Australian Tax Treaties and Transparent Entities After the Flood

Australian tax treaties are undergoing the most extensive revision in their history as ratifications under the Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting (MLI) begin to take effect. One area of change concerns the application of treaties to income of fiscally transparent entities.

My recent study (which this post partly summarises) surveys in detail the application of Australian tax treaties to transparent and hybrid entities after the MLI.

Transparent entities

The problem reflects the history of treaty design. In early work of the League of Nations and the OECD, model treaties were designed with individuals and classically taxed corporations in mind. The few references to other entities were confusing and uncoordinated.

This was generally not an issue if both countries’ tax laws agreed about the attribution of income and the status of the entity as transparent (so that its income was attributed to a participant, such as a partner, a beneficiary, or the grantor of a trust) or non-transparent in respect of that income. Otherwise, there was a real risk of double taxation going unrelieved or of taxation in one country being limited or eliminated without actual taxation in the other.

These risks were generally associated with hybridity of the entity — international differences in fiscal status or transparency — in respect of particular income, giving rise to differences in the attribution of that income. These problems became increasingly acute from the last quarter of the 20^th century.

In response, the OECD/G20 Base Erosion and Profit Shifting (BEPS) Project recommended that a transparent entity clause and a saving clause be included in the OECD Model. These two clauses reflected the guiding principles of the majority view in the 1999 OECD Partnership Report, which tried to fix problems in the application of treaties to partnerships by changing the standard interpretation of treaties.

The Partnership Report suffered from two major limitations: several economically significant countries rejected its interpretation of the model treaty text as it then stood; and it only applied to partnerships. It did not apply to trusts, entities that had the option of transparency or non-transparency under US check-the-box regulations, or other non-corporates.

The BEPS recommendations promote the Partnership Report principles to the status of treaty provisions and extend their application to fiscally transparent entities generally. The new provisions run closely parallel in design and drafting to the transparent entity clause and saving clause in US treaty practice, which had served as inspiration for the Partnership Report. They have been included in the 2017 update of the OECD Model and the MLI.

The transparent entity clause operates in a specific and carefully designed way that tries to minimise disruption to the rest of the treaty text. It does not directly confer treaty benefits, but governs access to other provisions which do so — the distributive and double tax relief rules. It opens or closes the door to those rules, and it also (up to a point) governs how they operate. Where income is derived by or through a fiscally transparent entity as perceived by either country, that income has the necessary status of ‘income of a resident of a contracting state’ only insofar as the residence country attributes the income to its resident. The effect is limited to treaty benefits: it does not interfere with the attribution or classification of income in the national law of either country.

The saving clause preserves residence-country taxing rights from treaty abrogation, other than by the double tax relief article and other measures specifically aimed at residence taxation. If the parties to a treaty covered by the MLI both accept (in other words, do not reserve against) the transparent entity clause but either reserves against the general saving clause, a fall-back saving clause is engaged, limited to the context and effect of the transparent entity clause.

The transparent entity clause is designed to avoid double taxation and unintended non-taxation, but only in source-residence situations. Unless some further provision is added, it does not provide for relief against pure dual-residence double taxation.

Australian tax treaties

Four pre-BEPS Australian treaties already deal with transparent entities or some classes of them: the treaties with the United States (1982, 2001), France (2006), Japan (2008) and New Zealand (2009). The study shows that, for a variety of reasons, these treaties will not change the nature or scope of their treatment of transparent entities after the MLI.

The US treaty still has an old-fashioned partial residence clause which only applies to partnerships, trusts and estates and reflects US treaty policy of the 1980s. The clause is notorious for its shortcomings and difficulty of interpretation, which increase cost and uncertainty where such entities are involved. It is unfortunate that the two countries missed the opportunity to add a transparent entity clause when they amended the treaty in 2001.

Sixteen Australian treaties acquire a transparent entity clause for the first time under the MLI, some with a general saving clause and others with a contextual saving clause. The 2003 treaty with the United Kingdom includes a clause that deals with some hybrid partnership cases, but will be replaced by the MLI provisions.

Australia’s two most recent treaties — with Germany (2015) and Israel (2019) — take account of the BEPS recommendations and are not affected by the MLI. Both include a transparent entity clause.

Findings (and maybe some solutions)

The transparent entity clause is an elegant solution to a difficult problem of treaty design and policy, but it is not by itself a complete solution. Some important questions remain unresolved, or at least not beyond dispute. These include the scope of the transparent entity clause and how it interacts with distributive rules and other treaty provisions.

Many countries have special attribution regimes. How can we tell whether one of these attracts the transparent entity clause? Some commentators have suggested that the operation of offshore investment rules may count as residence-country attribution if they treat the income of an offshore entity, such as a controlled foreign company or a trust, as income of an owner or grantor. The issue may not have been fully anticipated by the drafters of the transparent entity clause, but it must be confronted. Ideally, it should be addressed explicitly.

One way to do this is by a super-saving clause, such as Australia has used in its treaties with the United Kingdom, Germany and Israel. This seems to be a developing treaty policy. The super-saving clause says that the treaty does not restrict the application of any legal rule designed to prevent the avoidance or evasion of taxes. It then proceeds to identify particular rules deemed to have that character. The explicit effect is to preserve special anti-avoidance rules. Unlike the general saving clause, the super-saving clause has no exceptions. This is strong medicine. Presumably at the insistence of the German negotiators, the treaty with Germany includes a requirement that, if double taxation results, ‘the competent authorities shall consult for the elimination of such double taxation’. It remains to be seen whether this is adequate protection.

The distributive rules of a treaty allocate primary taxing rights. They generally apply to income of a resident of a contracting state if it has a particular characterisation and connection with the other treaty country. A number of these rules apply additional criteria which can create difficulties where a transparent entity is involved. My study considers several of these: beneficial ownership of income, which is required for treaty benefits in relation to dividends, interest and royalties; the conditions for access to a lower treaty rate for source taxation of intercorporate dividends based on whether a recipient company ‘holds directly’ a sufficient stake in the paying company (the Australia–Germany treaty solves the problem by looking through the transparent entity); and the treatment of business income by reference to whether the relevant person conducts a business through a permanent establishment in the source country (Australia’s typical trust permanent establishment clause looks increasingly anachronistic in this context).

Another issue that has not been satisfactorily resolved (as highlighted by the Resource Capital Fund III and Resource Capital Fund IV cases) concerns the procedure by which a participant can procure treaty benefits in a source country which regards only the entity as liable to tax.

The general saving clause should not be problematic, but Australian treaty practice has long opposed its use on the basis that a similar principle is already implicit. This confusing doctrine was followed in the 2015 treaty with Germany but not in Australia’s reservations under the MLI. The 2019 treaty with Israel includes a standard OECD saving clause, which suggests that Australia has sensibly given up its objection.

Although the OECD Model still focuses on relieving source-residence double taxation, hybrid entities present a particular risk of dual-residence double taxation. The Australia–New Zealand treaty includes a special provision for dual-residence double tax relief leveraged on the transparent entity clause. This is an approach that could usefully be followed in other treaties.

 

Further Reading

Mark Brabazon, ‘After the Flood: Transparent and Hybrid Entities in Australian Treaties after the MLI’ (2019) 17(1) eJournal of Tax Research 1

Mark Brabazon, ‘The Application of Tax Treaties to Fiscally Transparent Entities’, in Richard Vann (ed), Global Tax Treaty Commentaries IBFD (IBFD online, 2018)

Mark Brabazon, International Taxation of Trust Income (Cambridge University Press, 2019) chapter 8