Australia’s Anti-Hybrid Rules

For decades, the Bloomberg Tax International Forum has convened international tax experts from all over the globe. In this new series, Forum members will regularly share their views on current and important topics.

Getting Here

It is now more than four years since the release of the Organisation for Economic Co-operation and Development (OECD) base erosion and profit shifting (BEPS) Action 2 report on “Neutralising the Effects of Hybrid Mismatch Arrangements,” in October 2015, and more than two years since the subsequent report on “Neutralising the Effects of Branch Mismatch Arrangements” in June 2017.

The U.K. was the first and fastest mover. Based largely on the OECD report, and anticipating the OECD work on branches, the U.K. introduced new rules, operative from January 1, 2017. The rules contain 63 provisions in 14 chapters, eight of which deal with a specific hybrid, and one which introduces a specific integrity rule. The U.K. set the foundations for the Australian legislative approach, which was formulated after significant consultation.

Indeed, the day after the OECD released its final report on BEPS Action 2, the Australian Treasurer asked the Board of Taxation to consider Australia’s response. The Board of Taxation recommended Australia adopt the OECD proposals with some minor modifications, which the government accepted in the May 2016 Budget. Usefully, Treasury issued two Exposure Drafts, in November 2017 and March 2018. The final legislation received Royal Assent on August 24, 2018, to apply from years of income starting on or after January 1, 2019 (with some minor exceptions).

The new legislation, contained in Division 232 of the Income Tax Assessment Act 1997, introduced 89 sections with 30 new defined terms, 90 pages of legislation and a 125-page Explanatory Memorandum. Since that time, we have seen two Law Companion Rulings, two Practical Compliance Guidelines and further, albeit minor, changes announced in April 2019 Budget. The Australian rules deal with six specific categories: hybrid financial arrangements; hybrid payers; reverse hybrids; branch hybrids; deducting hybrids; and imported mismatches. They also provide for an additional targeted integrity rule.

New Zealand adopted a parallel process with the introduction of rules from July 1, 2018 (again with minor exceptions). The New Zealand rules contain 15 sections and, like Australia, six specific categories of anti-hybrid rules, although they are not identical. The U.K., Australian and New Zealand models form a comparative framework for the introduction of 27 new regimes in Europe, which is mandatory under EU Council Directive 2017/952 (ATAD II). These models are required to be published by December 31, 2019 and will generally be operative from January 1, 2020.

This paper will look at the design and structure of the Australian rules, which have embraced a high level of complexity in the drafting. This raises the question of whether the rules could have been drafted more simply.

Is Simplicity Feasible?

At one level the anti-hybrid rules are simple: if there is a difference in international tax treatment of instruments or entities that leads to effective double deductions or deductions without income inclusion, and there is sufficient nexus between the entities involved, then the benefit arising will be neutralized.

This simplicity raises the question whether the anti-hybrid project could have been reduced to some very straightforward tax rules, with possibly five to ten uncomplicated principle-based provisions. I suspect, but only with the wisdom of hindsight, that this may well have been the case.

The project involved collecting a very large number of hybrid examples from across the world and grouping them into half a dozen categories and framing sets of rules around each of those categories, albeit with many principles in common. The process was reductive. Oddly, the perceived reductive limit seemed to be governed by the need to explain each category of mischief through the selected examples. Thus, this is complex example-led legislation from the bottom up, and not simple top-down, principle-based legislation.

Hybridity: Beyond the Level of Income and Deductions

There are three elements of the anti-hybrid rules. First there is the hybridity itself. This is derived from the terms of the instrument or the differential tax treatment of an entity. It is not a concept related to a tax rate, although that perceived shortcoming led to the Australian targeted integrity rule.

The essential feature of hybridity is the notion that particular arrangements that fall within the specific six categories mentioned above produce double deductions (D/D) or deduction and non-inclusion of income (D/NI).

However, the determination of hybridity, in the Australian rules, does not stop at the income or deduction level. Thus in ascertaining whether a payment is included in income, that is, whether it is “subject to foreign income tax” (Section 832-130), one needs to ask whether there is a credit, rebate or other concession including dividend received deductions that effectively undoes the income inclusion.

An essential feature of anti-hybrid rules is that a hybrid benefit can be negated, in part or in whole, by income that is included in two jurisdictions or “dual inclusion income.” However, determining whether one has dual inclusion income (Section 832-680) also moves beyond the income level and can become quite complex. Thus, an Australian tax credit for foreign tax paid can undo a dual inclusion income exclusion, but a foreign tax credit for Australian tax paid does not. This surprising lack of symmetry was put in place for compliance simplicity, mostly for U.S. groups.

Differences of treatment in foreign exchange are excluded from the concept of hybridity under the Australian rules. However, the benefit of a franking credit can give rise to hybridity, such that certain deductible (New Zealand branch)/frankable (Australia) distributions are caught by the provisions.

Nexus

There are three notions here: a 50% control group; a “structured arrangement”; and a 25% common ownership relationship, that are applied differentially to the six types of hybrid arrangements. They are not mutually exclusive. All three are applicable to a hybrid financial arrangement. Only the notions of a control group and structured arrangement are applicable to the remaining hybrids, except for a reverse hybrid, which unusually does not have a nexus rule at all if Australia is providing the primary response (see below), but does require a control group or structured arrangement if Australia is undertaking a secondary response.

The meaning of “structured arrangement” has received considerable attention in Australia. This is in part because there is a 12-month deferral of the start date for the imported mismatch rule unless there is a structured arrangement. An arrangement can be structured even though it involves a control group. Essentially a payment will be made under a structured arrangement if the benefit of the mismatch is priced into the terms of the scheme or if it is reasonable to conclude that a mismatch is a design feature of the scheme (Section 832-210).

