Journal of Australian Taxation
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Journal of Australian Taxation |
|
TIME IS OF THE ESSENCE: SUPPLIES, GROUPING SCHEMES AND CANCELLED TRANSACTIONS
By Rebecca Millar[*]
Time of supply is fundamental to the operation of the Australian GST legislation. GST is a tax on transactions, and the tax status of transactions can only be determined by assessing the elements of each transaction at a particular point in time: the time of supply. The character of the supply, the status of the supplier, the location and status of the recipient, and the location of goods are just a few of the elements that may need to be assessed at the time of supply. It is therefore critical to be able to identify the time at which a supply takes place, yet the legislation does not specify when that time is. It has been said that Australia has no “time of supply” rules, but this view stems from a misunderstanding of the role of time of supply rules in other jurisdictions. A comparative approach reveals that such rules are simply “attribution” rules, and Australia does have attribution rules. Moreover, as in Australia, other jurisdictions frequently determine the status of a supply by reference to the “real” time of supply in the “real world.” Two areas where the interaction between “attribution” time of supply rules and the “real” time of supply is significant are the treatment of transactions that span the entry into or exit from a GST group, and the treatment of cancelled transactions. Recent United Kingdom case law on VAT avoidance schemes provides a good illustration of the former, while recent New Zealand case law on land transactions deals with the latter. These cases reveal that Australia is not alone in its lack of legislative guidance on the real time of supply and the article concludes with some suggestions for “real time of supply” rules for Australian GST.
To every thing there is a season, and a time to every purpose.[1]
The aim of this article is to illustrate the fundamental importance of time in Australia’s Goods and Services Tax (“GST”).[2] Taking a comparative approach, the article examines the “time of supply” rules in other jurisdictions and concludes that the phrase “time of supply” has a distinct legal meaning, which differs from its ordinary meaning. The essential role of the rules commonly termed “time of supply” rules is to attribute tax liabilities and input tax credit entitlements to particular tax periods. In contrast, if it is dealt with at all, the real time of supply (the time at which a supply actually takes place in the real world) is governed by “taxable event” rules.
One aim of the article is to dispel the idea that Australian GST lacks “time of supply” rules. The alleged absence of time of supply rules has been described as one of the fundamental differences between Australian GST and other Value Added Tax (“VAT”) regimes.[3] In illustrating the reason why the opposite is true, the article focuses on two main issues: the significance of timing in the GST grouping regime and the treatment of cancelled transactions.
Australia does indeed have “time of supply” rules, though it labels them “attribution” rules, in reflection of their true purpose. What Australia lacks are clear taxable event rules. Yet while other jurisdictions do, to a greater or lesser extent, have taxable event rules, those rules rarely assist in identifying the actual time of supply across the full range of possible types of supply. As a result, other jurisdictions face many of the same issues that confront Australia in identifying whether, and when, a supply can be said to have taken place.
As a preliminary, two fundamental characteristics of GST are considered: the transactional nature of the tax and the way in which it characterises supplies. The article then demonstrates the role of time in the GST grouping regime, using the decisions in three United Kingdom (“UK”) VAT avoidance cases as illustrations. These three cases involve group exit and entry schemes, and are of interest because of the way the taxpayers attempted to take advantage of the interactions between the UK rules on time of supply, VAT grouping, and input tax credit entitlements in order to circumvent a denial of input tax credits on acquisitions by one member of the group. Unfortunately for the taxpayers, the schemes were spectacularly unsuccessful, and in some cases the taxpayers were worse off than they would have been if the schemes had not been entered into. The article considers whether such schemes would fare any differently under Australian GST.
The UK cases raise issues about the interaction between time of supply and the attribution rules, and call into question what it means for a supply to be cancelled when its “time of supply” has passed. On this issue, reference is made to recent New Zealand case law on cancelled transactions, which illustrates and reinforces what can be identified in the cases: a tendency in other regimes to interpret “time of supply” rules merely as “attribution” rules. The article concludes with a brief consideration of where the real differences between Australian GST and other regimes lie and some suggestions for change.
Throughout the article, the phrase “time of supply” is used interchangeably in both its legal and its ordinary meaning. In most cases, the sense in which it is used will be evident from the context. Where clarification is required, the phrases “attribution time of supply”, “real (or actual) time of supply”, and “taxable event” are used. The conclusions are briefly summarised below.
• Time of supply plays a critical role in Australia’s GST.
• The status of a supply as taxable is determined at the time when the supply is made, and in the absence of “taxable event” time of supply rules, it must be concluded that the relevant time is the actual time of supply in what UK Courts refer to as “the real world”.
• Few supplies take place at a single instant: most transactions span a period of time during which the supplier either passes something to the recipient, or creates something that the recipient acquires. Thus, a range of possible times of supply can be identified for any particular transaction.
• Exactly when a supply can be said to have been made is a question of fact, which depends on the character of the supply. Where more than one event could be viewed as the appropriate time, the identification of the time of supply should accurately reflect the true character of the transaction.
• The approach that fits most coherently within the GST regime is that the time of a supply is when, to use the less than elegant style of the Act, “the thing is done”,[4] or in more familiar legal parlance, at the time of performance of the contract for the supply.
• The attribution rules operate without reference to the actual time of supply when allocating GST payable (and input tax credit entitlements) to particular tax periods.
• If there is no supply, there can be no attribution; nonetheless, attribution can, and often does, occur before liability is certain.
• The special rules for deposits in Div 99 of the Act do not enable attribution to be universally deferred until the time of supply; nor is it likely that the principle enunciated in Arthur Murray (NSW) Pty Ltd v FC of T[5] has any application for GST – a receipt is sufficient to trigger attribution and it is not necessary that the amount received should also be treated as income derived for the purposes of income tax.
1.1.3 On The More Specific Grouping Issue
• Under the grouping regime, supplies are only treated as if they are not taxable supplies if they are in fact “intra-group” supplies.
• Whether a transaction involves an intra-group supply is a question of fact, which can only be determined by considering the true character of the supply. In most cases, this will require performance of the contract (the doing of the thing) to take place while the entities are grouped.
• It is not possible to circumvent the intended operation of Australian GST by manipulating the time of contract, invoice, or payment to take advantage of the favourable treatment allowed for intra-group supplies.
• Nonetheless, a group member’s liability to pay GST is transferred to the representative entity under the GST grouping regime for attribution purposes only, even if the relevant supply takes place while the entity is not a member of the group.
In all cases, similar conclusions can be reached about the treatment of acquisitions, but to simplify discussion the focus is on the treatment of supplies.
It is a truism that GST is a “consumption tax”.[6] It is also a value added tax, applied at each stage of the commercial chain in proportion to the value added by suppliers in that chain, and its burden is intended to be borne by end consumers. The value added by suppliers is identified from the payments made and received by those suppliers for the “acquisitions” and “supplies” they make in the conduct of their commercial activities. Thus, it is a tax on consumption, but not necessarily on consumption as economists know it. Sijbren Cnossen has described the difference between the economists’ view of consumption and the VAT concept as follows:
There is wide agreement on the nature of the VAT collection mechanism. VAT is simply a retail sales tax that is collected piecemeal throughout the entire production-distribution process. The consensus may be smaller, however, on some of the theoretical underpinnings of the tax. Is the VAT a tax on consumption activities or, more narrowly, on consumption expenditures? … the answers depend on whether one adopts a theoretical, economic point of view or whether one takes a more practical, legal approach to the VAT.[7]
He goes on to conclude that “the tax is on consumption expenditures rather than on consumption activities”[8] and also points out that the whole amount of the tax is imposed on the “taxable event” (the supply) even when the thing supplied is a durable good with lasting value. VAT is not charged over time in proportion to the actual consumption of the thing supplied; it is charged up front at the time when consumption expenditure is incurred to acquire the thing supplied. Thus:
Ultimately, the VAT is a tax on consumption expenditures as they are incurred. Although the taxation of transactions may not be a goal in itself (as it would be under, say, a stamp duty), it is sufficiently central to the nature of the VAT that it cannot be ignored.[9]
This “practical legal approach” to GST accords with the author’s own views. In the context of a tax on transactions, aimed at taxing consumption expenditures, the concepts of “supply”, “acquisition”, and “consideration” are fundamental. The making of a supply and the acquisition of the thing supplied establish the “taxable event” – the transaction in which consumption expenditure has been incurred. The consideration (a concept far removed from the contractual concept) is the consumption expenditure, the existence of which crystallises both the taxable status of the supply and the amount of tax payable, as well as having a role in determining the time at which tax should be paid.
What does this practical legal approach mean in terms of the characterisation of supplies? There are three main competing interpretations of the GST concept of supply:
(i) every supply has a true or essential character, which can be determined by identifying the subject matter of the transaction. That subject matter is the thing that the supplier supplies and that the recipient gives consideration to acquire;
(ii) every supply can be re-characterised as a supply of the creation of a right to acquire something (generally a contractual right);
(iii) every supply involves both a supply of rights and a supply of the thing to which the rights relate, each supply being separate, and neither being incidental to the other.[10]
Ecclesiastes and common sense should be sufficient support for the proposition that “to every supply there is a time”. But it is difficult to say with certainty when that time of supply is. Each of the above views will have a corresponding time of supply:
(i) the time of supply is the time at which the thing is done; in other words, time of supply is determined by reference to performance;
(ii) the time of supply is the time of contract, because this is when the rights are created;
(iii) the time of supply is whichever occurs earlier, the creation of the rights or the performance; or alternatively, it is only when both events have occurred.
This article is predicated on the view that the first option is correct. In most cases, it is hard to justify the view that the formation of a contract involves the making of a supply for GST purposes. The better view is that a contract creates the legal setting in which a supply will take place: the supply is made under the contract.[11] A recipient pays consideration to a supplier for the thing supplied, rather than for the right to have it supplied. While the rights are clearly created (and therefore supplied), they are ancillary or incidental and “do not have economic value and independent identity separate from the transaction”.[12] Not only is this a practical, commonsense view, but it correctly identifies what it is that recipients are giving consideration “for”. This is clearly the view of Australia’s Commissioner of Taxation: the concept of the true and essential character of a supply peppers the GST rulings.[13] For GST purposes, unless the creation of contractual rights is the intended subject matter of the transaction, or the rights are dealt with separately after their creation, the only feasible approach is to focus on performance of the “real transaction”, characterising the supply according to its true or essential character.
The Canadian GST specifically provides that the entry into an agreement is a supply, with the performance of the agreement being part of that supply rather than a separate supply. While not directly relevant to the interpretation of Australian GST, the presence of this rule illustrates a choice to over-ride the ordinary meaning of “supply” by legislatively favouring a “contractual rights” characterisation.[14] There is no such specific legislative provision in Australia.
Those in Australia who prefer one of the “contractual rights” approaches (thus concluding that the position is the same as that in Canada), rely on para 9-10(2)(g), which provides that a supply includes “the entry into or release from an obligation”, and reinforce their views by reference to one aspect of Lord Millet’s judgment in Customs and Excise Commissioners v Redrow Group plc.[15] The author’s view is that Lord Millet’s focus in Redrow was on the input tax credit entitlements of the recipient and not on re-characterising the nature of the supply. Nothing in his judgment supports the view that the real estate agents provided anything other than real estate agency services. His statement that “the doing of those acts constituted a supply of services to Redrow”,[16] supports the view that the underlying nature of the supply was unchanged, and that its time of supply was when the services were performed.
Australia provides particularly fertile ground for this focus on contractual rights, in part due to the High Court decision in FC of T v Orica.[17] But the fact that contractual rights are Capital Gains Tax (“CGT”) assets, and that their satisfaction is a CGT event,[18] does not mean that the subject matter of every executory contract should be characterised as a supply of contractual rights, nor that every executed contract should be treated as involving two separate supplies, an initial supply of contractual rights with a subsequent supply at the time of performance, when the rights are extinguished by satisfaction.[19]
Those in favour of the third view of a supply place reliance on para 9-15(3)(a): at the point of entering into the contract, there is a supply of a contractual right to acquire something, the consideration for which is the payment (or the promise to pay); when the contract is later performed, there is no additional consideration, and so no further taxing point. Thus, where there is a prepayment and future performance, the time of supply is the time at which the contractual rights are created. The future performance might be a supply, but it is made for no consideration and so has no consequences. This approach requires the elevation of a rule about consideration to the status of both a time of supply rule and a characterisation rule and it is questionable whether this is a valid approach. It is submitted that para 9-15(3)(a) is neither intended for, nor capable of, turning all supplies into supplies of contractual rights made at the point of entry into a contract. In keeping with its placement in the definition of consideration, its purpose is to deal with those specific situations where the contractual rights have a separate identity and a sufficiently independent existence to be the subject of an independent transaction. Such transactions involve an initial supply whose “true character” is a supply of a right or option. Paragraph 9-15(3)(a) operates to prevent the payment for the right being counted twice (as consideration for both the supply of the right and the supply made when the right is exercised). The provision is aimed at intentional two-step transactions, not at simple prepayments.[20] It is true that the consideration for the right or option is, in common parlance, a “prepayment” for the later supply made when the right or option is exercised. But it is more than a prepayment for the performance of future supplies: it is also a current payment for the right to acquire those future supplies. If the recipient does not exercise the right or option, he/she is not entitled to complain about the supplier’s non-performance, because the right or option had a separate identity sufficient to make its supply the subject matter of a separate transaction.
