AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2021 >> [2021] MelbULawRw 17

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Cooper, Graeme S --- "Fixing the Defective Jigsaw" [2021] MelbULawRw 17; (2021) 45(1) Melbourne University Law Review 362

critique and comment


Graeme S Cooper[1]*

This article is about the flaws in the design and execution of our tax laws — problems which are not accidents or mistakes but rather the unfortunate outcomes of misguided policy choices. The article examines various design defects such as regime duplication and overlap and demonstrates just how hard it is to undo the problems they cause, and then considers the current drafting preference for rules that are deliberately indistinct and imprecise and the difficulties this approach creates for the tax community — difficulties which were readily acknowledged but have not been remedied by the ‘solutions’ advanced by the rules’ proponents. It is time for our policymakers to return to a targeted approach to both design and execution.


I wish to start this article with two aphorisms. The first is often ascribed to Mark Twain: ‘The weather ... is a matter about which a great deal is said and very little done’.[1] The second is usually attributed to the less well-known American satirist and curmudgeon HL Mencken: ‘there is always a well-known solution to every human problem — neat, plausible, and wrong’.[2] Taken together, they capture nicely the argument of this paper: faced with perennial complaints about the complexity of Australia’s tax system, our policymakers prevaricated; but when they decided to act, they pursued approaches they believed to be clear and simple; they turned out to be wrong.

So, this article is about reform of tax laws, but not tax reform writ large. I say that because the term ‘tax reform’ is employed in public debates as if it were a single idea with an agreed meaning, but as the late Justice Hill reminded us, ‘tax reform being subjective can have no fixed meaning’.[3] Ideas routinely floated under the label ‘tax reform’ are driven by different visions of the evil to be remedied: our taxes interfere too much in economic decisions (the work–leisure choice, the save–spend choice, and the invest here–invest there decision); our tax system discourages foreign investors; we must address bracket creep; the tax burden is regressive; our tax system is gender-biased; and so on. Different evils imply different (and often contradictory) actions: expand the goods and services tax (‘GST’); decrease the corporate tax rate; impose a wealth tax or a digital services tax; repeal stamp duty; flatten the personal income tax rate scale; tax female wages at rates lower than male wages; and so on.

These visions of tax reform raise principally economic questions, but my focus is on the part of the tax system that lawyers created, and which they alone can fix: the design and execution of our tax laws. By ‘design’ I mean the architecture of the tax law — the assembly of the different components of a statute into a whole.[4] By ‘execution’ I mean the odd drafting styles that have been

used by succeeding generations of drafters since 1995.[5] I will use the income tax for my examples, although the same story could be told about other

Commonwealth taxes.

I choose this topic because I want to revisit topics that have occupied me for almost 25 years. I have written elsewhere, on many occasions and at tedious length, about the problems of Australia’s tax legislation and I am going to revisit that work in this article.[6] Let me apologise to those who have heard this argument before and to those who don’t need convincing, but I don’t think one can say too often or too loudly just how poorly we are served by our legislation. Australia’s income tax assessment Acts must be among the worst statutes one could ever encounter.[7] The tax legislation of any country is invariably complex and difficult — tax statutes everywhere are unique beasts — but our legislation excels in needless obscurity and difficulty.

I should say something about the title of this article. In 1986, Professor Ross Parsons delivered a famous lecture in which he likened the income tax to a supernova,[8] but I prefer a different metaphor: the income tax is more like a jigsaw puzzle, and a defective jigsaw at that — important pieces are missing, there are many duplicates, and the image we are to replicate has been deliberately blurred. My point is that, even if our legislators were minded to adopt a modern reform plan — say, a wealth tax or a destination-based, cash-flow corporate

tax — that reform would likely be beset by exactly the same problems evident in the income tax. The legislation would be oddly written, and over time any coherence to the original design would be lost.

In my experience, while these problems of structure and execution are not unknown in other statutes, they are very pronounced in tax laws. The decisions about design and drafting are not accidents or mistakes — they are deliberate policy choices. And they are not viewed as unfortunate or unavoidable or misguided — they are seen as virtues. So, any change to address the problems I see would require abandoning current fashions and convictions. That might not be wishful thinking — there is evidence that some in government have now

recognised the preoccupation of the last 25 years with a language experiment was misguided:

While useful in addressing a particular aspect of complexity, the overall value of simplifying the drafting of legislation without any change in underlying outcomes is questionable. Simplifying language can only do so much if the underlying policy remains highly complex. In many cases, it will simply make the complexity of the policy more apparent and, in practice, only benefits the very small section of society using the tax legislation itself or related guidance material.[9]

While this epiphany — that focusing on different drafting styles is not especially productive — is welcome, the drafters of the ‘Re:think: Tax Discussion Paper’ have the diagnosis only half right.[10] They are right to doubt the value of the drafting experiments, but attributing the problems of our tax legislation to

this new culprit — complex policy prescriptions — is misguided. None of the

examples below involves complicated policies: they are instances of simple

policies, badly delivered.

II Design

My catalogue of structural flaws in the design of the income tax law has many entries: regime duplication and overlap (typically by legislative accretion); legislating inconsistent regimes; legislating regimes that are internally flawed; failure to remove legislative detritus; legislating by omission; legislative amnesia; and so on.

Another pervasive problem is the unending chase. The simplest way to demonstrate what I have in mind is with an example: one taxpayer borrows a machine from another to use in its business, and the borrower agrees to pay for the privilege. The borrower will want to know whether it can deduct the payment it is making to the lender, but in order to answer that question it now has to navigate a course through a plethora of regimes:

• Is the payment deductible or should it be absorbed as a cost or neither? That will require a journey through the Income Tax Assessment Act 1997 (Cth)

s 8-1, divs 40, 240, 242, 250 (‘ITAA 1997’).

• That answer might then be affected by whether the arrangement amounts to a ‘debt interest’,[11] which will mean inquiring into ITAA 1997 div 974 or

ss 25-85, 26-26.

• There is also a remote possibility that this might be a ‘financial arrangement’ which means looking at ITAA 1997 div 230 to be sure the conditions for the licensing exception are met.[12]

• The tax system also requires the borrower to think about compliance obligations such as no-ABN withholding in the Tax Administration Act 1953 (Cth) sch 1 div 12 (‘Tax Administration Act’), and this brings into play ITAA 1997 s 26-105.

• The borrower will also have to examine how much debt it is currently carrying and consider whether the arrangement will be affected by the thin-capitalisation rules in ITAA 1997 div 820.

• If the owner happens to be a foreigner, a new complication arises: the borrower will need to establish whether the asset is connected to a permanent establishment of the lender here and to look at the Income Tax Assessment Act 1936 (Cth) s 128B (‘ITAA 1936’) to decide whether the lender is liable to interest-withholding tax or royalty-withholding tax, because if either is relevant, that will raise different compliance obligations for the borrower under the Tax Administration Act and bring into play ITAA 1997 s 26-25.