There is considerable discussion on what this means in the Australian Taxation Office’s (ATO’s) Practical Compliance Guideline 2019/6, but essentially there will be a structured arrangement if a benefit was priced into the terms of the arrangement or a taxpayer had actual knowledge or it was reasonable to conclude that they should have had knowledge of the hybridity. Practically, this means that the deferral for the start date for the imported mismatch rule will rarely apply.

Neutralization

Where there is a hybrid mismatch, the rules seek to neutralize the benefit. To achieve that, a division is drawn between a primary and a secondary response. The secondary response only applies if the foreign jurisdiction does not have corresponding anti-hybrid rules or rules that have substantially the same effect. Thus, if a hybrid financial instrument results in a deduction in Australia and no income inclusion in the foreign jurisdiction, given that the primary rule is the deduction denial, there is no need to consider a secondary rule. However, if there is a deducting hybrid where deductions are undertaken for the same interest expense and the foreign country is the primary response, but does not have anti-hybrid rules, then Australia would provide a secondary response and deny the deduction.

One question that has arisen is whether the U.S. Dual Consolidation Loss rules are considered to be corresponding rules for the purpose of determining whether a secondary response is required under the Australian rules. Initial indications are that they are, but we await formal guidance from the ATO on this point.

Another element of the neutralization rules is that the provisions apply to timing differences for hybrid financial instruments. Unlike the U.K. and the BEPS Final Report, Australia has adopted a three-year rule, such that if there is a deferral of less than three years then the rules do not apply. However, if there is a deferral of more than three years, then the deduction is deferred until the period of the income inclusion.

These three general principles apply to six categories of hybrid, each of which forms a sub-division of Division 832.

Categories of Hybrids

Hybrid Financial Instruments

This D/NI category applies where there is debt, equity, derivative, repurchase agreement, securities lending arrangement or similar that gives rise to a deduction, but no income inclusion and there is not a deferral for three years or less. This rule applies not only to parties within a control group or a structured arrangement, but also to 25% related parties. This category will tend to have limited application for outbound investment, because Australia also has a rule which turns off the participation exemption (Division 768-A) for non-portfolio dividends that are deductible in the paying jurisdiction.

Hybrid Payers

This D/NI category applies commonly for U.S. check-the-box entities where the Australian entity is disregarded or transparent for U.S. purposes, but opaque for Australian purposes. The primary response is to deny the deduction. The dual inclusion income rules become very important in this scenario.

Reverse Hybrids

This D/NI category generally involves an entity making a payment in Australia to a reverse hybrid in Country B, with an investor in Country C. The mischief arises because Country B treats the reverse hybrid as transparent (and thus not taxable in B) and Country C treats it as opaque (and not taxable in C), which is the “reverse” of the hybrid payer mismatch. The primary response is to deny the deduction in Australia.

Branch Hybrids

Australia has an exemption for branches (Section 23AH) which is effectively switched off where there is a D/NI arrangement or where there is a mismatch in attribution between the residence country and the branch country. Many fear that there will be inadvertent circumstances that will give rise to the application of these rules.

Deducting Hybrids

This is a D/D category often involving a transparency—opaque mismatch where payments made to a third party are deductible in two jurisdictions. As noted above, if Australia is the primary response, no nexus rule is required; although it is if Australia is the secondary response.

Imported Mismatches

This is a complex set of rules, whereby a deduction could be denied in Australia for an up-chain un-negated hybrid of any of the forms mentioned above, following direct or indirect payment flows. This is Australia acting as a global police officer. Part of the complexity of these provisions is that they operate on a complicated formulaic apportionment basis, rather than one based on a “reasonable” amount allocated to Australia.

Integrity Rule

Treasury feared there would be a subversion of the hybrid rules by refinancing through a low or tax free jurisdiction. The standard case would be where a parent company equity finances a company in a low tax jurisdiction which debt finances an Australian entity. This does not give rise to a hybrid mismatch as such, the perceived mischief is simply the low tax rate.

The integrity rule to deal with this applies to interest, or similar amounts and derivative payments from Australia where the receipt is subject to foreign tax at 10% or less (ignoring foreign tax credits) and had the payment been made to the ultimate parent it would have been subject to a higher rate of tax, and it was reasonable to conclude that the principal purpose was to obtain a deduction and enable foreign tax to be imposed at 10% or less.

Restructuring

The introduction of the anti-hybrid rules has given many cause to consider restructure out of the rules. In doing so, consideration needs to be given to the potential operation of Australia’s general anti-avoidance rule in Part IVA of the Income Tax Assessment Act 1936. The Commissioner of Taxation has published guidance on this in PCG 2018/7. Essentially, the Commissioner will consider the restructuring to be “low risk” provided the restructure results in an ordinary commercial structure without contrived features. This is a sensible position as the whole raison d’être of the anti-hybrid rules is drive business to adopt ordinary commercial structures.

Precedential Value

In short, Australia’s approach to anti-hybrid rules has been to produce a set of very complex example-led provisions, largely following the U.K. and the OECD. The main risk for global taxpayers is that an up-chain un-negated hybrid will give rise to a denial of a standard interest deduction in Australia because of the imported mismatch rule. The second risk is that inadvertent mismatches will arise for branches which will be neutralized by the rules.

This is a difficult territory: here be dragons.

Grant Wardell-Johnson is Head of the Australian Tax Centre at KPMG.

This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.