Simple prepayments can be more effectively dealt with by the nexus requirement. A single-step transaction involving a prepayment is not consideration for the creation of the contractual rights, it is consideration for the future supply of goods, services, or whatever is the true subject matter of the transaction. Even if para 9-15(3)(a) has a theoretical application, its assistance is not required, because the initial payment is not consideration “for” the right or option to acquire something; the payment does not “attach” to the right because the right is ancillary or incidental to the supplies made on performance.[21] If the supplier does not perform, the recipient does have something to complain about.[22]
As indicated, the author does not support the rights-based approach to GST. Such mind-numbing complexity hardly seems appropriate for what is intended to be a tax capable of being understood and applied by the modern Australian version of “the man on the Clapham omnibus”. That GST is intended to be capable of such comprehension is supported by Justice Graham Hill, who has suggested that a rule of interpretation which “requires GST legislation to be interpreted in a practical or business-oriented way … that is not overly technical” might be appropriate.[23] Both the self-assessing nature of GST, and the fact that it applies generally to a wide variety of transactions, lead him to conclude that “it should therefore be inferred that the legislation is intended to be intelligible to all who are concerned with administering entities made liable to output tax or entitled to input tax credits”.[24] With well over two million entities registered for GST, the number of individuals who must regularly make decisions about how GST applies to particular transactions must be similarly high.[25]
Moreover, a conclusion that the subject matter of the tax is performance rather than the creation of contractual rights is consistent with both Cnossen’s practical legal concept of consumption, and a more theoretical economists’ concept of consumption.
While the Australian GST law contains comprehensive real time of supply rules for transitional purposes,[26] the A New Tax System (Goods and Services Tax) Act 1999 (Cth) (“GST Act”) itself has almost no rules identifying the time of supply. However, this simply means that there are no legislated rules for identifying the real time of supply.
Of course, all blanket statements should be followed by disclaimers, and it is best to start by acknowledging that there is in fact (almost) one time of supply rule in the GST Act, albeit one that is also a characterisation rule. Section 9-10(3A) prevents a retention of title clause from transforming one type of supply (a taxable supply of livestock or game for slaughtering or processing into food) into another type of supply (a GST-free supply of food). It achieves this both by characterising the supply as a supply of livestock or game, and by providing that the time of supply is not on or after the transfer of title. So, in reality, it is a “not the time of supply” rule: the time of supply occurs some time before title transfers, but we are not told exactly when. What we are told is that it is a “for the avoidance of doubt” rule, clarifying what would have been the rule in any case, and to this extent subs 9-10(3A) confirms what should be second nature to GST practitioners: that transfer of title is not always a critical element of a supply. Since “supply” is defined as “any form of supply whatsoever”,[27] aspects of a transaction for supply can clearly take place before, after, at the same time as, or without a transfer of title.
Subsection 9-10(3A) brings to the fore the sorts of questions that inevitably arise: What is a supply? When does it take place? How, without specific rules in the Act, can a supplier be certain when it has made a supply? This article takes the view that the time of supply is the time of performance, but that does not solve the problems because even performance is not easy to pinpoint. When does a livestock owner make a supply of livestock to the slaughterhouse? Is it at the time of delivery to the slaughterhouse? Or at the point in time when it becomes impossible for the supply to be cancelled? Is it in that nanosecond before the slaughtering happens, or at the moment when the slaughter happens but before the processing transforms the livestock into something we might call food?
If and when they matter, these questions, being unanswered in the Act, will fall to the Courts to answer. For the purposes of this article it is unnecessary to attempt a definitive exposition of the time of performance for every possible type of supply. The issues of why time of supply matters and what role it plays in the grouping regime can be illustrated even without such knowledge.
Even a brief perusal of the GST Act uncovers numerous provisions for which the real time of supply is relevant. Many of those provisions also make it clear both that time of supply and time of attribution are different things, and that performance appears to be the most likely focus for determining liability. The list below identifies the most obvious time-dependent provisions.
• Determining whether a supply is taxable.
(The point in time at which fulfilment of each of the elements of s 9-5 must be considered can only be the time when the supply is made).
• Determining whether a supply is connected with Australia.
(Under s 9-5 the supply must be connected with Australia at the time of supply, and s 9-25 determines this by reference to factors that are clearly indicative of performance).
• Determining whether an acquisition is creditable.
(The elements of s 11-5 must also be considered at the time the acquisition is made).
• Characterising a supply.
(The nature of a supply must also be determined at the time of supply. As the livestock-for-slaughter example shows, time is important to the question of whether a supply is of food – s 38-2. The same is true for medical and other health services – ss 38-7 and 38-10, and no doubt for other types of supply).
• Characterising a right to a supply.
(Paragraphs 9-30(1)(b) and 9-30(2(b) require a supplier to predict at the time it supplies a right to receive a supply, whether the later supply will be input taxed or GST-free. Whether that decision should be made as if current factors apply, or on the basis of the anticipated situation when the later supply will be made, the decision itself must be made at the time of the initial supply of the right).
• Determining whether an export of something other than goods or real property qualifies for GST-free treatment.
(The application of s 38-190 depends, in some cases, on the location of particular entities at the time when “the thing is done”. In many cases, this will be synonymous with the time at which the supply is performed, whether or not that is the same as the time at which the supply is made).
• Determining whether a sale of a business qualifies for a GST-free treatment as a supply of a going concern.
(Section 38-325 requires the supplier to carry on the enterprise until the day on which the supply is made).
• Determining whether a supply can “be treated as if it were not a taxable supply” under the GST grouping regime.
(Paragraph 48-40(2)(a) only applies to a supply that is made by one member of a GST group to another, something that can only be determined at the time when the supply is made).
• Determining the tax period in which a supply to your associate first “becomes” a supply connected with Australia.
(For certain supplies between associates for no consideration, ss 72-15 and 72-50 attribute the GST and input tax credits to the tax period when the supply first becomes a supply that is connected with Australia. This is essentially an attribution rule based on time of supply).[28]
• Determining whether an incapacitated entity or its representative has an adjustment.
(Paragraph 147-20(1)(a) refers to supplies or acquisitions made before the representative was appointed).
• Making adjustments on cessation of registration.
(Section 138-15 presupposes that the taxable status of the supply was determined when it was made, at a time when the entity was registered, and it is merely the trigger for attribution that has not yet occurred).
• Measuring a taxpayer’s current annual turnover and projected annual turnover.
(Sections 188-15 and 188-20 refer to the value of supplies made during particular periods, not to payments received during those periods, nor to amounts that would be attributed to those tax periods if the supplier were registered).
No doubt there are others, but those listed should be sufficient to illustrate the importance of time in Australian GST. Section 29-25 is also significant, because it makes clear the separation between attribution and liability rules, and also because of the way it highlights the various factors that might be relevant to identifying time of supply. Paragraph 29-25(2)(d) empowers the Commissioner to vary the attribution rules for a supply that occurs before the supplier knows it has occurred, thereby presupposing that there is a time of supply. Paragraph 29-25(2)(e) makes a similar assumption when it empowers the Commissioner to vary the attribution rules for a supply that occurs before the supplier knows the total consideration. Both paragraphs apply equally to the attribution of the input tax credit for the acquisition by the recipient, which presupposes that time of supply and time of acquisition are contemporaneous, ie that supplies do not slip into time warps and emerge later (or earlier) as acquisitions.[29] Such a view is in keeping with the Commissioner’s view that the concept of a supply involves something passing from one entity to another,[30] although certain types of supply (creations of rights, entries into obligations, grants of interests in property) involve the supplier creating something which the recipient thereby acquires, without the thing acquired necessarily passing from one to the other.
The remaining paragraphs in subs 29-25(2) make reference to features of supplies that could assist in identifying time of supply: passing of possession or title in goods, the use or enjoyment of a thing supplied, the time when a supply becomes certain (for conditional contracts or those subject to statutory cooling off periods). These events represent aspects of the performance of a contract, and the section clearly contemplates that performance is the stuff of supplies. Supplies are made under contracts rather than at the point when the contractual rights arise. What subs 29-25(2) also makes apparent, if it were not already so, is that Australia’s attribution rules operate independently of the actual time of supply. In most cases the disconnection between time of attribution and time of supply will not be an issue. However, because attribution can occur before the time of supply (see Section 4.3 below), the Commissioner is given power to deal with a range of situations where there is significant uncertainty about some aspect of the supply (including whether it will be cancelled before performance commences or is completed).
A foreign reader might be confused by the assertion that time of supply and attribution are independent, because the equivalence of the two concepts is ingrained by common usage. The Australian attribution rules look like the time of supply rules in other jurisdictions and perform the same function as time of supply rules, so how is it that they are different? As indicated, the difference is mainly one of terminology, but the UK cases illustrate one difference, which is the way that time of supply rules are sometimes interpreted to deem not only a time of supply but also a supply.
To compare the application of the GST Act to the UK cases it is first necessary to review the relevant UK provisions and consider the main differences and similarities between the two jurisdictions. The essential difference between the two regimes is that Australia never deems there to be a supply, whereas the UK law occasionally does so (but only if there is an underlying supply).
Under subs 1(2) of the Value Added Tax Act 1994 (UK) (“VAT Act”):
VAT on any supply of goods or services is a liability of the person making the supply and (subject to provisions about accounting and payment) becomes due at the time of supply.[31]
Australian GST deals with these concepts less succinctly, achieving the same effect through a combination of s 9-40 (who is liable) and Subdiv 29-A (attribution to tax periods). Australia’s GST maintains a strict separation between the time of supply (which is relevant to whether GST is payable) and the obligation to make payment. It will become evident that the UK provisions contain the seeds of a similar separation. Subsection 1(2) cannot operate unless a supply has been made. If there is a supply, the section determines the “Who?” and “When?” of liability: the supplier is liable and the payment becomes due at the time of supply.
To decide whether a liability arises, and when it is payable, one must first characterise the supply, because different time of supply rules apply for different types of supply. Three main categories of supply are relevant for time of supply purposes in the UK: supplies of goods, supplies of services, and continuous supplies of services.[32] It is significant that because transfer of “title” (or at least, of the right to dispose of goods as owner) is an essential element of a supply of goods, if this “taxable event” does not occur, one must conclude either that the supply did not take place, or that it was a supply of services.[33]
Australian GST does not subdivide supplies in the same way as the UK, but there is nonetheless a concept of taxable event. At the very least, supplies must be subdivided into supplies of “goods”, “real property”, and “things other than goods or real property” because of the structure of the “connected with Australia” requirements in s 9-25. Even within these categories, there will be different types of supply (sales of goods, leases of goods, rights to use goods, etc), each with a different “event” that confirms the supply has taken place. The taxable event will have an equally important role in determining time of supply in Australia’s GST. However, because the Australian attribution rules apply equally to all types of supply, the nature of the taxable event does not affect the time of attribution.
Under UK VAT law, the first point of call for identifying the time of supply is essentially what might be called “real time proxy” rules. These “actual” time of supply rules take one feature of the supply transaction, based in real world events, and determine that feature to be the time of supply.[34]
• For goods: the time of supply is:
− when the goods are removed, or
− if they are not to be removed, when they are made available to the recipient of the supply.[35]
• For services: the time of supply is when the services are performed.[36]
This use of real time proxies complicates the analysis of the concept of taxable events, since the real time proxies will tend to correspond with the actual time of supply in the real world. Although the distinction is subtle, even if a real time proxy has occurred, the supply may still not in fact have been made if the taxable event does not subsequently occur. It will be seen that in such cases the “real time proxy” time of supply does not create a supply if one has not taken place. This is most obvious for supplies of goods, where transfer of the right to dispose of goods as owner is essential.