• If the owner is foreign there is the possibility that the transaction might be a hybrid financial arrangement,[13] or the lender might be a hybrid entity or there might be a hybrid arrangement or entity one or more stages removed from these parties, and that would trigger ITAA 1997 div 832.

After that exhausting chase, the answer to that borrower’s question is: you may be able to deduct all of the payment, some of the payment, none of the payment, or something else entirely.

Too many matters in tax do not admit simple answers because of the way our legislation is constructed. In this article, I choose to dwell on regime duplication and overlap. Duplicating an existing tax regime with a newer version is habitual in Australia’s tax laws — there is a clear preference for being deliberately overambitious: ‘leave the existing tax claim in place, add another tax claim as well’. The difficulties of regime duplication can be seen in many places. It is perhaps the most annoying and unnecessary design defect and, once done, it is very hard to solve.

Take, for example, the taxation of financial arrangements regime (‘TOFA’).[14] When it was enacted in 2009,[15] it was decided that TOFA would simply overlay all the existing law for taxpayers who were inside TOFA: the multitude of rules dealing with the assessability (or deductibility) of interest; the treatment of discounts or premiums on the issue of securities; profit (or losses) on the sale of debt securities; writing off bad debts; forgiveness of debts; the treatment of debt denominated in foreign currencies; and so on. All of those rules remained in place and applied, and TOFA applied as well.[16] The duplication was to be handled by an ordering rule and a numerical calculation: in most cases, the TOFA rules would be applied first and the income tax rules applied second.[17] Any amount taken into account in making the TOFA calculations would be excluded in making the income tax calculation.[18] That sounds simple but problems still arise. Take debt forgiveness. The statutory debt-forgiveness rules are applied before TOFA.[19] That creates a circularity problem because the ordinary-income rules are applied before the debt-forgiveness rules, except that the TOFA rules are applied before the ordinary-income rules.[20]

A more complex example is fringe benefits tax (‘FBT’). When the 1985 Reform of the Australian Tax System: Draft White Paper proposed remedying the non-taxation of fringe benefits,[21] the obvious solution was to enact a series of valuation rules in the income tax and then insist the Australian Taxation Office (‘ATO’) get serious about collection and enforcement.[22] Such an approach would impose tax on the right person and at the right rate.

But instead of valuation rules, a new tax was proposed.[23] The tax might have been imposed on employees; instead, it was imposed on employers.[24] Being imposed on employers, the rate could not be differentiated for each employee, so it was imposed at a flat rate.[25] The legislation might have started at its intended target — non-cash benefits provided to employees.[26] Instead, it was drafted to encompass everything that meets the definition of ‘benefit’: a term intended to cover the field of both things that were already taxable — and actually being taxed — and those that weren’t.

This combination of design features — a separate tax, a flat rate, a different taxpayer and a different tax base covering ‘benefits’, not income — effectively duplicated the income tax for things that were already taxed as income: wages, commissions, redundancy and leave payments, employee shares, superannuation, cash allowances, and so on.

Ridding systems of duplication may seem simple — just enact some exceptions to the definition of ‘fringe benefit’[27] — but it required multiple steps in both the income tax and the FBT:

• Some amounts of cash, and some non-cash items, had to be excluded from the meaning of ‘fringe benefit’ so that they could be taxed just under the income tax.[28]

• Some things had to be excluded from income so that they could be taxed just under the FBT.[29]

• Some things had to be excluded either from income even though they were not taxed under the FBT, or from the meaning of ‘fringe benefit’ even though they would not be taxed under the income tax — ie benefits which were meant to be exempt entirely now needed to be immunised from

two regimes.[30]

• Then there were the ‘otherwise deductible’ rules for fringe benefits that were not exempt per se, but were not taxed under the income tax because of deductions.[31] In effect, we had to work out how to notionally pass the employee’s income deduction back to the employer to use to reduce its

FBT liability.

• And there were other instances of the otherwise deductible rules for fringe benefits that were taxed under the income tax, but not as employment


We have these explicit rules to reconcile the duplicated regimes, but other sources of overlap existed and were not so obviously solved, such as difference in timing[32] and valuation,[33] and transition between taxpayers.[34] Rules to handle these problems are not apparent on the face of the legislation.

My third example involves problems with the architecture of the capital gains tax (‘CGT’) provisions. When a tax on capital gains was added in 1986,[35] the regime was meant both to be internally coherent and to fit comfortably with the rest of the income tax. The experience of the last 30 years has revealed cracks, fault lines and holes: the CGT often sits uneasily with the rest of the income tax.

The obvious hole in the tax system which the CGT regime was intended to fill was the omission of a tax on the disposal of capital assets for a profit, and the original legislation used concepts and structures which captured that idea: a capital gain was triggered when an asset acquired on or after 20 September 1985 was disposed of and its cost base, indexed for inflation, was less than the consideration received on the disposal.[36] CGT had the ‘look and feel’ of a tax on the sale of assets at a profit. To put this another way, there was no need to enact legislation which imposed tax on the disposal of trading stock, revenue assets or depreciating assets since existing provisions in the legislation were more than capable of dealing with them. Most of the problems examined below arise from the decision to make the CGT regime duplicate tax claims that

already existed.

A Taxing Depreciating Assets

The decision to include depreciating assets in the CGT regime created an insoluble problem of regime conflict, a problem which survived until 1999:[37] it basically thwarted attempts to introduce a pooling system for recovering the cost of depreciating assets for a decade.

In 1991, the government decided to enact a system of pooling for depreciable items.[38] A pooling system recognises the reality that a company of any size cannot and does not track the life cycle of individual assets in its commercial accounts, and so it needn’t do this for tax purposes. But pooling never got off the ground. The attractiveness of pooling could not be doubted, but it was just not feasible because the CGT system would not permit a taxpayer the degree of flexibility that the pooling system was designed to accomplish. CGT treated each ‘asset’ separately and so a taxpayer had to track individual assets: their current cost base, their proceeds, and any resulting net capital gain identified.[39] In other words, taxpayers could adopt pooling for depreciation, but they would not meet their CGT obligations if they did — they would have to undo pooling and itemise capital gains.

The drafter of the legislation acknowledged that there might be some difficulties in the intersection between pooling and other regimes and attempted valiantly to solve them. ITAA 1936 s 62AAU attempted to manage the interaction with the CGT by reconstructing a notional cost for an item of pooled property if it was disposed of. CGT would then be applied using ‘the reconstruction assumptions’ in ITAA 1936 s 62AAM, which essentially reversed pooling. The taxpayer had to reconstruct a cost in the asset based on the assumptions that the asset had not been allocated to a pool and that it had been depreciated using a straight-line method but at rate applicable to the pool.[40] But in order to do that, the taxpayer would have to identify which asset had been sold, when it was bought, for how much, and so on — the very obligations that pooling had dispensed with. The drafter tried to navigate the interface between the two regimes, but they are, at heart, fundamentally inconsistent. It was not until ITAA 1997 s 118-24 was enacted in 1999[41] that the pooling system stood a chance of being useful — it simply switched off the CGT regime for most depreciating assets,[42] the position that should have been adopted in 1985.

