+] This paper is a revised and expanded edition of John Prebble, “Accumulation Trusts and Double Tax Conventions” [2001] British Tax Review 69. It takes into account amendments to the Commentary the Model Tax Convention, published as OECD Committee on Fiscal Affairs, OECD Model Convention on Income and on Capital, Condensed Version, Paris, 2003.
[∗] Barrister, Professor & former Dean of Law at the Victoria University of Wellington
[1] See generally, J.F. Avery Jones et al “The Treatment of Trusts under the OECD Model Convention–I [1989] British Tax Review 41, 50 ff.
[2] In April 2004 New Zealand was party to three additional double tax agreements that were signed but not in force.
[3] Income Tax Act 1994 s HH 4(1).
[4] 5th ed 1993, 6.1.2.
[5] Double Tax Treaty (footnote added).
[6] J.F. Avery Jones et al “The Treatment of Trusts under the OECD Model Convention–II” [1989] British Tax Review 65, 66. This article collects the relevant authorities,.
[7] OECD Model Convention Art 4(2) 1977.
[8] J.F. Avery Jones et al “The Treatment of Trusts under the OECD Model Convention–II” [1989] British Tax Review 65, 67, footnotes omitted.
[9] New Zealand Income Tax Act 1994 s HH 4(3B).
[10] Income Tax Act 1994 s OE 2.
[11] Section 3 of this paper.
[12] See also the view of Robert Venables QC, discussed above under heading 3.
[13] Convention between New Zealand and the United Kingdom for the Avoidance of Double Taxation (1983) Article 4 (1): “For the purposes of this Convention, the term ‘resident of a Contracting State’ means, as the context requires: (a) any person who is resident in the United Kingdom for the purposes of United Kingdom tax; or (b) any person who is resident in New Zealand for the purposes of New Zealand tax.”
[14] OECD Committee on Fiscal Affairs, OECD Model Convention on Income and on Capital, Commentary on Article 12, loose leaf, Paris, first published 1997.
[15] Eg, Australia-New Zealand, 1995.
[16] Informal discussions between the author and tax practitioners in several jurisdictions.
[17] OECD Committee on Fiscal Affairs, OECD Model Convention on Income and on Capital, Condensed Version, Paris, 2003, 173.
[18] See below, section 19, “Opinions of civil law jurists as to “beneficial ownership”.
[19] Cf J.F. Avery Jones et al, “The Treatment of Trusts under the OECD Model Convention”, 1989/12 European Taxation 391.
[20] See discussion above, under heading 13.
[21] Eg Wood Preservation Ltd v Prior (Inspector of Taxes) (1968) 45 TC 112, 132-133 per Harman LJ, CA; J Sainsbury plc v O’Connor (Inspector of Taxes) [1991] STC 318, 330 Per Nourse LJ, CA.
[22] Section 11 above quotes examples from different editions of the Model Convention.
[23] See discussion in section 13 above.
[24] K Vogel et al, Klaus Vogel on Double Taxation Conventions (1997) 562, cited by C du Toit, Beneficial Ownership of Royalties in Double Tax Treaties Doctoral Thesis for the University of Amsterdam (1999) 89.
[25] Idem.
[26] M Romyn, “De uitenindelijk gerechtigde; Wie geniet inkomsten voor verdragsdoeleinden? In Van Dijck Bundel (1988) 319, translated and quoted by C du Toit, Beneficial Ownership of Royalties in Double Tax Treaties Doctoral Thesis for the University of Amsterdam (1999) 94.
[27] Authorities are collected in C du Toit, Beneficial Ownership of Royalties in Double Tax Treaties Doctoral Thesis for the University of Amsterdam (1999) 98 – 99.
[28] See above, sections 13 and 18.
[29] (1979) 79 DTC 297, 304.
[30] The writer is indebted to Dr Philip Baker for drawing his attention to this argument.
[31] Income Tax Act 1994 s HH 4(3B).
[32] Income Tax Act 1994 s HH 3(4).
[33] Discussed above, section 13 and 18.
[34] Income Tax Act 1994 subpart EH, known as “the qualified accrual rules”.
[35] Income Tax Act 1994 s CD 4.
[36] Income Tax Act 1994 s CD 1.
[38] Proposition (b) assumes that there is no settlor of the trust who is resident in New Zealand. See above, section 2.
[39] No 365 of 1984 (Part I, page 1041).
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/eJlTaxR/2004/10.html