There are no equivalent provisions in Australian GST; the GST Act is simply silent on the real time of supply. Similar concepts appear at various places in the GST Act, in particular in s 9-25, where the concepts of delivery, making available, and removal of goods appear, as does the phrase “the thing is done”, which, as the Commissioner has noted, might be equated with the performance of services.[37] What is missing is any statement of the relevance of these events in determining a time of supply. Their presence in the GST Act might suggest to the Courts that they have some role in identifying steps in the process of making a supply, but the GST Act does not elevate them to the status of timing or taxable event rules.
For both goods and services, the time of supply for UK VAT can be brought forward if a VAT invoice is issued or payment is made before the “actual” time of supply.
• Earlier invoice/payment: if a tax invoice is issued or payment is received before the actual time of supply, then the supply is treated as taking place at that earlier time, but only to the extent of the payment or the amount invoiced.[38]
• Later invoice/payment: if a tax invoice is issued up to 14 days after the actual time of supply, the time of supply is the date of invoice, unless the supplier opts not to use this rule.[39]
For continuous services: a continuous supply of services over several tax periods is treated as a series of separate supplies, each of which is made at the earlier of payment or issuance of a tax invoice.[40]
Thus, in most cases, the time of supply depends on both the “actual” time of supply and the “deemed” time of supply. For continuous services, however, the deemed time of supply rule is the only rule that applies: if no payment has been made and no invoice issued, the supply is treated as not having yet been made. Moreover, as will become evident when the grouping schemes are discussed, the continuous services rule has been interpreted as deeming both a time of supply and a supply, whereas the rules for goods and non-continuous services merely deem a time of supply.
Again, there is no direct equivalent in Australian GST. The tax period to which liability for tax is attributed is determined by attribution rules, which make no reference to anything labelled the “time of supply”. Since the attribution rules make reference directly to the time of invoice and/or payment, rather than to the time of supply, there is no need to have deemed times of supply to
co-ordinate these events. For whatever reason, the legislature appears to have decided that if no deemed time of supply rules are required, there must also be no assistance required in identifying the real time of supply.
Nonetheless, it is clear that the Australian attribution rules are analogous to the UK deemed time of supply rules in function.
• For all supplies, irrespective of their character: GST is attributable to the earliest tax period in which either any of the consideration is received or an invoice is issued for the supply, subject to variations for taxpayers accounting for GST on a cash basis.[41]
For all supplies that are both provided and paid for progressively or periodically: the same rule applies as above, but applied as if each part of the supply were a separate supply. If those parts cannot easily be identified, then they may be correlated with the components of the payments.[42]
Thus, the point that has already been laboured is made once more: Australian GST achieves attribution without making reference to the real time of supply. If a supply takes place before any invoice or payment, no GST will be attributable at the time of supply. In contrast, under the UK rules, if a supply of goods and services takes place before payment (and more than 14 days before invoice) the full VAT liability becomes due.[43] Accruals based attribution in Australia achieves, for all supplies, a similar effect to the UK treatment of continuous supplies, except that it does so while remaining silent on the question of whether, or when, the supply has taken place.
The UK law includes a series of special time of supply rules, most of which are of little relevance to this article. They deal with matters such as intra-community acquisitions, supplies for which recipient created tax invoices are issued, applications to own use, disposals of assets for no consideration, warehoused goods, etc. The Commissioners may, at the request of a taxpayer, agree to treat supplies as taking place at a particular time, based on the occurrence of some event, or on the normal time at which such an event would occur – in other words, to use a different “real time proxy” than the one used in the basic “actual” time of supply rule.[44]
The Commissioners also have powers to issue regulations about time of supply, and there is an extensive list which includes special rules for goods for private use, free supplies of services, compulsory acquisitions of land, certain leases that are treated as supplies of land, supplies of water, gas, power, heat, ventilation, etc, supplies of goods under consignment, retention payments, royalties, construction, and supplies spanning a change in rate. Most of these rules use similar factors to those above: they focus on a real time proxy, on payment or invoice, or on the beginning or end of a tax period. In some cases, they deem a supply to take place, but in most cases they do not; they simply deem a time for the underlying supply.
Australia has equivalents for some of these rules in the Commissioner’s Determinations under s 29-25, which vary the attribution rules in particular circumstances, and in Div 99, which defers attribution where deposits are held as security for the performance of an obligation.
The basic purpose of allowing group registration for VAT purposes is essentially the same in the UK and Australia, but certain key differences in the way in which VAT grouping is implemented in the UK deserve mention:
• the group is treated as if it were a single entity for VAT purposes;[45]
• the representative member is treated as carrying on all the businesses carried on by the group members; and
• supplies by or to a member of the group are treated as being made by or to the representative member.[46]
As a result, during the period for which companies are grouped, only the representative member is registered for VAT (since the other members are treated as if they are not making any supplies or acquisitions). This approach is similar to the single entity rule for income tax consolidation in Australia.[47]
The Australian GST position is more complex: a GST group is “approved” under Div 48, but does not have a separate registration. Each member of the group must continue to be separately registered for GST.[48] No departure is made from the real identity of the entity making the acquisitions and supplies, and for that reason there is no need to treat the representative member as carrying on the enterprises of its members. Nonetheless, certain key decisions or calculations are made as if the GST group were a single entity, in particular whether an acquisition or importation by a group member gives rise to an input tax credit entitlement[49] and the amounts of any input tax credits and adjustments.[50] Moreover, as in the UK, the actual payments, credits, and adjustments are made by the representative member, and the other members are relieved from the requirement to submit GST returns while they are members of the group.[51]
The key effect of grouping on potentially taxable intra-group supplies is similar in both jurisdictions.
• In the UK: “any supply of goods or services by a member of the group to another member of the group shall be disregarded”.[52]
• In Australia: “a supply that an entity makes to another member of the same GST group is treated as if it were not a taxable supply” and “an acquisition that an entity makes from another member of the same GST group is not a creditable acquisition”.[53]
As indicated, the UK provisions treat the group as a single registered entity. “Disregarding” intra-group transactions fits logically with that approach. While the Australian “multi-entity” approach is intended to achieve much the same effect, it is significantly more complex in operation.
Both the Australian and UK laws have rules to deal with two different aspects of the entitlement to input tax credits:
• initial entitlement rules; and
• adjustments to reflect actual use.
Under the VAT Act, “input tax” is defined as tax charged on supplies to the taxpayer of goods or services used or to be used for the purpose of any business carried on or to be carried on by him.[54] “Credits” for input tax are only allowable if the input tax is attributable to the making of taxable supplies.[55] Where input tax has initially been attributed to taxable supplies the entity intends to make, but before that intention is fulfilled, and within a 6 year period after the start of the tax period in which the attribution was determined, the entity uses or forms an intention to use the acquisition wholly or partly for making exempt supplies, the taxpayer is required to account for an amount equal to the input tax that is no longer attributable to taxable supplies. This adjustment, to use Australian terminology, must be made at the time of the altered use, or the time of formation of the altered intention, whichever is applicable.[56]
Like the UK VAT law, Australian GST makes initial entitlement to input tax credits dependent on a connection between the acquisition and the taxpayer’s enterprise.[57] However, whereas UK taxpayers must establish a positive link with making taxable supplies before they are entitled to credits, Australian taxpayers need only show that their acquisitions are neither private/domestic in nature, nor related to making supplies that would be input taxed.[58] The initial input tax credit entitlement is based on the acquirer’s intended use of the acquisition. A taxpayer can be required to make adjustments if the initial assessment of entitlement was incorrect. Depending on the nature and value of the acquisition, its use (“application”) must be monitored for between one and ten adjustment periods.[59] Although the principles for change in use are broadly similar in both regimes, Australian adjustments are based on the actual use or application of an acquisition, and do not encompass changes in intention. Moreover, adjustments are made annually in the June tax period, rather than at the time when a change in creditable purpose occurs.[60]
After a somewhat extensive scene setting, the significance of time of supply can now be illustrated by reference to the three UK cases on group exit and entry schemes:
• CCE v Thorn Materials Supply Ltd and Thorn Resources Ltd;[61]
• CCE v Svenska International plc;[62] and
• BUPA Purchasing Ltd v CCE.[63]
In simple terms, each of the schemes involved three entities:
• an exempt group member (“E”): for one reason or another E has less than a full entitlement to input tax credits;
• a non-group supplier (“S”): it is acquisitions from this entity for which E would be denied input tax credits; and
• an intra-group supplier (“GS”): the interposition of this entity between S and E is necessary for the operation of the schemes.[64]
GS and E are at all times members of the same wholly owned corporate group. If they are not grouped for VAT, tax will flow through the transaction chain until it reaches E, where it will be blocked. If GS and E are grouped for VAT, S will charge VAT on its supply, which will be treated as being made to the representative member of the group. The “group” will be unable to claim input credits because of E’s use of the acquisitions for purposes other than that of making taxable supplies. Clearly, it is not grouping per se that forms the basis of the schemes: the single entity treatment simply shifts the input tax denial to the representative member.
The critical step in the schemes involves manipulating the timing of the transactions. The time of performance is separated from the time of payment/invoice under the intra-group supply contract, with most or all of the contract price being paid by E to GS while the entities are VAT-grouped. All aspects of the external contract take place while E and GS are not grouped, so that it can clearly be said that GS “makes” its acquisitions when it is not VAT-grouped with E. The objective is to trigger a deemed time of supply for the intra-corporate group supply while the companies are VAT-grouped (so that the supply can be ignored under the grouping rules). Group exit schemes do this using prepayments; group entry schemes use deferred payments. Although the purpose of the schemes is to circumvent a denial of input tax credits to E, in all three cases the taxpayer whose returns are challenged is the interposed group supplier.
In Thorn Materials, the objective was to circumvent the denial of input credits on acquisitions of cars intended for retention rather than for re-sale.[65] The taxpayer and E (the intended purchaser of the cars) were members of the same corporate group and were grouped for VAT. While grouped, the taxpayer contracted to sell cars to E, and a prepayment of 90% of the purchase price was made (an amount close to £34 million).[66] At this point in time, the taxpayer had not acquired, contracted for, been invoiced, or paid for any of the cars it would have to purchase in order to meet its obligations under the contract with E. The two companies then ungrouped for VAT purposes, after which the taxpayer acquired the cars and “on-supplied” them to E, invoicing the remaining 10% of the purchase price, which was duly paid.
The taxpayer argued that the actual time of supply rule was displaced by the deemed time of supply rule because of the earlier payment, with the result that to the extent of the 90% prepayment, the supply took place while the companies were grouped and so must be disregarded. There was a further time of supply at the time of removal (when the goods were physically supplied), which was taxable to the extent of the 10% final payment. This final payment ensured that the taxpayer would be seen to have acquired the cars for an onward taxable supply, albeit at a price considerably lower than its cost of acquisition. This would, the scheme designers presumed, cement the taxpayer’s entitlement to input tax credits. Thus, what was in the real world a single supply of cars to E would escape tax on 90% of the price, effectively converting E’s zero% input tax recovery rate into a 90% recovery rate.[67] Not surprisingly, the House of Lords appeared eager to reject this line of argument. In doing so, their Lordships read the UK legislation as if it operated in the way that the Australian legislation does: they recognised that VAT grouping did not, in reality, undo the separate existence of the group members, and more importantly, they read the time of supply rules as if they were merely “attribution” rules, whose function was to determine the time at which VAT was payable.
The key points in the reasoning were as follows:[68]
• a supply cannot be disregarded under the grouping rules unless the supply takes place while the entities are grouped; if not, there is nothing that can be disregarded;
• the purpose of the grouping rules is to simplify the collection of tax, it is not “to confer exemption or relief from tax”;[69]
• requiring intra-group supplies to be disregarded does not mean that the existence of the transaction or the separate identities of the group companies should be disregarded;
• while the supplies must be disregarded at the time of the prepayment, once the taxpayer left the group it was once again a separate taxable person for VAT purposes, and when it then delivered the goods, it made a supply of goods in the course of its business;
• the only relevant supply was the actual supply of the goods, which took place when the companies were no longer grouped;
• in keeping with the requirements of the UK VAT law and art 11A(1) of the Sixth Directive,[70] the taxable value of the supply was everything that the taxpayer had received for making the supply, including both the 10% payment at the time of performance and the 90% prepayment made while the companies were grouped.