B Taxing Revenue Assets

The decision to apply CGT to revenue assets was similarly unnecessary. Unlike trading stock and (eventually) depreciating assets,[43] there is no exclusion from the CGT regime for these assets; instead, the duplication of both income tax and CGT is meant to be handled by ITAA 1997 s 118-20.[44] At first glance, the rule seems straightforward and effective: priority is given to the income tax, with the CGT operating as a fallback for cases where the income tax produces a lower amount. But there are obvious problems in the drafting.

The first problem is the requirement that the amount be included in assessable or exempt income by a provision elsewhere in the ITAA 1997because of the event’.[45] Most amounts that are assessable as receipts will not involve amounts being included because of a CGT event. For example, in the transaction in ATO Interpretative Decision (‘ID’) 2008/110 (now withdrawn), a solicitor renders an invoice and is paid.[46] The solicitor is assessable under ITAA 1997 s 6-5 because they have supplied services to the client and rendered an invoice. CGT event C2 then happens, being the collection of the receivable.[47] The rule in ITAA 1997 s 118-20 does not solve the problem because the tax under s 6-5 is not triggered by an ‘event’. The drafter, when made aware of the problem, attempted to fix it:

Subsection (1) applies to an amount that, under a provision of this Act (outside of this Part), is included in ... your assessable income or *exempt income ... in relation to a *CGT asset as if it were so included because of the *CGT event referred to in that subsection if the amount would also be taken into account in working out the amount of a *capital gain you make.[48]

Just how this subsection solves the problem is not obvious. It says the ‘because-of-the-event’ test will be satisfied if ‘an amount ... is included in ... your assessable income ... in relation to a *CGT asset’,[49] but it is hard to see how the fees invoiced by the solicitor are included in their assessable income ‘in relation to a *CGT asset’.[50] The ID was withdrawn in 2018 on the basis that it was otiose: ‘Guidance on the issue contained in this ATO ID can be found in [Taxation Determination (‘TD’)] 2’.[51]

That document describes a different transaction and misses the point


In other respects, ITAA 1997 s 118-20 is more effective: it is capable of

dealing with timing discrepancies, and it is able to deal with some discrepancies in the identity of the relevant taxpayer, at least so far as partnerships are


ITAA 1997 s 118-20 also deals specifically with deliberate exemptions in the income tax so that income tax exemptions are not negated by the imposition of CGT. But the decision to have the CGT duplicate existing tax claims means that, where governments intend to deliver an exemption, tax claims have to be switched off twice: an exemption from CGT does not immunise an amount from income tax.

The same is true of rollovers, as the taxpayer in Ransley v Federal Commissioner of Taxation discovered: the rollover of her capital gain under ITAA 1997 sub-div 124-M did not immunise her from an income tax liability on the share exchange.[52]

Not only is it important to switch off both CGT and income tax, it can be important to do it in the right way. CGT can be eliminated by playing with one or more of the elements that trigger the tax, and the legislation is not especially consistent or coherent in deciding which method to use. The ITAA 1997 offers several paths to the same destination: the event is switched off;[53] the event happens but it does not involve an asset;[54] any gain or loss is disregarded;[55] or the gain is reduced.[56] But so far as removing duplication is concerned, these four methods aren’t perfect substitutes. For example, ITAA 1936 s 47 treats a distribution by a liquidator as a dividend to the extent that it represents ‘income’ derived by the company, and ITAA 1936 s 47(1A) defines ‘income’ for this purpose to mean ‘a net capital gain ... (except a capital gain that is disregarded)’. If CGT is switched off by saying that no event happens, or that the amount of the gain is reduced, or that no asset is involved, this method is ineffective to switch off ITAA 1936 s 47, but if CGT is removed by having the gain disregarded, tax under s 47 falls away as well.[57]

While ITAA 1997 s 118-20 can address amounts that are assessable (or exempt) somewhere else in the ITAA 1997, the rule does not address other ways in which tax law might treat an amount, such as applying the amount to reduce the cost of some asset. This has a similar effect to including an amount in income, albeit one step removed, but ITAA 1997 s 118-20 will not address

this outcome.

And there is an assumption in the drafting that the computation for income tax and CGT will have similar values: they will either both be positive or both be negative. The legislation does not contemplate the possibility that there may be an income tax gain and a capital loss, or an income tax loss and a capital gain. This might seem unlikely, but it happens for a variety of reasons, such as where: assets enter the different tax regimes at different times; income and gains are computed on values at different times; or different rules apply about the

immediate deductibility of costs versus absorption of costs into cost base.

The last problem of ITAA 1997 s 118-20 I will mention is the problem of trying to reconcile a gain-based regime with a flow-based regime. CGT is a tax on net gains — individual payments and receipts are ignored; what matters is the sum of the two items. The income tax will sometimes apply to a net figure, but more often it focuses on gross flows which are assessable or deductible. This means there needs to be a way of accommodating the situation where the income tax recognises a gross receipt or payment, whereas CGT recognises the difference between the two.

If that sounds too oblique, consider a case like Cliffs International Inc v Federal Commissioner of Taxation (‘Cliffs International’).[58] The company successfully claimed a deduction for payments it made in connection with a purchase of shares.[59] While the reasoning in the High Court was often opaque and inconsistent, the result was that the taxpayer bought a subsidiary by paying a price, part of which was allowed as a deduction.[60] ITAA 1936 s 160ZH(6) had provided from the outset that an amount which was deductible could not be included as an incidental cost of acquisition, but the assumption in the drafting was obviously that the price paid for a capital asset would always be non-deductible. While that will be true in 999 cases out of 1,000, Cliffs International had been decided and thoroughly absorbed into the tax culture by the time the CGT rules were being drafted. But it wasn’t until 1997 that the legislation was amended to provide a comprehensive rule that an amount could not both be an allowable deduction and form part of cost base.[61]

Interestingly, there is no equivalent rule in regard to capital proceeds: the ATO seems convinced that one amount can be both assessable as ordinary income or statutory income and treated as capital proceeds in respect of some CGT event. For example, TD 2001/27 treats the one amount as both statutory income (a dividend deemed to arise under ITAA 1936 s 47) and capital proceeds (from the cancellation of the shareholder’s shares).[62] The structure of legislation does not obviously reject this position, although logically it seems hard to characterise the one amount as both the fruit of the tree and proceeds of selling the tree itself.[63] The legislation might have excluded amounts which are already assessable from being capital proceeds — in other words, adopted the same solution used for deductible costs.[64] Instead, the assessable amounts

remain assessable as ordinary income or statutory income and remain

capital proceeds.

III Execution

The drafting of legislation is a perennial topic for commentators and practitioners. There is a long list of usual complaints: statutes are prolix; wording is cumbersome; there is too much jargon; usage is often arcane; punctuation is on

holiday; and so on.

In 1993, Parliament was convinced by the argument that the complexity of Australia’s tax system could be traced to problems of language and could be solved by changing the words used in the statute.[65] The Joint Committee of Public Accounts’ report led to a number of textual and presentational changes that we are familiar with.[66] It also began the processes that would lead to principles-based drafting.[67]

To some extent, the plain-language movement of the 1990s had elevated usage issues, at least so far as obscure and arcane terminology was concerned.[68] It is true that the modern drafting style does not use as much jargon: we now use ‘economic benefit’ where once we might have used ‘money’ or ‘property’ or ‘services’.[69] But we still manage to assemble simple words into inscrutable


Tax drafting, however, added a new dimension: in tax, there would be a deliberate policy for rules to be indistinct and imprecise — not just in the choice of words but also in the articulation of operative rules. It is one thing to avoid using terms that are esoteric; it is another thing to want rules that are vague and imprecise. That approach was deliberate, but our experience of the project to date has not been promising. The two examples below demonstrate the difficulties of considered ambiguity: accurately expressing the principle, and then deciding what it does — and what it doesn’t — mean.

A The TOFA Regime

My first example is the TOFA regime because it was conspicuously chosen as the project to trial principles-based drafting.[71]

In the 2005 draft, the basic definition of a ‘financial arrangement’ was proposed: ‘a legal or equitable right to receive something of economic value in the future’.[72] The explanatory material argued that the definition captured the ‘principle’ of a financial arrangement by taking into account ‘the common economic substance underpinning all [financing and risk-shifting] arrangements’.[73]

The general principle which defined the discrete group of ‘financial arrangements’ utilised a single, observable, generic characteristic: someone had the right to get (or an obligation to provide) something of value tomorrow.[74] But the attempt to capture the ‘principle’ of a financial arrangement using this formulation, rather than a more traditional definition, suffered from obvious difficulties. First, there was nothing in the definition which expressly or by implication limited it to transactions involving finance. But more importantly, the only ‘principle’ being captured was simply a deal which is incomplete.

It was not a surprise when the definition of a ‘financial arrangement’ was substantially modified in succeeding versions of the TOFA legislation. In the January 2007 exposure draft, the definition of a ‘financial arrangement’ was significantly narrowed and refined.[75] The definition, which had been just five lines in the 2005 draft,[76] now occupied three pages. The regime would now apply to ‘a ... right to receive a *financial benefit that has a *monetary nature’.[77] Under the new test, money became important.[78] And the draft added an entirely

new qualification:

(6) You do not have a financial arrangement ... if:

(a) under the arrangement you also have a right or obligation to receive or provide something that is not a financial benefit of a monetary nature; and

(b) that right or obligation is not insignificant in comparison with your rights and obligations under the arrangement to provide financial benefits that are of a monetary nature.[79]

This represented a fundamental re-conception of the principle. No longer did the principle refer just to a single, observable, generic characteristic; the principle now looked at the two legs of a transaction — someone had a right to receive money tomorrow and did not have an obligation to provide non-money tomorrow. The principle now required a consideration of multiple characteristics: the presence of money, futurity and the absence of non-money.

But even with the reconceptualisation, the scope of transactions covered was extremely broad. For example, the definition might capture any seller or buyer of goods, any unpaid employee, a legatee under a will, a superannuant, or a tort victim — at some stage in their relations, these people will likely have rights to receive money (or obligations to pay it) and they might not, at the same time, have an offsetting non-monetary obligation. Timing is important to the definition because it was meant to be ambulatory: an arrangement could be a ‘financial arrangement’ one day and not the next.[80] But these rules were never meant to apply to wages, legacies, superannuation, or damages payments; so again the overarching principle had to be circumscribed to ensure a multitude of

transactions could not be ‘financial arrangements’ at any time.

But even with these carve-outs, the new version still has difficulties — evidenced by the now-deferred project to rewrite the TOFA rules[81] — and they stem from a fundamental difficulty which the drafters chose to ignore: that there is nothing coherent about the scope of a regime directed to ‘financial transactions’ or ‘financial arrangements’. The scope of such a regime has to be constructed. That should have been obvious because the TOFA project was pursuing not one target but two: both financing arrangements and risk-shifting arrangements (ie derivatives and hedging). These are not related topics — raising money to operate is different to holding assets or liabilities to shift risk — and it was unrealistic to expect that a single principle could be devised that would encompass both in any meaningful way. Eventually, the principle-based approach could not be sustained: the two other types of financial arrangements which were grafted onto the system — equity interests[82] and foreign currency holdings[83] — are simple matters of form.

B The Consolidation Regime

I will now turn to Australia’s consolidation regime to demonstrate the other problem — just what is this principle meant to imply? Most countries combine the tax positions of related entities either by an attribute transfer mechanism, or by single joint filing with an ‘aggregate and eliminate’ rule.[84] The central provision of Australia’s corporate consolidation regime — the single entity rule (‘SER’) — adopts a different approach: it works by undoing the recognition of group members as separate legal entities.[85] The SER stands in splendid isolation in the consolidation rules, without support or buttressing from other more elaborate deemings. It, and it alone, does most of the work of the entire consolidation edifice for an established group — the vast bulk of legislative detail is about entries and exits.

I think the legislative history is clear that the mechanism chosen by the drafters to achieve consolidation is to undo the fact of incorporation and thus the separate existence of the related entities, so it is more than a little surprising that our judges so far have read this section in quite a different way. In Channel Pastoral Holdings Pty Ltd v Federal Commissioner of Taxation (‘Channel Pastoral’), Pagone J said:

The statutory direction in s 701-1(1) is not that a subsidiary of a consolidated group is to be treated as non-existent, or that it ceases to be a taxpayer or that it does not derive or make assessable income or gains, or does not incur losses or outgoings ... The requirement that the subsidiary is to be taken to be part of a composite whole is an indication of a statutory direction that the parts continue to exist as parts of the composite whole, and that their individual portions of income and gains, or losses and outgoings, will be taken into account in the
calculation of the aggregated composite whole.[86]

This passage was endorsed by Davies J in Glencore Investment Pty Ltd v Federal Commissioner of Taxation.[87] The ATO responded by trying to reassert its view of the SER,[88] so it seems we have a fundamental disagreement about exactly what the principle is and how it works.