Lord Clyde, in his concurring judgment, rejected arguments that the deemed time of supply rules effectively created two supplies, one to the extent of 90% made when the prepayment occurred and another to the extent of 10% at the time of removal. He concluded that the requirement to disregard intra-group supplies applied only to “… supplies completely carried out between the members of the group” and not to the “the kind of deemed part supply” created by the deemed time of supply rules.[71] There was one single supply of cars by the taxpayer, and that supply was fully taxable because it took place when the companies were no longer grouped for VAT.
The next decision of the House of Lords involved a group entry scheme based on deferred payments. The exempt supplier in the Svenska scheme was the newly formed UK branch of a Swedish bank. The taxpayer was the existing UK subsidiary of the bank, and was the representative member of a VAT group. Since the newly formed branch had no staff or premises, it entered into an arrangement with the taxpayer for the provision of “management services”, which effectively covered its entire operations. Under the agreement between them, the branch would pay the taxpayer a proportion of the latter’s total costs, calculated by reference to their comparative revenues. For VAT purposes, this arrangement was characterised as a continuous supply of services.
The branch was unable to claim input tax credits, initially because it was not registered for VAT, and then later because its supplies were predominantly exempt financial services. The outsourcing of its entire operations put the branch at a significant disadvantage compared with other financial service providers. Grouping could alleviate this problem, at least to the extent of the taxpayer’s untaxed inputs such as wages. Unfortunately, the branch did not meet the requirements for being a member of a VAT group because it was a branch of an unregistered foreign entity, and at that time only resident companies could be grouped.[72]
During the time that the branch and the taxpayer were unable to be grouped for VAT, the taxpayer continued to provide services to the branch, and to make taxed acquisitions that related to those services. However, presumably because moves were on foot to have the grouping rules changed, and in anticipation that the branch might eventually be able to be grouped, the taxpayer deferred charging for the management services. No invoices were issued, and no payments made, for five years, until nearly a year after the branch was eventually brought into the VAT group.
The taxpayer argued that, to the extent that its taxed acquisitions during the five year period were related to its on-supplies to the branch, it could claim full input tax credits because the acquisitions related to supplies to the branch, which at the time could only be taxable. When an invoice was eventually issued for the management services, the time of supply rule for continuous supplies would deem a supply to be made at that point. That deemed supply would be made between two VAT-group members and would therefore be disregarded for VAT purposes. Since the branch would not be charged tax, there would be no input tax for which to deny deductions.
After Thorn Materials, one might have expected the House of Lords to look past the deemed time of supply rules and treat the taxpayer’s services as taxable, on the basis that they took place before the companies were VAT-grouped. Instead, their Lordships approached the scheme from the perspective of the taxpayer’s entitlement to input tax credits for its earlier acquisitions. In his dissenting judgment, Lord Lloyd of Berwick suggested that the case could have been decided on the same basis as Thorn Materials, but it appears that counsel for both sides had been unwilling to argue on that basis.[73] Given the reasoning of the majority, one might wish that they had thought again.
The majority took the view that the time of supply rule for continuous supplies was in different terms from the other time of supply rules. It precluded any reference to actual time of supply and deemed not only a time of supply, but also the making of a supply at the time of each payment. This made it impossible to follow Thorn Materials by treating the pre-group performance of the management services as a taxable supply, for which the payment between VAT-group members would have been consideration. The only relevant supply was the deemed supply, which the time of supply rules treated as having been made when the companies were grouped. Thus, this deemed supply had to be disregarded.
Because no tax was payable on the intra-group supply of services, the House of Lords could only attack the scheme by blocking the taxpayer’s input tax credits. There was no question but that the taxpayer had initially been entitled to its input tax credits. When it actually made the acquisitions (and the on-supplies) the input tax was rightly attributed to intended taxable supplies, being the on-supplies of management services to the branch, with which it was unable to be grouped. This link to taxable supplies was not affected by the fact that the intended supplies would only be recognised as happening when a tax invoice was issued or a payment made. Despite this initial entitlement, the House of Lords took the view that the change of use provisions applied. The taxpayer’s initial intention to make taxable supplies to the branch had been superseded by a subsequent change of intention. At the time when the branch was brought into the VAT group, the taxpayer had formed an intention to use the earlier acquisitions for a purpose other than making taxable supplies, and had later actually used them for that purpose when payment triggered the deemed supply.
This conclusion required the suspension of all consideration of actual time of supply. In “the real world”, the taxpayer had made acquisitions and immediately consumed them in supplying services to the branch. Thus, in the real world, the acquisitions clearly were not used again later, either at the time of formation of the group or the later time of invoice. Nonetheless, the House of Lords held that the suspension of reality required by the time of supply rule for continuous supplies should be maintained in interpreting the interaction between that rule and the adjustment rules:
The requirement … in relation to continuous supplies of services that services which, in the real world of commerce, have actually been supplied to, and already used by, another person are not to be treated as supplied until a payment has been received or a tax invoice has been issued gives support to the view that such supplies can be deemed to be used at a time subsequent to their “actual” use and gives weight to Mr Pleming’s submission that such supplies do not simply “disappear into limbo”… [A]rtificial though the concept is, … the effect of [the then input tax credit rules] … is that the supplies received by Svenska, for which it received credit, were used or appropriated for use by it in making partly exempt supplies to the customers of the London branch after [the date of formation of the VAT group].[74]
In other words, if the acquisitions were used to make deemed supplies, then they could be deemed to be applied at the time of making those deemed supplies, and the entitlement to credits on that deemed application could be measured by reference to the earlier supplies by the branch to its customers. It was irrelevant that all actual supplies and acquisitions had long been completed. Perhaps the oddest aspect of this interpretation was that the supplies to the customers of the branch, to which the input tax was now attributed, had in fact taken place earlier, at a time when the companies were not grouped. The past was simply deemed to be the present, in every way.
Although the phrase “make-believe world of VAT” appears only once in the judgment, its counterpart phrase “the real world” appears twelve times, and on each occasion, it is used in contrast with the decision being made. Concerned to explain this approach, Lord Hope of Craighead said:
This conclusion is not easy to grasp if regard is had to what was happening in the real world. But the statutory scheme does not always follow the real world. The guiding principle as to relief for input tax as against output tax is that of fiscal neutrality …[75]
Thus, a less than coherent interpretation of the grouping provisions was justified by what, in the European Union, has come to be an over-riding principle of VAT. The principle of fiscal neutrality essentially ensures that the effective rate of VAT is as close of possible to the intended rate. It requires that suppliers should not remit VAT on more than the actual amount they have received,[76] nor on less than they have received.[77] Nor should the VAT be more than the appropriate percentage of the amount paid by the recipient of the supply.[78] It ensures that input tax credits are allowed where appropriate, and denied when not appropriate. In short, with its focus on the amounts paid for supplies, the principle of fiscal neutrality embodies the view of VAT as a tax on private consumption expenditures. Whether Australia has, needs, or will develop a version of the principle of fiscal neutrality only time will tell.
BUPA Purchasing is a more recent High Court Chancery Division case, which combined the issues in Thorn Materials and Svenska because it involved a group exit scheme for continuous supplies of services. The exempt supplier in the group was a large private hospital which supplied exempt medical services. Its normal input tax recovery rate was between 5 and 10%. At a time when the companies were grouped for VAT, the taxpayer contracted to provide ongoing services to the hospital and received a 98% pre-payment for the future services. The taxpayer then exited the group, after which it proceeded over a period of time to make acquisitions and on-supply services to the hospital in return for payments that gradually made up the outstanding 2% of the contract price. No doubt on the authority of Svenska, both the taxpayer and the Commissioners agreed that the time of supply rule for continuous supplies required the taxpayer’s supplies to be treated as separate deemed supplies, made to the extent of the payments.
In contrast with Svenska, where the House of Lords had struck down the group entry scheme by requiring the taxpayer to make an increasing adjustment for its earlier acquisitions, the BUPA Purchasing scheme was a group exit scheme, which meant that the taxpayer’s initial claim for input tax credits was in issue. The taxpayer argued that it should be entitled to full input credits because its acquisitions related to making taxable supplies, being the deemed supplies made to the extent of the 2% final payments. The Commissioners argued, and Park J agreed, that the taxpayer’s acquisitions related to both sets of deemed supplies: the earlier 98% supply (which was disregarded under the grouping rules) and the later 2% supplies (which were taxable). If the real supplies were apportioned 98% to 2% to deemed supplies, then as a matter of logic, the proportion of the acquisitions that related to those deemed supplies should also be divided 98% to 2%. Since credit for input tax requires a positive link to taxable supplies, only 2% of the input tax was recoverable.[79]
Unfortunately for the taxpayer, this meant that the effective recovery rate was worse than it would have been had the scheme not been entered into. Having attempted to improve the hospital’s recovery from between 5 and 10% to 98%, recovery for the corporate group in fact fell to less than 2% as a result of the decision.[80]
How would similar cases be decided in Australia? In the absence of artificial deemed supply rules, liability must be decided on the basis of the real time of supply. It follows that all three cases would be decided on the basis of the taxpayer’s output tax liability for its on-supplies, rather than on the basis of its input tax credit entitlement for the related acquisitions. This was the outcome in Thorn Materials, which relied on the actual time of supply, rather than the time of supply rules, to determine when the supply took place. Of course, this conclusion necessitates a decision about how the real time of supply should be identified.
As stated earlier, characterising a supply as performance does not, in itself, solve the problem of determining when a supply takes place. Performance is not an instantaneous process. Even a simple supply such as a sale of goods takes place over a period of time, no matter how short that time might be. A supply by way of sale is a process under which possession and ownership of the goods pass from the supplier to the recipient. The time of supply might be any, all, or some combination of the following:
• the time at which the goods are in some way identified or put aside by the supplier for the recipient;
• the time at which the goods are delivered or made available to the recipient of the supply (or to someone else at the recipient’s direction);
• the time at which it becomes certain that the supply will be or has been made; or
• the time at which title to the goods transfers to the recipient.
The “actual” time of supply rule in the UK focuses on one of the key moments in the early stages of the supply process: the time when the goods are delivered or made available. The Australian GST Act provides no legislative guidance on identifying the actual time of supply and makes no recourse to real time proxies. Neither, for that matter, does the New Zealand GST law.[81] It seems likely therefore, that both Australia and New Zealand must use a more complex approach, determining time of supply on a facts and circumstances approach, with the thing supplied and the means by which it is supplied being relevant factors.[82] Any rules about time of supply will have to be established by case law and will inevitably be shaped by the different types of supply that come before the Courts for consideration.
In the case of services, the process of supply is more nebulous. Depending on the nature of the services, the supply may involve the supplier gathering and processing information and/or goods, forming these into a coherent package, and delivering the benefit of the services to the recipient of the supply in some way. Even the UK rule that the actual time of supply is the time when services are performed leaves open the question of exactly when that is. In this regard, the UK law is in the same position as the Australian and New Zealand laws – it is effectively silent on how the time of performance should be identified.
Tim Kyle proposes a range of possible times of supply for a simple supply of legal services, including the time of contract, the time when the advice is prepared, the time of postage or receipt of the letter of advice, the time of payment, or the start or end of the period to which the supply is attributable.[83] For reasons already outlined, this author takes the view that the time of contract can only represent the “doing of the thing” if the subject matter of the transaction is a right or obligation that the contract creates. It is therefore inappropriate to identify the time of supply for legal services on the basis of time of contract. The time of postage or receipt of the letter of advice is a rule based on delivery, and may be appropriate in circumstances where delivery is the essence of the supply and is the event that cements the occurrence of the supply. Depending on the terms of the contract, it may be equally valid to rely on the time at which the advice is prepared. On the other hand, neither the time of payment nor a time in the tax period of attribution have a sufficient relationship with the making of the supply by the supplier to be accepted as time of supply rules for legal services. Nonetheless, Kyle’s range of options illustrates the dilemma created in all jurisdictions by the absence of a clear time of the taxable event for supplies of services/things.[84]
In Australia, faced with a prepayment between two entities that are members of the same approved GST group, for a future taxable supply of goods to be made after the entities cease to be grouped, one need only turn to para 48-40(2)(a) to conclude that, if the supply is not made by one member of a GST group to another, it cannot be treated as if it is not a taxable supply.