But even if we could agree what the principle is, the second question is: just what does it accomplish? Here, we encounter the problems of unbounded

implications leading to ‘principle creep’ and unintended consequences.

Some of the implications of the SER are obvious and seem to be well and easily captured by the principle. It is generally understood that, by virtue of the SER alone, the intra-group provision of services, sales and purchases of inventory, depreciables or capital assets does not give rise to income or gain, cost or loss;[89] no income, deduction or franking consequences attach to intra-group flows of interest, dividends or royalties.[90]

The ATO takes matters somewhat further, and expands the domain where implications of the SER arise to include aspects of tax administration and procedure: lodgement of returns, record-keeping obligations and the imposition of penalties.[91] This seems inoffensive, though they are not strictly matters that go to the calculation of the taxable income or tax loss of the group’s members.

But the implications of the SER have been expanded even further by the ATO to carry implications about characterising transactions for members of corporate groups: the claim is not simply that the transaction is ascribed to the head entity, but that in fact a different transaction occurs because of the SER. One example is the transaction in Federal Commissioner of Taxation v Myer Emporium Ltd (‘Myer Emporium’): the assignment of the right to receive interest on a loan.[92] The issue in dispute was whether the $45 million received on selling future interest coupons was income or capital in the hands of Myer Emporium.[93] The High Court concluded that the $45 million received was assessable as income.[94] That conclusion was expressly based on the view that the transaction in question was a sale.[95]

One can contrast that decision with the ATO’s view in TD 2004/33, which says that no CGT event happens on the assignment of an intra-group loan because ‘[t]he income tax laws treat the conferring of rights under the debt in the non-group entity ... as the borrowing of money or obtaining of credit’.[96] The view is repeated in TD 2004/83, which asserts that the transaction involves the raising of finance and the issue (not sale) of a debt interest for the purposes of Australia’s debt–equity rules,[97] and in TD 2004/84 and TD 2004/85, which provide that an assignment of the interest or principal is governed by the rules on the issue of discounted bonds.[98]

These conclusions apparently follow from an ATO view that, because of the SER, the loan between group members has no tax consequences during consolidation, and so it follows that the SER also means the loan does not actually exist and so cannot be sold.

Delimiting the implications of the SER becomes even more problematic when these TDs are contrasted with a different series of TDs issued by the ATO dealing with the sale of a different kind of intra-group asset. TD 2004/40 says the sale of shares in a member of the corporate group is treated as a sale, not a share issue.[99] And for intangibles, TD 2004/34 says that the assignment of an option over an asset held by another group member is the assignment of that option, not the grant of the option,[100] while TD 2004/35 says that the assignment of a licence to use an asset owned by another group member is the

assignment of that licence, not its creation.[101]

No explanation is offered why the SER means the sale of intra-group debt is to be seen as a borrowing while the sale of a share, an option or a licence

remains a sale; nor is the inconsistency with Myer Emporium addressed.

The second problem noted above is the problem of unintended consequences — vague rules can do too much. If the subsidiary members of a consolidated group no longer exist for tax purposes, it carries the implication that the identities of the parties to a transaction are changed. Consider, for example, a subsidiary member of a consolidated group, which borrows money from a related offshore entity. Obviously, the interest rate to be charged on the

loan — a ‘core purpose’[102] — would ordinarily be set by reference to the assets of the subsidiary rather than the parent or the entire group, unless the loan is guaranteed or the other companies offer their assets as security. But if the subsidiary member is disregarded for tax purposes, then who is the borrower, and what rate should be charged to it? Are we to determine the interest rate charged to the Australian company on the basis that the loan was made to the head entity? Commercially, we know that the loan can only be enforced against the subsidiary, but for tax purposes we are told this subsidiary doesn’t exist: it

cannot be the borrower; the borrower must be somebody else.

The lesson is relatively clear. Principles are not easy to capture and the attempt to find them can be just wrongheaded. And even if they appear to have a confined domain, it is not obvious that their implications will be restricted to just that domain. The SER can be understood just to carry implications in the process of calculating income and deductions — indeed, its defined scope is expressly limited just to calculating income or loss. But the ATO reads it as carrying implications about administration and compliance, even the existence and ownership of assets and the identities of people who are engaged in

transactions. This is a long way from a principle that appears so simple.

IV The ‘So What?’ Question

This article so far has been mired in detail in an attempt to demonstrate two points which may not have needed proving to a tax-literate audience. And since it is easy to select samples and present them as representing the whole, we should not discount the possibility that my examples are simply idiosyncratic, though my view is they are typical and teach a general lesson.

But it is now time to ask the ‘so what?’ question. Even if we assume that these practices are systematic rather than isolated, and that they are deliberate rather than accidental, are they anything more than nuisances to taxpayers and their advisers? There are a number of answers to this question; perhaps none of the answers is conclusive on its own, but taken together they point to problems in our tax system which should be fixed.

With regard to regime duplication, an obvious difficulty which should concern government is that their policies can be thwarted by duplication and imprecision. The pooling story is an obvious example — taxpayers were denied the ability to save compliance costs because of regime duplication.[103] But consider also the need to enact multiple exemptions if amounts that are meant to be exempt will remain exempt — every income exemption needs an entry in ITAA 1997 div 53 and s 118-27. And the same applies to rollovers and deferrals. Government policy — whether it be on pooling, exemptions or rollovers — is undermined if the reconciliation is imperfect.

And as the CGT discussion demonstrated,[104] perfect reconciliation is decidedly hard. It can rarely be done by simple definitional delimitation. As the FBT example showed,[105] it can require aligning all the features of a tax: the definition of the base, the identity of the taxpayer, the timing rules, and the valuation rules. And it can also require rules to govern transitions between the taxes.

Then there is simply economic waste. Duplicating existing tax claims and deliberate imprecision aggravate compliance costs for taxpayers. Fewer questions can be answered without engaging advisers because the obvious regime is rarely the end of the journey. And as the rulings and determinations I have referred to show, these practices can confuse administrators as well. All of this aggravates the deadweight drag of uncertainty — taxpayers can never be sure that their positions will not be challenged until the statute of limitations expires.

But as a lawyer, I regard as more compelling the proposition that this is simply an affront to the rule of law: citizens are entitled to have their obligations clearly expressed in advance, especially when non-compliance carries sanctions, and they are entitled to require our politicians to share that goal, and not to turn a blind eye to practices that foster (and are designed to foster) ambiguity and uncertainty.

The decision to be deliberately imprecise was not taken without some thought to its difficulties, but its adherents proposed many and various reasons why it would be successful. They are worth examining, because if the benefits have not been delivered, we have only the detriments. The main justifications were: comprehensiveness (we cannot foresee all the circumstances which might need to be regulated, but a principle will cover the topic without needing to itemise every conceivable case); adaptability (imprecision means the law will adjust and deal appropriately with things that did not exist at the time the rule was drafted); better legislation (legislation will have to capture a coherent policy position and that means there must be one); legislative efficiency (tax law will be shorter, simpler, more stable and need fewer amendments); certainty (principles will be clearer and admit of fewer areas for dispute); better adjudication (judges will apparently find interpretation easier because the intended target will not be lost behind excessive detail); reduced compliance costs (clearer rules and more stable laws mean less need for tax practitioners); and less scope for avoidance (drafting with precision in effect creates a roadmap to avoidance).[106] Some of the claims may seem sensible, some exaggerated, and others implausible, but more importantly, the experience has not matched the promise: the government has said it will have another attempt at doing TOFA,[107] and

consolidation is currently the site for a disagreement between the courts and the administration.

There are other causes for disquiet. There is always a level of concern about the ‘second line of defence’ offered by proponents of principles-based

drafting — that the missing detail can be supplied by the administrative agencies in rulings or other documents.[108] That is possible, but the lesson of the Australian rulings system is that the administrator, confronted with a system by which it will be held to its utterances, can also simply clam up: in 1999, there were 19 Tax Rulings, while in 2019, there were 6, and 5 Law Companion Rulings; in 1999, there were 84 TDs, while in 2019, there were 14.[109] That is not to say the administrator has been on holiday, but the documents that are now issued instead, such as Law Administration Practice Statements, Practical Compliance Guidelines and ATO IDs, are expressed not to be binding. Non-reliable administrative announcements are not a sufficient answer.

And then there is the great unknown — how will judges react to statutes (in fact, just to tax statutes) drafted using principles? The lesson of Channel Pastoral is that what seems obvious to the drafters may not be quite so obvious when read by others. Sir Anthony Mason indicated the kind of reception that

principles-based legislation might receive:

We need to recall that tax liability is determined by Australian judges and lawyers who are attuned to interpreting specific, detailed provisions rather than provisions formulated in more general, abstract terms. ... [Tax] laws ... will not have a predictable operation if they are couched only in abstract and general terms which leave a very large element of leeway to the courts in applying very general principles to particular fact situations.[110]

And it is all very well to argue that if statutes are drafted around principles, judges will have a better idea what Parliament would have wanted to happen, but it is no small leap of faith to believe that judges will conclude that Parliament actually legislated a provision achieving that objective. Indeed, the argument that ‘this is what Parliament would have wanted’, inevitably carries with it the unspoken implication that they didn’t actually enact it.

As I said at the outset, I chose this topic because I think one can never say too often, or too loudly, just how poorly we are served by our legislation. Our policymakers would do well to take the advice of Kitto J and ‘take the trouble to analyse [their] ideas and define [their] intentions with precision before putting pen to paper’.[111] Tax law is hard enough without legislation which pursues structural redundancy and needless imprecision.

* Professor of Taxation Law, Sydney Law School, The University of Sydney.

[1] See, eg, Fred R Shapiro (ed), The Yale Book of Quotations (Yale University Press, 2006). The original form of the quote is probably more properly attributed to Charles Dudley Warner: see ‘The Author of “My Summer in a Garden”’ (1889) 6(2) Book Buyer 55, 57.

[2] HL Mencken, Prejudices: Second Series (Jonathan Cape, 1921) 158.

[3] Justice Graham Hill, ‘Tax Reform: A Tower of Babel’ (2005) 1(2) Journal of the Australasian Tax Teachers Association 1, 1.

[4] Elsewhere, I have referred to this issue as raising problems with the ‘architecture’ of the tax law: Graeme S Cooper, ‘The Defective Jigsaw’ (2015) 30(4) Australian Tax Forum 783, 785–6. See also Graeme S Cooper, ‘Text, Style, Presentation’ in Graeme S Cooper (ed), Executing an Income Tax (Australian Tax Research Foundation, 2008) 123, 123.

[5] The language dimension to the problem is examined more fully in Graeme S Cooper, ‘Legislating Principles as a Remedy for Tax Complexity’ [2010] (4) British Tax Review 334 (‘Legislating Principles’). See also Tony Frost and Graeme Cooper, ‘Trading One Uncertainty for Another? Ten Years’ Experience with the Debt–Equity Rules’ (2013) 17(1) Tax Specialist 2; Graeme S Cooper, ‘“A Rose Is a Flower Is a Plant”: Tax Simplification South of the Equator’ in Report of Proceedings of the Forty-Seventh Tax Conference (Canadian Tax Foundation, 1996) 3:1 (‘A Rose Is a Flower Is a Plant’).

[6] See Cooper, ‘Legislating Principles’ (n 5); Frost and Cooper (n 5); Cooper, ‘A Rose Is a Flower Is a Plant’ (n 5); Graeme S Cooper, ‘Sources of Complexity in the Income Tax: Culprits Real and Imagined’ in Graeme S Cooper (ed), Executing an Income Tax (Australian Tax Research Foundation, 2008) 1.

[7] Income Tax Assessment Act 1936 (Cth) (‘ITAA 1936’); Income Tax Assessment Act 1997 (Cth) (‘ITAA 1997’).

[8] Ross Parsons, ‘Income Taxation: An Institution in Decay?’ [1986] MonashULawRw 6; (1986) 12(3) Monash University Law Review 77, 77.

[9] Department of the Treasury, ‘Re:think: Tax Discussion Paper’ (Discussion Paper, March 2015) 176 (‘Re:think’).

[10] ‘Re:think’ (n 9).

[11] See ITAA 1997 (n 7) s 974-15.

[12] See ibid s 230-460(2)(e).

[13] See ibid s 832-1.

[14] Ibid div 230.

[15] See Tax Laws Amendment (Taxation of Financial Arrangements) Act 2009 (Cth).

[16] See Explanatory Memorandum, Tax Laws Amendment (Taxation of Financial Arrangements) Bill 2008 (Cth) ch 11 (‘2008 TOFA Explanatory Memorandum’) .

[17] Ibid 343–54 [11.8]–[11.46].

[18] ITAA 1997 (n 7) s 230-20.

[19] Ibid s 230-470.

[20] Ibid ss 230-20(4), 245-85(1).

[21] See Commonwealth, Reform of the Australian Tax System: Draft White Paper (Australian

Government Publishing Service, 1985) 88 [8.6] (‘Draft White Paper’).

[22] See ibid 88 [8.8].

[23] Ibid 90 [8.20].

[24] Fringe Benefits Tax Assessment Act 1986 (Cth) s 66(1) (‘FBTAA’).

[25] Fringe Benefits Tax Act 1986 (Cth) s 6.

[26] See Draft White Paper (n 21) 87–8 [8.3]–[8.5].

[27] See FBTAA (n 24) s 136(1) (definition of ‘fringe benefit’).

[28] Ibid.