The more difficult questions are how to determine whether the supply is made between members of a group, and how any GST payable should be attributed. On the specific facts of Thorn Materials, the answer is simple: every factor that could possibly be relevant to the time of performance of the supply (removal, delivery, passing of possession, and transfer of title) occurred when the companies were no longer grouped. Moreover, it was always intended that they would do so. It is not only unnatural but also unreflective of the true character of the supply to say that anything resembling a relevant supply occurred at the time of the prepayment. Moreover, the taxpayer was unable to actually supply the goods while the entities were grouped, because at that time it had not yet acquired the goods for on-supply. There should be no question but that the supply took place while the entities were not grouped and so would not meet the requirements of para 48-40(2)(a). Thus, the supply would be taxable and the price of the supply would be the full amount of consideration received, including the prepayment and the final instalment.[85]
As indicated, Australian GST uses the real time of supply for all types of supply, which means that the result should be the same for both a Svenska type group entry scheme and a BUPA Purchasing group exit scheme. Section 156-5 does not deem a supply to be made and so it apparently has no impact on the question of when the supply is made, but merely directs that the attribution rules be applied “as if each progressive or periodic component of the supply were a separate supply”.[86]
The services provided in both Svenska and BUPA Purchasing were “management services”. In both cases, all the relevant acquisitions and on-supplies occurred while the entities were not grouped. Thus, in Australia, such supplies would be taxable supplies and would not be excused from taxable treatment under Div 48 simply because the attribution “trigger” (the prepayment or deferred payment) took place when the entities were grouped.
In a Svenska group entry scheme, the taxpayer would initially be entitled to input tax credits because its pre-group acquisitions are made in carrying on its enterprise and are neither private/domestic in nature, nor related to making input taxed supplies. Nothing in the Act requires the taxpayer to consider its entitlement to input credits from the viewpoint of the future GST group. There is also nothing that would cause the deferred payment to trigger a Div 19 adjustment; entering the group does not cause the acquisitions to cease being creditable acquisitions. Nor is there any need to make a Div 129 adjustment, because the taxpayer has applied the acquisitions by on-supplying them to the branch before they become grouped. Having been on-supplied, it is not possible for the taxpayer to re-apply the goods. A wholesale departure from real-world-time to make-believe-world-of-VAT-time is simply not required.
Similarly, the taxpayer in the BUPA Purchasing arrangements would have been entitled to all its input tax credits. Its acquisitions of services for on-supply were made when the entities were no longer grouped; they were made in carrying on its enterprise; and they were neither private/domestic in nature, nor related to making input taxed supplies.
How did it get so late so soon?
It’s night before it’s afternoon.
December is here before it’s June.
My goodness how the time has flewn.
How did it get so late so soon?[87]
Unfortunately, establishing liability is not the end of the matter. If the supplies in these grouping schemes are not made between group members, but the time for attribution occurs while the entities are grouped, who is liable to pay the GST?
The triggers for attribution are “accounting” features of the transaction – the issuance of an invoice or the receipt of some or all of the consideration. Attribution focuses on aspects of the payment process, rather than the supply process. In most cases, there will be a timing difference between the event that establishes liability (the time of supply) and the event that triggers attribution (the time of invoice or payment). This difference immediately raises the possibility that, unlike supplies, attribution can in some cases be in danger of disappearing into “time warps” if the attribution rules have not been properly integrated with the liability rules and/or the special rules.
In the normal course of events, the disconnection between attribution triggers and liability triggers should not matter: if a supply is made in one tax period and then invoiced and paid for in the next tax period, attribution will take place in the second period. But how does the Act deal with situations where liability is certain only after the tax period in which the attribution trigger occurs? One answer might be that, since there can be no liability until a supply is made, attribution must be deferred until that time. But while there is specific provision for deferring input tax credits until the recipient holds a tax invoice,[88] there is simply no mention of deferring attribution until an acquisition/supply is made. If it was necessary to ignore the attribution trigger until it was certain that a supply had been made, taxpayers would be constantly having to amend their past GST returns. Apart from the irritation this would entail, it would also generate a liability for interest on underpayments.[89]
With the exception of Div 99 security deposits, the normal expectation is that output tax (and input tax credits) will be included in the return for the tax period when the event that triggers attribution occurs, whether or not the supply has occurred. In the case of a full or part payment, attribution is triggered if the supplier has received the payment; but is it possible to import the principle in Arthur Murray[90] into Australian GST in order to defer attribution until supplies are made? It is submitted that it is not. The question in Arthur Murray was not whether the payments had been received, but whether, having been received, they had the character of income derived. The attribution rules only require a deposit to be received by the supplier, it is not necessary that it be derived as income.
This is what Lord Nolan had to say on the matter in Thorn Materials, referring to the Sixth Directive provisions on which the UK time of supply rule is based:
The article authorises the tax to be charged … in advance of the transfer of ownership where there has been a prior payment on account, but that does not displace the necessity for a transfer of ownership to follow in fact. Otherwise there is no chargeable event, and no justification for the imposition of the tax.[91]
and:
What happens, then, if there has been a payment on account, from which tax has duly been charged … but the goods are destroyed or for some other reason their supply does not occur? My Lords, tax legislation tends to be more explicit in its provision for the collection of tax than for its repayment, but Mr Prosser suggested, rightly to my mind, that the position would be covered by art 11C(1) of the Sixth Directive which provides that in the case of cancellation, refusal, or total or partial non-payment, the taxable amount is to be reduced.[92]
In other words, in the UK, you can pay VAT in advance of a supply, but if you do so and the supply does not take place, there will be an adjustment event. For goods, payment before supply can occur under both the “actual” and “deemed” time of supply rules in the UK, because both may require VAT to be paid in advance of transfer of title. These passages highlight the importance of the taxable event concept in UK VAT: for a supply of goods, the taxable event is the transfer of the whole property in goods; if this does not occur, there is no supply of goods. As indicated, for services it is less clear what aspect of performance is the taxable event, such that it can be used to confirm that the services have in fact been supplied.
Thus, even under a regime that includes an explicit “time of supply” rule, the UK VAT still looks to see that the supply has actually taken place. With the exception of continuous supplies, the time of supply rule merely identifies the time for payment. If such a regime has been interpreted to require “payment before supply”, it is difficult to see why the Australian regime should not be interpreted to require “attribution before supply”.
The attribution-before-supply problem might be thought to encourage those who prefer the “everything is contractual rights” approach, but there is a simpler analysis:
• there is no liability unless there is a supply;
• if an attribution event happens before the supply is made, output tax is attributed for the intended future supply;
• if the supply does not go ahead, there is a Div 19 adjustment event (because the altered circumstances have the effect of cancelling the supply);
• if, for some reason, the recipient of the supply had neglected, in the earlier tax period, to attribute the input tax credit for the acquisition of the now “cancelled” supply, it cannot go back to amend the earlier return because it was not entitled to the credit in the first place – the adjustment event operates ab initio. Similarly, if the supplier had neglected to remit the GST, the Commissioner cannot go back and require the GST to be remitted.
Thus, there is no need for constant amendments of past GST returns. The adjustment provisions allow the supplier and recipient to account for the altered situation when it first becomes apparent, rather than having to go back and amend past returns.
This approach can be seen quite clearly in recent New Zealand case law dealing with cancelled transactions, where despite the reliance on a deemed time of supply rule based on the earlier of invoice or payment, the Taxation Review Authority (“TRA”) has approached the issue in the same way as the House of Lords in Thorn Materials. The deemed time of supply is treated as being merely an attribution rule, the function of which is to allocate GST and input tax credits to tax periods, but only if there is in fact a supply. If a cancellation occurs, it cancels both the supplier’s initial GST liability and the recipient’s initial input tax credit entitlement. One function of the adjustment rules is to allow the cancellation to be accounted for in the period in which it occurs, rather than in the earlier period: for transactions that do not proceed, the recipient must repay the input tax credit previously claimed and the supplier can recover the GST previously paid.
In Case W22,[93] the TRA dealt with a scheme designed to take advantage of the mismatch between cash and accruals accounting in order to generate funds with which the taxpayer could finance its property development activities. The taxpayer claimed input tax credits of nearly $9 million for the “acquisition” of 114 plots of land. The input tax credit entitlement was said to result because the part payment ($10 per property) and invoice (the land contract) triggered a time of supply for the “acquisition” of the properties.[94] The scheme fell through when the Commissioner of Inland Revenue refused to pay the refunds; the taxpayer could no longer fund the acquisitions and the vendor treated the contracts as cancelled and proceeded to sell some of the land to other purchasers.
Judge Willey took the view that the taxpayer had never been entitled to input tax credits because the supply did not take place. He reasoned as follows:[95]
• the contracts had clearly been cancelled;
• there were therefore no enforceable transactions on which an input tax credit could be claimed;
• the adjustment provisions in the NZ GST law provided that if a contract was cancelled, then the recipient must refund to the Commissioner any input tax claimed.[96]
Hence there is no output tax payable at the time of supply, and therefore the full amount of the input (sic) which would have been payable to the purchaser had the contracts not been cancelled must be repaid to the Commissioner or foregone by the purchaser if the refund has not actually been made.[97]
• the payment of a deposit was not a supply: a supply is not divisible into legal and equitable interests; it is a supply of “all the rights” in the land, and since the supply had been cancelled, so the input tax credit entitlement had been cancelled;
• in determining the taxpayer’s input tax credit entitlement at the time of supply, the Court was not required to ignore the later cancellation of the supply;
• thus, the taxpayer had never been entitled to the input tax credits.
In reaching this conclusion, Judge Willey relied on his two earlier decisions in the same year, both of which also related to cancelled land transactions.[98] His view was that:
There can only be one “supply” which is affected by GST. That “supply” is of the property the subject of the contract and payment of the deposit. It is complete upon the execution of the contract.[99]
While recognising that this was problematic in light of a “time of supply” based on invoice/payment, he reasoned that the time of supply rules merely determine when a supply is made “for the purposes of liability for GST”.[100] In this, he was following previous New Zealand judicial pronouncements on the role of time of supply, including the following statement by Durie J:
Section 9 does no more than fix a time of supply for the purposes of the Act, as the section says. The purpose of the Act is to tax the supply of goods and services in the course of trade. In furtherance of that purpose s 9 appears to be no more than a mechanical provision to assist the imposition and collection of that tax by determining such matters as when tax becomes payable or deductible.[101]
Thus, in New Zealand, as in the UK, time of supply rules do not create a supply: if there is no supply, there can be no GST payable and no input tax credit entitlements. If GST has been paid, or input tax credits claimed, they are refunded under the adjustment rules. In other words, the opposite of the statement from Ecclesiastes is also true: if there is nothing, it cannot have a time. If there is no supply, it cannot have a “time of supply”. If a supply appears to have taken place, later events have undone it: like an annulled marriage, the supply never was.
Judge Willey’s views in Case W22 were recently confirmed on appeal in Ch’elle Properties (NZ) Limited v Commissioner of Inland Revenue.[102]After some discussion of whether the TRA had been entitled to rely on the s 25 adjustment rules (the significance of which had not been recognised until the TRA had asked for submissions), Hansen J rejected the contractual rights approach to characterising supplies. In response to the taxpayer’s argument that the supply for which the input tax credits were claimed had not been altered by the cancellation of the contracts because it was still technically possible for the contracts to be completed, his honour stated:
[58] This submission implies that the question of whether the vendor companies owned or had the right to acquire the land they were selling is irrelevant to the question of supply. That is not so. In analysing the nature of supply, a careful consideration of the legal arrangements actually entered into and carried out should be made in the light of the factual background …
[59] At the time of the original supply, the A companies at least had a legal right to acquire the land they were selling. That right was lost when the contracts were cancelled. The basis on which supply originally took place was utterly changed. Ch’elle recognised this by filing an amended return which reflected the cancellation. Had the Commissioner paid Ch’elle the input tax credit claims, it is unarguable, in my view, that Ch’elle would have been required under s 25 to refund the money. The Authority’s finding on this point is upheld.[103]
Thus, a number of key points can be taken from both the UK and New Zealand cases:
• even with a deemed time of supply rule, liability only arises if there is an actual supply made in “real time” in “the real world”;
• only a deemed supply rule (interpreted in the same way as the UK continuous supplies rule) can affect timing in such a way as to alter actual liability, as opposed to timing of payment (but query whether this extends to deeming a supply where there has been a payment but no services at all have been delivered);[104]
it is common for VAT and GST regimes to provide for output tax to be payable before a supply is made; and
if a supply for which the time of supply has already occurred does not go ahead, the liability never existed.
It is submitted that the same applies in Australia. There is support for this approach in the Australian GST Act. Paragraph 19-10(1)(c) refers to an adjustment event that causes a supply to become a taxable supply. The example to the paragraph refers to a supply that is intended to be a GST-free export but becomes taxable because the export does not take place. In other words, if the status determined initially turns out to be incorrect, the “error” can be fixed through an adjustment. It is not that the supply was an export and then became a taxable supply, it is simply that the expected export did not eventuate. This is no different from a supply that is initially intended to be made between members of a group, but later turns out to be made when the members are no longer grouped.
This approach also makes sense of the Commissioner’s view that there are no adjustments for adjustment events which occur in the tax period to which the initial GST is attributed – such adjustments are accounted for in that period.[105] While this is eminently practical, it only makes sense in the context of the GST Act if the effect of changing the initial consideration alters the liability ab initio. Otherwise, there is no specific mechanism in the Act for dealing with same-period adjustments. If the UK/NZ approach is correct, adjustments for adjustment events in later periods can be seen as merely an accounting mechanism, removing the need for amendment of past returns.
In Section 4.3, it was suggested that the disconnection between liability and attribution could lead to attribution disappearing into “time warps”. One source of such time warps is the GST grouping regime. The taxpayers in the group entry and exit schemes made their taxable supplies while they were not grouped, but the attribution triggers for those supplies occurred while they were grouped. During the period when the taxpayers were grouped, they would not have been required to submit GST returns unless they were the representative members of their group.[106] Two key issues arise:
(i) By whom is the GST payable – the taxpayer or the representative member of the group?
(ii) In which tax period is it payable – the period of the prepayment, the period immediately before or after the taxpayer leaves the group, the period in which the supplies are made, or some other period?
Division 48 makes no mention of attribution. Rather, it provides that the GST payable by a member of a group is payable by the representative member.[107] But what exactly does this mean? When is an amount payable on a taxable supply made by a member of a group, such that it becomes payable by the representative member? At the time of supply, when the liability first arises? At the time when the amount payable would normally become attributable? The transfer of responsibility for payment from a GST group member to the representative member is time-dependent, but not in the same way that time is relevant in determining whether a supply is an intra-group supply.
The word payable is used inconsistently in the GST Act. In some places it refers to the quality of an amount owed or due to be paid, in others it is used to refer to the quality of responsibility or accountability to pay the amount. An amount of tax arises on a particular supply; its “payability” cannot be shifted to other supplies. In contrast, the responsibility for paying can be shifted from one taxpayer to another, as it is in the grouping rules.
The grouping provisions provide that the GST payable on a taxable supply that a member of a GST group makes is payable by the representative member.[108] In this context, it appears that Div 48 uses payable to mean attributable. The question arises whether this only applies to tax attributable on supplies made while the members are grouped, or to any tax attributable during the period in which they are grouped.
Division 48 makes no specific provision for:
• payment/invoice while grouped, for supplies to be made when not grouped;
• payment/invoice before grouped, for supplies to be made when grouped;
• supplies made when grouped, which are paid/invoiced once ungrouped;
• supplies made before grouped, which are paid/invoiced once grouped; and
• the same scenarios for acquisitions.[109]
If the responsibility for payment is only shifted to the representative member for supplies made while the entities are grouped, the attribution in these circumstances will indeed disappear into a time warp. Where the attribution trigger occurs while grouped, for a supply made while not grouped, the taxpayer is not required to submit returns (unless it is the representative member), so how and when would the tax be paid? It is true that the Commissioner could always require an additional return to be furnished, but to do so he must aware of the situation.[110] Where the attribution trigger occurs while not grouped, for a supply made while grouped, there is no mechanism for determining who should pay.
It is submitted that this is not the correct approach. The Act simply presumes that the attribution and adjustment provisions will effectively deal with these issues. This can only be true if it is accepted that for attribution purposes, Div 48 transfers the responsibility for payment to the representative member if the GST is payable/attributable while the entities are grouped, irrespective of when the supply is made. This is consistent with the Explanatory Memorandum which gives examples of an entity joining and leaving a GST group, and indicates that Div 48 will transfer responsibility for payment to the representative member only to the extent that the GST or input tax credits are attributable to the time when the entities are grouped.[111]
If, as in the UK group exit schemes, the taxpayer knew from the outset that the supplies would be made when it was no longer a member of the GST group, it follows that it knew the supply would be taxable. Thus, the GST would be payable at the time of the prepayment.
In other group exit situations where avoidance schemes are not involved and there is an initial intention to make the supplies while grouped, the supply should initially be treated as not taxable at the time of prepayment. When it becomes evident that the supplies will be made outside the group, an adjustment event will occur (because ceasing to be a member of a group is an event that causes the supply to become taxable). An increasing adjustment will arise (for both the supplier and recipient) on or before the last day they are grouped because this is when they will become aware of the adjustment.[112] The adjustment is not one to which s 48-110 applies, because it does not relate to a supply made while the supplier was a member of the GST group and it does not arise after it ceased to be grouped.
This is consistent with the result in Thorn Materials. It is also consistent with the view proposed earlier, that the adjustment provisions allow changes to be made in the period when the supplier becomes aware of them, even if the adjustment changes the character of the supply ab initio.
4.4.2 Svenska: Attributing GST for a Group Entry Scheme
The theoretical issues are the same for group entry schemes. The supplies took place before the attribution was triggered. Thus, liability and responsibility for payment are certain before the deferred payment is made.
The specific facts of the Svenska scheme would have made the solution simple: the taxpayer was the representative member of the group. Thus, there was no need to shift the payability, and the GST would simply be payable by the taxpayer in the period of the deferred payment. If, on the other hand, the taxpayer was not the representative member, s 48-40(1) would have the effect of shifting responsibility for paying to the representative member.[113]
What, in the end, does it really mean to say that Australian GST has no time of supply rules? It has been shown that once the role of time of supply rules in other jurisdictions is understood, it is clear that we do have such rules, but we simply call them attribution rules. The significance placed on the absence of rules that go by the name of “time of supply” in Australia stems from a misconception of the role of such rules in other jurisdictions. The misconception is that time of supply rules create supplies, whereas it appears from the case law that in most cases they do not.
In the final analysis, the similarities between regimes are more significant than the differences. In the United Kingdom, New Zealand, and Australia:
• time of supply or attribution rules are simply mechanisms for dealing with when GST is payable (or input tax credits are creditable);
• no GST or VAT is payable unless there is actually a supply made in real time in the real world;
• this position can be changed by a provision that actually deems a supply to take place (such as the time of supply rules for continuous supplies in the UK) but it is questionable whether even then they would be treated as creating supplies if no underlying supply ever took place;
• whether a supply has happened in real time in the real world depends on the character of the thing being supplied;
• other jurisdictions sometimes have rules identifying a critical aspect of a supply that must take place before that type of supply can be said with certainty to have been made;
• these taxable event rules invariably relate to tangible supplies such as supplies of goods and/or real property;
• Australia, unfortunately, has no such rules and must determine whether a supply has taken place on a facts and circumstances basis;
• however, in this, the other two jurisdictions are in the same position as Australia when dealing with supplies other than goods, because the exact time at which the taxable event occurs is always more difficult to identify for services.
As it turns out, one advantage of not calling its attribution rules time of supply rules is that Australian GST deals more effectively with group exit and entry schemes, because it requires a focus on the transactions taking place in the real world, and avoids the need for the circumlocutions required in cases like BUPA Purchasing. Nonetheless, every silver lining has a cloud, and Australia’s reliance solely on time of invoice or payment may turn out to have its own difficulties. For example, the absence of any special rules to require attribution for supplies between related parties leaves open the possibility of attribution being deferred indefinitely, whether or not the entities are grouped. If supplies are made “for consideration”, but the supplier simply never issues an invoice and the recipient never makes a payment, no attribution trigger will occur.[114] This is the reason why New Zealand, whose “attribution time of supply” rule is essentially the same as Australia’s attribution rules, has a special attribution time of supply rule for supplies to associates. The rule requires attribution on the basis of real time proxies (delivery or making available for goods, and performance for services) if these would require attribution before the normal time of supply rules based on payment or invoice.[115] This special attribution rule for associates is the only significant difference between Australia’s attribution rules and New Zealand’s time of supply rules. In time, it is likely that Australia will encounter the clouds behind its attribution rules, and will respond by enacting a range a special “time of attribution” rules to deal with this and other situations, just as other regimes have special time of supply rules.
It is perhaps less likely that any amendments will be made to clarify when the real time of supply takes place. Perhaps it would be unwise to attempt a full exposition of the time at which different types of supply take place. However, to dampen the enthusiasm of the contractual rights apologists, provide guidance to the Courts on when transactions are intended to attract the tax, and improve certainty in the application of the Act, a series of “real” time of supply rules (as opposed to “attribution” time of supply rules) might be desirable. The function of these rules would effectively be to define the taxable events in Australian GST – to specify what it is that constitutes a supply by specifying when a supply takes place. This would counter any tendency to characterise all supplies as “an entry into an obligation” or “a creation of rights”.
The following suggested “actual” time of supply rules use the standard real time proxies used in other jurisdictions, both because these proxies focus on performance of the supply, and because they are consistent with the “connection with Australia” provisions in s 9-25.
(1) Section 10-10 applies where, for the purposes of this Act, it is necessary to determine the time at which a supply or acquisition takes place.
(2) For the avoidance of doubt, if a supply is made at a particular time, the acquisition of the thing supplied is made at the same time.
This provision ensures that there are no “time warps” into which supplies can fall before they become acquisitions.
Supplies of goods wholly within Australia or from Australia
(1) Subject to subsection (2), a supply of goods is made when the goods are delivered or made available to the recipient of the supply.
Supplies of goods to Australia
(2) A supply of goods that involves the goods being brought to Australia is made:
(a) if the supplier imports the goods, when the goods are delivered or made available, in Australia, to the recipient of the supply; or
(b) if the supplier installs or assembles the goods in Australia, when the installation or assembly is done.
Supplies of anything of than goods
(3) A supply of anything other than goods is made when the thing supplied is done.
Supplies made at more than one time
(4) If a supply is made at more than one time, it is made at each time to the extent that it is made at that time.
Consideration would need to be given to whether it is more appropriate to include real property in the goods rules or the services rules. Other jurisdictions do both, depending on whether the particular real property is being treated as goods (generally the default position for sales) or services (for lesser interests, though there is no consistency of practice between treating these as goods or services). The “connection with Australia” rules refer only to the location of the underlying land, making no reference to anything for which time is relevant. The transitional rules more closely approximate the goods rules, providing that a supply of real property occurs when the property is “made available” to the recipient of the supply.[116] The choice has been made here to opt for the “services” treatment, because Australia lumps together all types of supply of real property, making no distinction between leases, licences, sales, etc.[117]
Subsection 10-10(4) is intended to prevent suppliers constructing schemes to avoid GST by arguing that their contract involved only one supply of goods, which was made at the time when they first delivered a small proportion of the goods, despite the fact that they then deferred delivery of the remaining goods for some tax-advantageous purpose (such as a group exit scheme).
(1) This section applies to a progressive or periodic supply whether or not Div 156 applies to that supply.
(2) Where a supply is progressive or periodic and Div 156 applies to the supply, section 10-10 applies to each component of the supply as if it were a separate supply identified in the same way that the separate supplies are identified under section 156-5.
(3) If a component of a supply that is progressive or periodic is treated as a separate supply under subsection (2), that component may itself still be progressive or periodic.
(4) Where a supply, or a component of a supply that is treated as a separate supply under subsection (2), is progressive or periodic, the supply is made at a particular time to the extent that it is actually made at that time.
(5) A supply that is made for a period, including a supply to which section 156-27 relates, is treated as being made continuously throughout the period.
(6) For the avoidance of doubt, this section applies to a supply that is made for no consideration.
It is more likely than not that this section changes the current position under the GST law. As the law is presently drafted, Div 156 only has effect for attribution purposes.[118] This creates problems for leases, licences, and other supplies that are granted at the outset, but which cover a period that spans times for which different GST treatments would apply (such as on entering or leaving a GST group). It is submitted that the proposal above is the only practical solution to these issues. This will have no impact on the nature of GST as a tax on expenditures at the time they are incurred, because the rule does not alter the time of attribution, but merely requires the tax status of the supply to be apportioned if it is clear that part of the “supply” will attract a different treatment. However, the implications for leases of goods from outside Australia would need to be considered.