[29] ITAA 1936 (n 7) s 23L(1).

[30] See, eg, FBTAA (n 24) s 136(1) (definition of ‘fringe benefit’ para (f)).

[31] See, eg, ibid ss 19, 24, 37, 44, 52.

[32] The statute has to be able to address the situation where each tax recognises the taxable event in different years: eg the fringe benefit arises in year one but the income arises in year two, or the FBT sees a one-time event while the income tax sees an ongoing series of transactions.

[33] The statute should provide what happens if the income and fringe benefit amounts differ: eg the value for FBT is $100 but the amount of income is $150 or vice versa. Do the allocation rules allocate and immunise items or amounts?

[34] The statute has to deal with the transition from the employer to the employee if, say, providing a piece of property is a fringe benefit taxed to the employer, but the sale of that property

generates a gain taxed to the employee. What is the employee’s cost in an asset bought with money already taxed, albeit to another person, but probably passed on to the employee in some


[35] See Income Tax Assessment Amendment (Capital Gains) Act 1986 (Cth) s 19 (‘Capital Gains Amendment Act’), inserting ITAA 1936 (n 7) s 160L.

[36] See Capital Gains Amendment Act (n 35) s 19, inserting ITAA 1936 (n 7) ss 160L, 160Z(1)(a).

[37] See New Business Tax System (Capital Allowances) Act 1999 (Cth) sch 1 item 14, inserting ITAA 1997 (n 7) s 118-24.

[38] Department of the Prime Minister and Cabinet, Building a Competitive Australia (Australian Government Publishing Service, 1991) 5.25. See also Taxation Laws Amendment Act 1992 (Cth) (‘TLAA’).

[39] See ITAA 1997 (n 7) s 100-45.

[40] See ITAA 1936 (n 7) s 62AAM, as inserted by TLAA (n 38) s 24.

[41] See New Business Tax System (Capital Allowances) Act 1999 (Cth) sch 1 item 14.

[42] ITAA 1997 (n 7) s 118-24.

[43] Ibid ss 118-24–118-25.

[44] Ibid s 118-20(1) reads: ‘[a] *capital gain you make from a *CGT event is reduced if, because of the event, a provision of this Act (outside of this Part) includes an amount (for any income year) in ... your assessable income or *exempt income’, and sub-s (4) extends this to an amount that is treated as non-assessable non-exempt (‘NANE’) income.

[45] Ibid s 118-20(1) (emphasis added).

[46] See Australian Taxation Office, Capital Gains Tax: Debt Arising from the Provision of Services (ATO ID 2008/110, 1 August 2008) (‘CGT ID 2008’). This ID was withdrawn on 23 February 2018.

[47] Ibid. See also ITAA 1997 (n 7) s 104-25.

[48] ITAA 1997 (n 7) s 118-20(1A) (emphasis added).

[49] Ibid (emphasis added).

[50] Ibid.

[51] CGT ID 2008 (n 46). See also Australian Taxation Office, Capital Gains: What Are the CGT Consequences for the Lender (Creditor) When a Debt Is Waived? (TD 2, 10 September 1991).

[52] (2018) 109 ATR 27, 31 [4], [9], 89 [249] (Jagot J).

[53] See, eg, ITAA 1997 (n 7) ss 240-85, 242-85.

[54] See, eg, Income Tax (Transitional Provisions) Act 1997 (Cth) s 108-5.

[55] See, eg, ITAA 1997 (n 7) s 118-5.

[56] See, eg, ibid s 118-20.

[57] Despite this, the ATO decided to ameliorate the strictness of the language in ITAA 1936 (n 7) s 47(1A) in Australian Taxation Office, Income Tax: Capital Gains (TD 2001/14, 13 June 2001), though strictly speaking the TD only applies to the non-taxable portion of a capital gain

attributable to goodwill.

[58] [1979] HCA 8; (1979) 142 CLR 140.

[59] Ibid 143, 151 (Barwick CJ), 175 (Jacobs J), 176 (Murphy J).

[60] Ibid 144–6, 151 (Barwick CJ), 175 (Jacobs J), 176 (Murphy J).

[61] See ITAA 1997 (n 7) s 110-45.

[62] Australian Taxation Office, Income Tax: Capital Gains (TD 2001/27, 7 November 2001) 2


[63] See Eisner v Macomber, [1919] USSC 119; 252 US 189, 206 (Pitney J for the Court) (1920).

[64] See ITAA 1997 (n 7) s 110-45.

[65] Joint Committee of Public Accounts, Parliament of Australia, An Assessment of Tax: A Report on an Inquiry into the Australian Taxation Office (Report No 326, 1993) 78–9 [5.9]–[5.12],

81–4 [5.24]–[5.38] (‘Report No 326’).

[66] The project also involved adding diagrams, lists, tables, typeface conventions, asterisks, signposts, annotations, examples, notes and other quasi-descriptive material to the statute. It also led to guides explaining the operation of divisions and subdivisions: see, eg, Australian Taxation Office, Debt and Equity Tests: Guide to the Debt and Equity Tests (Guide, 16 April 2018). The use of objects clauses was not common but it was not unknown in Australian tax

legislation at that time.

[67] See below n 71.

[68] See, eg, Report No 326 (n 65) 78 [5.9]–[5.11].

[69] See, eg, ITAA 1997 (n 7) sub-divs 727-B, 727-D.

[70] I have referred to ITAA 1997 (n 7) s 118-20(1A), and there are many more. One of my favourite examples is s 165-200(1), which provides that s 165-165 does ‘not affect how *shares, and rights carried by *shares, are counted for the purposes of determining ... the total voting power in the company; or ... the total *dividends that the company may pay; or ... the total distributions of capital of the company’. That seems to say the ‘same share rule’ in ITAA 1997 (n 7) s 165-165 does not operate for the purpose of counting dividends, capital returns or shareholding. But counting shares is the only thing the same share rule does: it was enacted to prescribe what to count and how to count for the purpose of determining voting power etc. Did ITAA 1997

(n 7) s 165-200 really just switch off the same share rule? If not, just what did it do?

[71] The draft Bill released for consultation in December 2005 drew attention to the claim that it had been drafted using the coherent principles approach: see generally Exposure Draft, Tax Laws Amendment (Taxation of Financial Arrangements) Bill 2006 (Cth) (‘2005 TOFA Exposure Draft’). Chapter 1 of the accompanying explanatory material was dedicated to describing principles-based drafting and explaining its anticipated advantages: see Explanatory Material, Tax Laws Amendment (Taxation of Financial Arrangements) Bill 2006 (Cth) ch 1 (‘2005 TOFA Exposure Draft Explanatory Material’).

[72] 2005 TOFA Exposure Draft (n 71) sch 1 item 1, inserting ITAA 1997 (n 7) s 230-30(1)(a).