(1) Where a supply is made under an agreement with an entity but the supply is provided, or the agreement requires it to be provided, to another entity, the time of supply is the time when the supply is provided to that other entity.
This is aimed at those sections which specifically require the tax status of a supply to be determined by reference to the provision of the thing supplied to an entity that may not be the recipient of the supply.[119]
(1) Nothing in this Division deems a supply to take place if the supply does not in fact take place.
(2) Where an adjustment event has the effect of cancelling a supply after the time when the supply would have been treated as taking place under this Division, this Division applies as if the adjustment event had taken place before that time.
This is intended to prevent the rules being used to abuse the GST regime, and clarifies that these “time of supply” rules are real time proxies, but do not create supplies where no supply exists.
As indicated, these real time of supply rules effectively define the “taxable event” concept for Australian GST and are not intended to be “time of supply” rules in the sense that phrase is used in other jurisdictions. They would remain independent of the existing attribution rules, which would continue to be used to allocate payment responsibility to particular periods in the same way that time of supply rules do in other jurisdictions. As indicated earlier, a special attribution rule for supplies between associates would be desirable to require attribution at the proposed Div 10 time of supply, if that occurs before the Subdiv 29-A attribution triggers occurs.[120]
Ultimately, this is merely a wish list for Australian “real time” of supply rules. At this stage, it is highly unlikely that such provisions will be enacted, because there has been too little time for the development, implementation, and discovery of schemes taking advantage of timing issues. No doubt if such schemes do come to light, the Commissioner will issue rulings with views similar to those above, and will challenge the schemes in Court, where similar principles may emerge through judicial decisions in due course. What would be undesirable would be for the Commissioner to challenge such schemes on the basis of the anti-avoidance provisions in Div 165, without also pressing for coherent and practical changes to the law.
[*]Senior Lecturer, Faculty of Law, University of Sydney. I am grateful to Richard Vann (Challis Professor of Law, University of Sydney) for his feedback on this article, and also to Rick Krever (Professor of Law, Deakin University) and Stephen Bladwell (Partner, PKF Sydney) for their feedback on an earlier version, which was presented at the ATAX 16th Annual GST and Indirect Tax Weekend Workshop, Noosa, Queensland, 29 April-1 May 2004. Thanks also to the anonymous referee for his/her useful and incisive feedback. Any errors herein are nonetheless entirely of my own making.
[1] Ecclesiastes 3:1, King James Version (or, for some of us, The Byrds).
[2] References to GST law in this article are to the principal Act, the A New Tax System (Goods and Services Tax) Act 1999 (Cth) (“GST Act”). All references hereafter are to that Act, unless otherwise stated.
[3]One author in particular has made this point on many occasions, see P Hill, “GST and Employees: As a Matter of Principle the FBT Gross-up Factor is Not Justified” [2000] 21 GST Today 1, 2, where the author refers to “the total absence of a time of supply rule” as a fundamental principle “enshrined in our GST law”. These views are reiterated in P Hill, “GST and Employees: What a Mess!” [2000] 39 Weekly Tax Bulletin para 1626; P Hill, “Conveying GST” [1999] 13 GST Today 12, 14; P Hill, “GST Rulings – Facts & Fallacies” (2001) 1 Australian GST Journal 21, 28; and P Hill, “Progressive Supplies” [2000] 22 GST Today 8, 11. Other authors have expressed similar views, see P Stacey, “Rent – A GST Musical” [1999] 5 GST Today 1, 1; T Kyle, “Consumptive Disorder: GST and International Supplies of Things Other Than Goods and Real Property” (2002) 17 Australian Tax Forum 115, 147; and A Sommer, “Commentary on Division 11” in GST Commentary (2004).
[4] GST Act, s 9-25(5)(a).
[5] [1965] HCA 58; (1965) 114 CLR 314 (“Arthur Murray”).
[6] Explanatory Memorandum to the A New Tax System (Goods and Services Tax) Bill 1998 (Cth) (“EM”) states: “GST is effectively a tax on final private consumption in Australia”.
[7] S Cnossen, “VAT Treatment of Immovable Property” in Thuronyi (ed), Tax Law Design and Drafting (1996) 231, 232.
[8] Ibid.
[9] Ibid 234.
[10] Some might concatenate options two and three; others choose one or other of them depending on the desired outcome (particularly when applying s 38-190(3)).
[11] A concept recognised in the GST Act itself – see, eg, s 29-25(2)(f) and (g), s 79-5(1)(a), s 38-190, and s 38-325.
[12] GSTR 2001/6, para 80.
[13] The concept of the “true character” of a supply appeared in GSTR 2000/11, paras 31 and 81 and GSTR 2002/6, paras 71, 82, 83, and 89, then morphed into “essential character” in GSTR 2001/8, paras 19, 74, and 81, GSTR 2003/2, para 40, GSTR 2003/7, paras 15, 98, 104, 107, and 108, GSTR 2003/8, paras 20 and 43 (where “true character” also appears at para 50), and GSTR 2003/D7, paras 92 and 258. “True character” has made a recent return to favour in GSTR 2003/11, paras 25 and 120, GSTR 2003/16, para 107, GSTR 2000/19A, para 29, and GSTR 2004/1, para 101.
[14] Excise Tax Act 1985 (Can), s 133 (headed “Agreement as supply”) states: “where an agreement is entered into to provide property or a service (a) the entering into of the agreement shall be deemed to be a supply of the property or service made at the time the agreement is entered into; and (b) the provision, if any, of property or a service under the agreement shall be deemed to be part of the supply referred to in paragraph (a) and not a separate supply.”
[15] Customs and Excise Commissioners v Redrow Group plc [1999] UKHL 4; [1999] 1 WLR 408, [1999] 2 All ER 13, [1999] STC 161 (“Redrow”). Such views focus on those parts of the judgment where Lord Millet characterises the supply as one of contractual rights, rather than on other parts where he characterises the supply to Redrow as the performance of the services under the contract. For examples of reliance on Redrow for characterisation, see: P Stacey and C James, “The GST Treatment of Tripartite Arrangements” (2002) 31 Australian Tax Review 192; P Stacey and C James, “The Limits of Supply: Part Two” (2002) 2 Australian GST Journal 61; and M Cridland, “Supplies ‘Provided to’ Entities in Australia. When is a Supply of Services ‘Made to’ a Non-resident Considered to be ‘Provided to’ an Entity in Australia?” (2002) 2 Australian GST Journal 149. But compare this with P Stacey, “Input Tax Credits Lost and Gained – Redrow Re-interpreted and Fuel Cards” (2004) 8 Tax Specialist 17.
[16] Redrow [1999] STC 161, 171. This passage, which appears in Lord Millet’s extensive discussion under the heading “Identifying the recipient of the services”, does not appear in the Weekly Law Reports version of the judgment. Lord Millet clarified his views in Customs and Excise Commissioners v Plantiflor Limited [2002] UKHL 33, [2002] STC 1132, para 50 where he discussed his judgment in Redrow and clearly characterised the supply to Redrow as being the performance of the estate agency services, which were simultaneously supplied to the Redrow customer whose home was being sold.
[17] [1998] HCA 33; (1998) 194 CLR 500.
[18] Income Tax Assessment Act 1997 (Cth), ss 108-5 and 104-25 (“ITAA97”).
[19] In ID 2003/90, entitled “Acquisition of CGT Asset – Satisfaction of Rights to Have Asset Transferred to Purchaser”, the Commissioner refers to the “real transaction”, a concept which corresponds with the “true” or “essential” character of the supply concept used in the GST rulings (see note 13 above). The Commissioner also takes a similar approach to compensation payments in both the income tax context (TR 95/35) and the GST context (GSTR 2001/4).
[20] These views have been canvassed previously by the author in the context of vouchers − R Millar, “The Vouchers Problem: an Insoluble Conflict or an Illustration of the Nature of Consideration in the ‘Complex Parallel Universe’ of GST?” (2003) 18 Australian Tax Forum 107.
[21] The supply is a single (composite) supply rather than a multiple (mixed) supply − GSTR 2001/8.
[22] For completeness, paras 9-30(1)(b) and (2)(b) should also be noted. These classify a supply as GST-free or input taxed if it is a supply of a “right to receive” a supply that would be GST-free or input taxed. There is no corresponding provision that the right to receive a supply that would be taxable should itself be treated as taxable, an omission that will prove problematic if the contractual rights approach proves to be the one accepted by the Courts. If all supplies are of contractual rights and take place at the time of contract, then the group exit schemes discussed in this article can proliferate in Australia without even the necessity to make prepayments or issue invoices. All that will be required is for the contract to be entered into, at which point s 48-40(2)(a) will operate to prevent the supply “made” between two group members being treated as taxable. Division 165 would then be the Commissioner’s only recourse.
[23] G Hill, “Some Thoughts on the Principles Applicable to the Interpretation of the GST” [2003] JlATax 1; (2003) 6 Journal of Australian Taxation 1, 18.
[24] Ibid.
[25] The Australian Bureau of Statistics reported 3.5 million entities with an Australian Business Number (ABN) as at 30 June 2001, of which 2.2 million were registered for GST, see 1369.0.55.001 Australian Business Register – Counts of ABNs, accessed at www.abs.gov.au on 9 November 2004. The Australian Business Register showed that the number of active ABNs had increased to 4.67 million by 9 November 2004 (from Statistical Search on www.abr.business.gov.au on that date). If the proportion registered for GST has remained the same, approximately 2.9 million entities would have been registered at that date.
[26]A New Tax System (Goods and Services Tax Transition) Act 1999 (Cth), ss 6, 12, 14, 15 and 24.
[27] GST Act, s 9-10(1).
[28] Division 72 does not use time of supply for attribution where there is inadequate consideration, presumably because the normal attribution rules would apply, albeit using the Div 72 GST-inclusive market value.
[29] Which is not to say that the attribution events must coincide. Clearly, Subdiv 29-A makes no link between attribution of the supplier’s output tax and of the recipient’s input tax credit, a fact that lay the seeds for the New Zealand GST avoidance scheme discussed in G Hill, “Scheme New Zealand or An Example of The Operation of Div 165” [2003] eJlTaxR 8; (2003) 1(2) eJournal of Tax Research 147.
[30] GSTR 2001/4, para 23.
[31] Emphasis added.
[32] In the UK, the starting point is to determine whether a supply is a supply of goods, since anything that is not a supply of goods is a supply of services – VAT Act,
s 5(2)(b). The concept of continuous supplies is analogous to the Australian progressive or periodic supplies concept.
[33] Schedule 4 of the Act specifies particular things that are to be characterised as a supply of goods, the two most fundamental being “any transfer of the whole property in goods” (VAT Act, Sch 4, para 1) and “the grant, assignment or surrender of a major interest in land” (VAT Act, Sch 4, para 4).
[34] In Commissioners of Customs and Excise v Svenska International plc [1999] STC 406, the judgments are peppered with the phrase “actual supply”, referring to the real time performance of services, in contradistinction to the “deemed supplies” with which the House of Lords was dealing. Similarly, in Commissioners of Customs and Excise v Thorn Materials Supply Ltd and Thorn Resources Ltd [1998] STC 725, 731 Lord Nolan noted that the transfer of ownership in the goods, which is the subject matter of a supply of goods, normally takes place at the time of delivery. The first rule for time of supply is thus essentially a proxy for actual time of supply.
[35] VAT Act, s 6(2). The section also provides that if the goods are removed before it is certain a supply will be made, the time of supply is the earlier of 12 months after removal or the time when the supply becomes certain.
[36] VAT Act, s 6(3).
[37] GSTR 2000/31, para 64.
[38] VAT Act, s 6(4).
[39] VAT Act, s 6(5).
[40] VAT Act, s 6(14). The details are in The Value Added Tax Regulations 1995 (UK) SI 1995/2518, reg 90. From 1 October 2003, reg 94B, which was introduced in response to the grouping schemes discussed in this article, forces a periodic taxing point for connected or group “undertakings” unless the supplier can show that the recipient was entitled to full input tax credits.
[41] GST Act, s 29-5. To avoid complicating the discussion, cash basis accounting is largely ignored in this article. Nonetheless, it should be clear that a liability to pay which is proportional to payments received is based on the same detachment from the taxable event (the making of the supply) that occurs for accruals based accounting. Cash accounters are not immune from the problems discussed.
[42] GST Act, s 156-5.
[43] Of course, commercial practices probably see the rules operate in an essentially similar manner in many cases, since there is generally a close coupling between the time of supply and the time of payment or invoice for many transactions.
[44] VAT Act, s 43. The Commissioners can also, to the extent that the earlier invoice rule has not triggered a time for part of the supply, agree that the time of supply is the beginning or end of a tax period.
[45] Under VAT Act, s 43A companies can be grouped if they are under common control, a concept defined by reference to, amongst other things, holding a majority of the voting rights or the rights to appoint directors. Section 43AA was introduced by the Finance Act 2004 in an attempt to limit abuses of this wide entitlement (see HM Customs and Excise, Tackling VAT Group Abuses (December 2003)). The new section gives the Treasury wide powers to modify the operation of s 43A, in particular to narrow eligibility for grouping by reference to the accounting standards for consolidated group financial statements. In contrast, Australia has a narrower 90% common ownership requirement but a broader range of entities that can group, so long as they meet the requirements set out in Div 48 of the GST Act 1999 and its associated regulations. There is no restriction on grouping of non-residents under Australian GST.
[46] VAT Act, s 43.
[47] ITAA97, s 701-1.
[48] GST Act, s 48-10(1)(c).
[49] GST Act, s 48-45(2).
[50] GST Act, s 48-55.
[51] GST Act, ss 48-40, 48-45, 48-50 and 48-60.
[52] VAT Act, s 43(1)(a).
[53] GST Act, ss 48-40(2)(a) and 48-45(3). There are exceptions for reverse charged “imported services” and, in relation to supplies, for certain joint venture transactions.
[54] VAT Act, s 24.
[55] VAT Act, s 26.
[56] Value Added Tax Regulations 1995 (UK), regs 108 and 110. Regulation 109 requires a similar adjustment to be made where the input tax is initially attributed to exempt supplies, and later used or intended to be used for taxable supplies.
[57] GST Act, s 11-15(1).
[58] GST Act, s 11-15(2).
[59] GST Act, Div 129.
[60] GST Act, s 129-20. Under s 129-55 the concept of “apply” includes the supply, consumption, disposal, or destruction of the acquisition. The Commissioner has taken the view that simply holding an asset for a particular purpose does not amount to an application of the asset – see para 80 of GSTR 2003/6 (on matrimonial property distributions) and the more extensive discussion in the Property and Construction Industry Partnership Issues Register Section 04 on www.ato.gov.au. If this is correct, a change of intention for an acquisition that has not yet been used will neither change an initial input tax credit entitlement, nor require an annual adjustment to be made. Whether the Commissioner is correct is not significant for the purposes of this article.
[61] [1998] STC 725 (HL) (“Thorn Materials”).
[62] [1999] STC 406 (HL) (“Svenska”).
[63] [2003] STC 1203 (“Bupa Purchasing”).
[64] Manipulation of timing is the key aspect of the schemes, and although the interposed entity is necessary to the scheme, it is not critical to be able to argue that its role is in any way artificial.
[65] [1998] STC 725, 729.
[66] The prepayment was immediately loaned back to E for three years at 5.77% interest, on terms that the loan was repayable at will or could be set off against liabilities to the taxpayer, which would include a 5% markup on the taxpayer’s cost of eventually acquiring the cars. While not a factor in the judgment, this clearly didn’t assist the “flavour” of the transaction, about which Lord Nolan noted “the appellants have not suggested for a moment the these transactions were designed for any commercial purpose, or indeed for any purpose other than the avoidance of VAT” – [1998] STC 725, 729.
[67] Slightly less due to the 5% markup on the intra-group transaction.
[68] The list is a précis of the main points in the judgment of Lord Nolan, with which Lord Browne-Wilkinson and Lord Lloyd of Berwick agreed.
[69] [1998] STC 725, 733 (per Lord Nolan).
[70] Sixth Council Directive of 17 May 1977 on the harmonisation of the laws of the Member States in relation to turnover taxes – common system of value added tax: uniform basis of assessment, 77/388/EEC [1977] OJL 145/1 (“Sixth Directive”).
[71] [1998] STC 725, 741 (per Lord Clyde).
[72] [1999] STC 406, 409 (per Lord Lloyd of Berwick).
[73] Ibid 414.
[74] Ibid 423 (per Lord Hutton).
[75] Ibid 416.
[76] Boots Co plc v Customs and Excise Commissioners (Case C-126/88) [1990] ECR I-1235; [1990] STC 387, ECJ; Elida Gibbs Ltd v Customs and Excise Commissioners (Case C-317/94) [1996] EUECJ C-317/94; [1997] QB 499; [1996] ECR I-5339; [1997] All ER (EC) 53; [1996] STC 1387, ECJ.
[77] Freemans plc v Customs and Excise Commissioners (Case C-86/99) [2001] 1 WLR 1713; [2001]STC 960, ECJ.
[78] Argos Distributors Ltd v Customs and Excise Commissioners (Case C-288/94) [1996] EUECJ C-288/94; [1997] QB 499; [1996] ECR I-5311; [1996] STC 1359, ECJ.
[79] [2003] STC 1203, para 15 (per Park J).
[80] This result can be gleaned from the following passages in the judgment in BUPA Purchasing: the average rate is identified at para 9 and the adverse consequences of the Court’s decision are discussed at para 52. The overall recovery would have been less than 2% because not only did the taxpayer lose 98% of its credits, but the hospital lost 5 – 10% of the credits on the tax charged to it on the deemed 2% supply.
[81] Section 9 of the Goods and Services Tax Act 1985 (“NZ GST Act”) relies on time of invoice or payment to set a deemed time of supply. Note, however, that for associated person transactions, the New Zealand law does specify a time of supply using the standard indicators of removal or making available of goods or the performance of services – NZ GST Act, s 9(2)(a). This rule prevents deemed time of supply rules being used to gain a GST advantage in group entry and exit schemes.
[82] The Commissioner has noted that s 9-10(2) identifies two aspects of supplies: the thing being supplied and the means by which it is supplied – GSTR 2001/4, para 25. As s 9-10(1) notes, the list is not comprehensive. Indeed, s 9-10(2) makes no mention of one of the key means for making a supply – a “sale”.
[83] T Kyle, “Consumptive Disorder: GST and International Supplies of Things Other Than Goods and Real Property” (2002) 17 Australian Tax Forum 115, 147.
[84] This author only partially agrees, however, with Kyle’s conclusion, that the absence of “a readily identifiable time of supply is a defect” in Australian GST: ibid. As stated, in this regard, Australia’s GST is not out of line with other jurisdictions and while at the end of this paper a range of possible “taxable event time of supply” rules are suggested, they are drafted to leave open the full range of possible times of performance that might be appropriate for particular transactions.
[85] Sections 9-75 and 9-15. The particular facts of Thorn Materials would not have given rise to an input tax credit blockage in Australia because there are no exactly equivalent provisions denying input tax credits for cars.
[86] Emphasis added. The interaction of time of supply with Div 156 creates problems for supplies such as real property leases, where it could be argued that a single supply takes place at the time when the interest in land is created. Thorn Materials and Svenska were referred to in Royal and Sun Alliance Insurance Group plc v Customs and Excise Commissioners [2003] UKHL 29, para 36 where Lord Hoffman said, in discussing the application of the continuous supplies rule to leases: “I find the notion of the same goods or services being supplied over and over again too hard to grasp. In my opinion the plain effect of the regulations is to treat each successive supply as different from the one before”. Despite Lord Hoffman’s protestations, this surely only confirmed Sedley LJ’s view of VAT as a world “in which factual and legal realities are suspended or inverted” – Royal and Sun Alliance Insurance Group plc v Customs and Excise Commissioners [2001] STC 1476, para 54. Query whether such a practical analysis is possible in Australia. Section 156-22 allows Div 156 to apportion the attribution for leases, but the initial liability might still be determined up front when the lease is granted.
[87] Theodor Seuss Geisel, aka Dr Seuss, 1904-1991, cited from the Online Dictionary of Quotations at http://www.quotationreference.com. Despite extensive efforts, the original citation for this poem was not located, but it was too delightful, and too apposite to the point being made, to leave it out.
[88] GST Act, s 29-10.
[89] Example 2 in para 4.34 and the examples in para 4.40 of the EM (above n 6) clearly anticipate the possibility of attribution before a supply or acquisition is made.
[90] [1965] HCA 58; (1965) 114 CLR 314.
[91] Thorn Materials [1998] STC 725, 731.
[92] Ibid 732.
[93] Case W22 (2003) 21 NZTC 11,212 (Decision No 019/2003).
[94] The “stakeholder” clause (requiring the deposit to be held by a third party on behalf of the purchaser) was struck out from the standard form contract to ensure the payment would be treated as a deposit received by the supplier and so would trigger the time of supply.
[95] Case W22 (2003) 21 NZTC 11,212, paras 110-140.
[96] NZ GST Act, s 25.
[97] (2003) 21 NZTC 11,212, para 127 [Emphasis added].
[98] Case W11 (2003) 21 NZTC 11,100 (Decision No 005/2003); and Case W9 [2003] NZTRA 3; (2003) 21 NZTC 11,083 (Decision No 003/2003).
[99] Case W11 (2003) 21 NZTC 11,100, para 19.
[100] Ibid para 74.
[101] CIR v Capital Enterprises Ltd (2002) 20 NZTC 17,511, para 49 (per Durie J) [Emphasis added]. In support of this statement, his honour cited the president of the Court of Appeal in Pine v CIR (1998) 18 NZTC 13,750 (CA), 13,573, where P Richardson stated: “section 9(1) in defining the time of supply, which is important in deciding when tax becomes payable or deductible, does not determine the character of the supply that is being made”.
[102] Ch’elle Properties (NZ) Limited v Commissioner of Inland Revenue (2004) 21 NZTC 18,618.
[103] Ibid paras 58-59. From the final sentence of the passage cited it is clear that his honour did not fully engage with the TRA’s conclusion that the initial input tax credit entitlement had been undone by the cancellation, so that s 25 would operate merely as a mechanical provision for repaying the input tax credit to which the taxpayer was no longer entitled. It is also odd that the High Court, and to a lesser extent the TRA, decided the case on the basis of the general anti-avoidance rule in NZ GST Act, s 76, despite having found that there was no input tax credit entitlement (and so no tax benefit to be undone).
[104] Note that NZ GST Act, s 9(3)(aa) also contains a continuous supply rule in similar terms to the UK rule.
[105] GSTR 2000/19, para 15.
[106] GST Act, s 48-60.
[107] Similarly, the input tax credit entitlements of a member, and any adjustments the member has, are entitlements or adjustments of the representative member – ss 48-40(1), 48-45(1) and 48-50(1).
[108] GST Act, s 48-40(1).
[109] These same issues arise in respect of commencement and cessation of registration, despite the existence of some specific provisions to deal with cessation.
[110] The power to do so appears in s 31-20. Section 48-60 is expressed to operate despite s 31-5, but makes no reference to s 31-20, so the Commissioner is not precluded from requiring a return to be furnished.
[111] EM, above n 6, paras 6.22 and 6.23. This analysis is not entirely without difficulties. It presupposes that the reference in s 48-40(1) to “GST payable on any taxable supply … that a member of a GST group makes” is different from the reference in s 48-40(2) to “a supply that an entity makes to another member of the same GST group”. Perhaps it could be argued that there are inherent differences between the two provisions. Section 48-40(2) requires a reference to the time of supply, because if it did not, any supply made between entities who have ever been, or ever become, members of the same GST group would be treated as if it were not a taxable supply. In contrast, s 48-40(1) is not a relieving provision. It is merely a provision shifting payment responsibility while entities are members of the same group.
[112] GST Act, s 29-20(1) and Div 19.
[113] As per para 6.22 of the EM, above n 6.
[114] See, eg, the discussions in GSTD 2004/D3 on book entries that do not involve the netting off of mutual liabilities.
[115] NZ GST Act, s 9(2). There is also a range of special attribution time of supply rules dealing with situations similar to the list in Australia’s s 29-25.
[116] GST Transition Act 1999, s 6(3).
[117] See s 38-187.
[118] See note 67 above.
[119] See, eg, s 38-190(3) and (4), and also s 87-5(1)(b).
[120] This would, of course, affect the timing of GST and input tax credits for companies engaging in Svenska type group entry schemes.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/JlATax/2004/6.html