[73] 2005 TOFA Exposure Draft Explanatory Material (n 71) 25 [3.5]. The explanatory material also states at 27 [3.12] that

the definition of ‘financial arrangement’ ... is cast in terms of what fundamental and common elements, in principle, characterise both the provision of finance and the shifting or allocation of risk. In this regard, key common elements of all financial arrangements are ... futurity, that is, entry into an arrangement now with performance in the future ... [and the] right of a party to the arrangement to receive, or obligation of a party to provide, something of economic value in the future, irrespective of whether the value or existence of the right or obligation is contingent on some event or other thing.

[74] 2005 TOFA Exposure Draft (n 71) sch 1 item 1, inserting ITAA 1997 (n 7) s 230-30(1).

[75] See Exposure Draft, Tax Laws Amendment (Taxation of Financial Arrangements) Bill 2007 (Cth) sch 1 item 1 (‘2007 TOFA Exposure Draft’), inserting ITAA 1997 (n 7) ss 230-35–230-50.

[76] See 2005 TOFA Exposure Draft (n 71) sch 1 item 1, inserting ITAA 1997 (n 7) s 230-30(1).

[77] 2007 TOFA Exposure Draft (n 75) sch 1 item 1, inserting ITAA 1997 (n 7) s 230-40(2).

[78] See, eg, this statement:

In a commercial context, arrangements commonly identified as ‘financial instruments’, ‘financial transactions’, ‘financial assets’ and ‘financial liabilities’ include ... debt instruments such as bonds, loans, bills of exchange and promissory notes ... [and] derivatives such as options, forwards and swaps ... A factor that is common to all of the above — and to equivalent arrangements — is that a party to the arrangement has either a right to receive, or an obligation to provide, cash or something equivalent to cash ...

Explanatory Material, Tax Laws Amendment (Taxation of Financial Arrangements) Bill 2006 (Cth) 40–1 [3.27]–[3.28].

[79] 2007 TOFA Exposure Draft (n 75) sch 1 item 1, inserting ITAA 1997 (n 7) s 230-40(6).

[80] See, eg, 2005 TOFA Exposure Draft (n 71) sch 1 item 1, inserting ITAA 1997 (n 7) s 230-30(1).

[81] The decision to revisit the TOFA rules was announced in the 2016–17 Budget and then deferred indefinitely in the 2018–19 Budget. See Commonwealth, Budget Measures: Budget Paper No 2: 2016–17 (Budget Paper, 3 May 2016) 37 (‘2016–17 Budget’); Commonwealth, Budget Measures: Budget Paper No 2: 2018–19 (Budget Paper, 8 May 2018) 26.

[82] ITAA 1997 (n 7) s 230-50.

[83] Ibid s 230-530.

[84] New Zealand and the United States of America, for example, use a joint filing system, while the United Kingdom uses an attribute transfer mechanism: see Income Tax Act 2007 (NZ)

s FM 3; IRC § 1501 (2018); Corporation Tax Act 2010 (UK) s 137.

[85] ITAA 1997 (n 7) s 701-1(1) provides:

If an entity is a *subsidiary member of a *consolidated group for any period, it and any other subsidiary member of the group are taken for the purposes covered by subsections (2) and (3) to be parts of the *head company of the group, rather than separate entities, during that period.

The ‘purposes’ in sub-ss (2)–(3) are working out the amount of the liability for income tax (or the size of a tax loss) of the head entity of the group and each subsidiary member.

[86] [2015] FCAFC 57; (2015) 232 FCR 162, 191–2 [119].

[87] (2019) 272 FCR 30, 128 [400].

[88] See Australian Taxation Office, Income Tax: Consolidation (TR 2004/11, 22 September 2004) (‘TR 2004/11’), which provides that ‘the interpretation and consequences of the SER as described [in] the judgment of Pagone J are not consistent with the Commissioner’s view of the SER’: at app 1 [42K].

[89] Ibid 2 [8].

[90] Ibid 7 [32], quoting Explanatory Memorandum, New Business Tax System (Consolidation) Bill (No 1) 2002 (Cth) 23 [2.18].

[91] TR 2004/11 (n 88) 2 [5], 5 [20], 5–6 [24].

[92] [1987] HCA 18; (1987) 163 CLR 199, 205–6 (Mason ACJ, Wilson, Brennan, Deane and Dawson JJ).

[93] Ibid 206.

[94] Ibid 220.

[95] Ibid 219–20.

[96] Australian Taxation Office, Income Tax: Consolidation (TD 2004/33, 22 September 2004) 1 [4].

[97] Australian Taxation Office, Income Tax: Can the Assignment of an Intra-Group Debt or Income Stream to an Entity That Is Not a Member of the Consolidated Group Give Rise to a Debt Interest for the Head Company of the Group under Division 974 of the Income Tax Assessment Act 1997? (TD 2004/83, 15 December 2004) 1 [1].

[98] Australian Taxation Office, Income Tax: Can Division 16E of Part III of the Income Tax Assessment Act 1936 Apply to a Head Company of a Consolidated Group where the Principal of an Intra-Group Loan Is Assigned by a Member of the Group to a Non-Member? (TD 2004/84, 15 December 2004) 1 [1]; Australian Taxation Office, Income Tax: Can Division 16E of Part III of the Income Tax Assessment Act 1936 Apply to a Head Company of a Consolidated Group where an Intra-Group Income Stream Is Assigned by a Member of the Group to a Non-Member (TD 2004/85, 15 December 2004) 1 [1].

[99] Australian Taxation Office, Income Tax: Consolidation (TD 2004/40, 6 October 2004) 1 [1].

[100] Australian Taxation Office, Income Tax: Consolidation (TD 2004/34, 22 September 2004) 1 [1].

[101] Australian Taxation Office, Income Tax: Consolidation and Capital Gains Tax (TD 2004/35, 22 September 2004) 1 [1]–[2].

[102] See ITAA 1997 (n 7) s 701-1(1)–(3).

[103] See above Part II(A).

[104] See above Part II.

[105] See above Part II.

[106] See, eg, 2008 TOFA Explanatory Memorandum (n 16) ch 1; Greg Pinder, ‘The Coherent Principles Approach to Tax Law Design’ [2005] (Autumn) Economic Roundup 75.

[107] 2016–17 Budget (n 81) 37.

[108] See, eg, Pinder (n 106) 84.

[109] See ‘Legal Database: Public Rulings’, Australian Taxation Office (Web Page) <>, archived at <>.

[110] Sir Anthony Mason, ‘Opening Address’ (Speech, ATAX International Conference on Tax

Administration, 20 April 2006) 3 [11]–[12].

[111] Federal Commissioner of Taxation v Newton [1957] HCA 99; (1957) 96 CLR 577, 596. I thank Mark Macrae for pointing me to the correct citation.

AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback