27^TH ATAX GST CONFERENCE, BRISBANE

CHRIS SIEVERS, Lonsdale Chambers

INTRODUCTION

The concept of apportionment is an inherent part of the GST Act.[1] The signpost for apportionment is the phrase “to the extent that”. More than 65 years ago the High Court observed that these words indicated that an apportionment was envisaged by the legislature.[2]

The words “to the extent that” appear in a number of places, including the following:

• Chapter 3 (the Exemptions):
  • Section 40-35 dealing with residential rent.
  • Section 40-65 dealing with the sale of residential premises.
• Chapter 2 (the Basic Rules):
  • Section 9-5 dealing with what is a taxable supply.
  • Section 9-30 dealing with supplies that are both GST-free and input taxed.
  • Section 11-15 dealing with acquisitions made for a creditable purpose.
  • Section 11-30 dealing with the entitlement to input tax credits where acquisitions are made partly for a creditable purpose.
• Chapter 4 (the Special Rules):
  • Section 129 dealing with the application of a thing for a creditable purpose.

Apportionment may also arise where the words “to the extent that” are not present. An example is s 9-80 which deals with the value of taxable supplies that are partly GST-free or input taxed. The concept of apportionment is found in the requirement that the value is “the taxable proportion of the value of the actual supply that the taxable supply represents” (emphasis added).

Identifying the parts of the GST Act where apportionment may be available is a relatively straightforward task. The difficulty is in determining whether apportionment is available given the particular facts and circumstances and, if so, how the apportionment is to be determined. No guidance is found in the legislation, although it does appear to be generally accepted that any apportionment must be “fair and reasonable”. The Commissioner has sought to assist taxpayers by publishing a number of rulings on the question of apportionment.[3] However, a review of the case history in the United Kingdom VAT regime shows that disputes between taxpayers and the Commissioner are likely to arise.

The areas where apportionment has caused the greatest difficulty for the Commissioner and taxpayers appear to be the following:

• Determining the extent of creditable purpose under Division 11 and adjustments to creditable purpose under Division 129; and
• Determining the “taxable value” of supplies which are both taxable and GST-free and/or input taxed.

The driver for dispute is that the apportionment methodologies applied have a direct revenue impact for taxpayers (and the Commissioner), through the entitlement of taxpayers to receive input tax credits and the liability of taxpayers to pay GST. Each of these areas is discussed in this paper.

I have also conducted a review of reported decisions in the United Kingdom VAT regime in recent years. While it must be remembered that the UK VAT regime is different, I consider that these decisions illustrate some of the issues that are likely to arise in Australia when dealing with apportionment.

Determining the extent of creditable purpose

Apportionment can play an important role in determining the extent of the taxpayer’s “creditable purpose” and the ability of taxpayers to recover the GST cost borne on inputs, or at least a proportion of those costs.

Division 11

The entitlement of an entity to input tax credits is addressed in Division 11 and the centerpiece is the concepts of “creditable acquisition” and “creditable purpose”. The elements of a creditable acquisition are set out at s 11-5 and the first element is that you acquire the thing “solely or partly for a creditable purpose”. Section 11-15 sets out the meaning of “creditable purpose” and sections (1) and (2) are as follows: (emphasis added)

• You acquire a thing for a creditable purpose to the extent that you acquire it in carrying on your enterprise.
• However, you do not acquire the thing for a creditable purpose to the extent that:
  • the acquisition relates to making supplies that would be input taxed; or
  • the acquisition is of a private or domestic nature.

If an entity makes an acquisition for a creditable purpose it is entitled to an input taxed credit equal to the GST payable on the supply of the acquisition: s 11-25.[4] If the acquisition is made only partly for a creditable purpose, the entity is entitled to an input tax credit to the extent of that creditable purpose expressed as a percentage of the total purpose of the acquisition: s 11-30.

The effect of s 11-15(1) and (2) is that there are three points where apportionment may apply:

• The extent that the entity acquires the thing in carrying on its enterprise: 11-15(1);
• The extent that the acquisition relates to making supplies that would be input taxed: 11-15(2)(a); and
• The extent that the acquisition is of a private or domestic nature: 11-15(2)(b).

It is paragraph 11-15(2)(a) that has caused the greatest difficulty.

For various reasons, Parliament determined that not all supplies were to be taxable and that some supplies were to be “input taxed” supplies. In HP Mercantile Pty Limited v Commissioner of Taxation (2005) 143 FCR 553 Hill J[5] made the following observations on the underlying policy behind the concept of an input taxed supply (at [16]-[17]):

16 In terms of GST theory, it is generally accepted that there are certain kinds of activities where the basic system of output tax on supplies and input tax credits on acquisitions will not lead to taxation on the value added by each supplier in the chain. The most important example is said to be financial transactions of financial institutions such as, but not confined to, banks, because they constantly borrow and lend and turn over money in a way that amounts, such as interest charged, will not represent the real value added by the financial institutions. Indeed, as the explanatory memorandum distributed with the bill which, as amended, later became the GST Act (“the EM”) says in Chapter 1 [5.140]: “…there is no readily agreed identifiable value for supplies consumed by customers of financial services”. In such a case, it is the margin or imputed margin that is the real economic subject of the supply. There are other examples where this may be the case, one of which is the leasing of, or other dealings with, residential property (not being new residential property).

17 By way of what may be seen as a compromise for the difficulties of applying the normal system of value added taxation to financial supplies and other difficult cases, value added taxation design has created a form of supply which is referred to in Australia as an input taxed supply but which, in international value added tax parlance, is referred to as an “exempt supply”. An input taxed or exempt supply (and financial supplies made by financial institutions will be the main example) will not, generally speaking, attract output tax, but the entity which makes financial supplies will, likewise, not obtain an input tax credit for the tax payable on acquisitions it makes in the course of its enterprise of making input taxed supplies.

The legislative policy is that no GST is payable on an input taxed supply and also that the supplier cannot claim input tax credits for those acquisitions that relate to the making of the input taxed supply. That policy is reflected in paragraph 11-15(2)(a) which operates as a “blocking” provision”[6]. If an entity acquires a thing in carrying on an enterprise, the effect of paragraph 11-15(2)(a) is that the entity does not make the acquisition for a creditable purpose to the extent that it relates to making supplies that would be input taxed. The statutory text envisages that acquisitions may be partly creditable in certain circumstances, leading to an apportionment of input tax credits – this is addressed by s 11-30 which introduces the concept of “extent of creditable purpose”.

The issue of apportionment may arise where an entity makes both input taxed and non-input taxed supplies (be they taxable or GST-free) in the course of its enterprise. The entitlement of those entities to input tax credits on acquisitions made in the course of their enterprises can be described as follows:

• There is a full entitlement to credits where the acquisition does not relate to making supplies that would be input taxed (acquisitions made solely for a creditable purpose);
• There is no entitlement to credits where the acquisition relates to making supplies that would be input taxed (acquisitions made for a nil creditable purpose); and
• There is a partial entitlement to credits where the acquisition relates to the making of supplies that would be input taxed as well as supplies that are taxable and/or GST-free (acquisitions made for a partial creditable purpose).

This third category was recognised by the majority of the Full Federal Court[7] in Commissioner of Taxation v American Express Wholesale Currency Services Pty Limited (2010) 187 FCR 398 where their Honours adopted the description of the statutory scheme of Hill J in HP Mercantile (extracted at paragraph [10)] above) and made the following observations on apportionment (at [105]):

Typically, where a corporation makes both input taxed and non-input taxed supplies, acquisitions in the nature of general expenses and overhead will be only partly creditable: see HP Mercantile at [37] 563. This is because such acquisitions have a real and substantial, though indirect, relationship to the corporation’s activities.

For acquisitions falling into the third category, it will be necessary for taxpayers to determine their “extent of creditable purpose” so that their input tax credit entitlement can be calculated. In the context of the GST Act, which as a general rule is intended to be cost-neutral to business, this is an important calculation because it has a direct cash impact on the taxpayer (and the Commissioner, who has to pay a refund).

Determining the recovery for acquisitions falling into the third category is of particular relevance to the following types of enterprises:

• Financial institutions that make both input taxed financial supplies and taxable supplies;
• Property developers that make both taxable supplies of the sale of new residential premises and input taxed supplies of the lease of residential premises;
• Retirement villages that make both taxable supplies to residents and input taxed supplies of the lease of residential premises.

However, the recent decision of the Federal Court in Rio Tinto illustrates that the issue of apportionment can arise for any taxpayer that makes input taxed supplies in the course of an enterprise which involves the making of taxable or GST-free supplies. This decision is discussed further at paragraph [16)] below.

Apportionment and paragraph 11-15(2)(a)

When is apportionment available?

A threshold issue to be determined is whether apportionment is appropriate for a particular acquisition, ie, whether the acquisition falls into the first, second or third categories referred to at paragraph [12)] above. It is only if the acquisition falls within the third category that apportionment may be available.

This threshold question is illustrated by the recent decision of the Federal Court in Rio Tinto. The taxpayer argued that it was entitled to full input tax credits for acquisitions made in respect of residential housing that was leased to employees and contractors. The taxpayer contended that while the lease of residential housing was an input taxed supply, the acquisitions did not relevantly relate to those supplies but related to the ultimate taxable or GST-free supply of iron ore by the taxpayer. Accordingly, paragraph 11-15(2)(a) did not apply. The alternative argument of the taxpayer was that the acquisitions related to taxable or GST-free supplies and also to the input taxed supply of leasing the housing to employees and there should be an apportionment of the taxpayer’s entitlement to input tax credits. The Court found (at [34]) that there was no need to address the apportionment issue because the acquisitions related wholly to the provision of accommodation. The acquisitions therefore fell within the second category and no credits were available.

The decision in Rio Tinto (which is on appeal) raises the question of what type of expenses may be apportionable. In Ronpibon Tin the High Court (at 59) made the following observation on s 51(1) of the Income Tax Assessment Act 1936:

The charges for management and the directors’ fees are entire sums which probably cannot be dissected. But the provision contained in s. 51 (1), as has been already said, contemplates apportionment. The question what expenditure is incurred in gaining or producing assessable income is reduced to a question of fact when once the legal standard or criterion is ascertained and understood. This is particularly true when the problem is to apportion outgoings which have a double aspect, outgoings that are in part attributable to the gaining of assessable income and in part to some other end or activity. It is perhaps desirable to remark that there are at least two kinds of items of expenditure that require apportionment. One kind consists in undivided items of expenditure in respect of things or services of which distinct and severable parts are devoted to gaining or producing assessable income and distinct and severable parts to some other cause. In such cases it may be possible to divide the expenditure in accordance with the applications which have been made of the things or services. The other kind of apportionable items consists in those involving a single outlay or charge which serves both objects indifferently. Of this directors’ fees may be an example. With the latter kind there must be some fair and reasonable assessment of the extent of the relation of the outlay to assessable income. It is an indiscriminate sum apportionable, but hardly capable of arithmetical or ratable division because it is common to both objects.

The observations of the High Court support the conclusion that, in general, apportionment may be available for acquisitions that relate to “mixed activities” or to undifferentiated general overhead outgoings that cannot be traced to a particular supply. That also appeared to be the view taken by the majority in Amex (see the extract at paragraph [12)] above). However, each case will likely depend on its own particular facts.

What is an appropriate apportionment methodology?

Once it is accepted that apportionment is available for an acquisition or for a group of acquisitions, it is necessary to determine a “fair and reasonable” apportionment methodology. This raises the question of what is “fair and reasonable”?

In GSTR 2006/4 ‘Goods and services tax: determining the extent of creditable purpose for claiming input tax credits and for making adjustments for changes in extent of creditable purpose’, the Commissioner outlines his views on this issue. Paragraphs 32-34 of the Ruling provide as follows: (footnotes excluded)

You may choose your own apportionment method, but the method you choose needs to be fair and reasonable in the circumstances of your enterprise. It needs to appropriately reflect the intended or actual use of your acquisitions or importations.

The ‘fair and reasonable’ principle was used by the High Court in Ronpibon Tin v. FC of T ,¹⁶ in the context of the apportionment of expenditure serving more than one object ‘indifferently’.¹⁷ The High Court did not, in that case, apply this principle in relation to the allocation of specific acquisitions wholly to specific ends, or to apportioning items of expenditure ‘distinct and severable parts of which’ can be identified as being devoted to such specific ends. The Commissioner’s view is that the ‘fair and reasonable’ principle applies equally to the choice of method for allocating or apportioning acquisitions in all circumstances.¹⁸

Following the principles set out by the High Court, the apportionment method you choose needs to:

* be fair and reasonable;

* reflect the planned use of that acquisition (or in the case of an adjustment, the actual use); and

* be appropriately documented in your individual circumstances (see paragraphs 98 to 100 of this Ruling).

The Commissioner therefore leaves it to the taxpayer to choose its own apportionment method, noting the following requirements:

1. the method is fair and reasonable in the circumstances of the taxpayer’s enterprise; and
2. the method appropriately reflects the intended or actual use of the acquisitions.

New Zealand

The position in New Zealand appears to be similar. The website for the Inland Revenue states as follows:[8]

When goods and services are acquired on or after 1 April 2011 a claim for input tax is allowed depending on the extent to which goods or services are:

* used for, or

* are available for use,

in making taxable supplies.

At the time they are acquired, an estimate of the intended use is made by choosing a method that provides a fair and reasonable result. This can be based on:

• previous records
• experience
• business plans, or
• any other suitable method.

The central requirement is that the taxpayer can choose a method as long as it provides a fair and reasonable result. As in Australia, the taxpayer is free to choose the method, save that there is a special rule for the concurrent use of land (land which is used concurrently for taxable purposes (for sale) and for exempt purposes (rental pending sale)): s 21E. The formula is as follows:

consideration for taxable supply   * 100

total consideration for supply

United Kingdom

The position in the United Kingdom is a little different. The regime is described in UK – VAT Notice 706 and appears to operate as follows:

• There are two types of apportionment methodologies (described as “partial exemption methods”):
  • the standard method specified in the law; and
  • a special method devised by the taxpayer to reflect the taxpayer’s unique business circumstances.
• The VAT Notice makes the following observation on partial exemption methods:

A partial exemption method must produce a result which enables you to recover a proportion of input tax which fairly reflects the extent to which the purchases on which it was incurred are used to make taxable supplies (and other supplies with the right to deduct). It should be easy for you to operate and for HMRC to check. Such a method is described as ‘fair and reasonable’.

The standard method is as follows:

Value of taxable supplies in the                                    = Recoverable

period (excluding VAT)                                    * 100      percentage of

Total value of supplies in the period                              residual input tax

(excluding VAT)

The standard method must be used unless the Revenue has given approval for the taxpayer to operate a special method. Also, the application of the standard method is subject to an override which deals with circumstances where the standard method does not produce a fair and reasonable deduction of input tax.

A taxpayer can obtain approval for a special method by writing to the Revenue with details of the proposal. The proposal must include a declaration that the method is fair from its effective date of application, and for the foreseeable future so that from its effective date a fair amount of input tax is recovered. If the Revenue subsequently finds the declaration to be incorrect, it may serve a Special Method Override Notice to override the special method. A declaration is incorrect if two conditions are met:

• the special method does not produce a fair and reasonable attribution of input tax to taxable supplies resulting in an unfair over-recovery of input tax; and
• the person signing the declaration knew or ought reasonably have known this at the time they made the declaration.

The Revenue has the power to direct a taxpayer to use a particular method or to stop using an existing special method. The powers are used in circumstances where the Revenue is unable to identify a mutually satisfactory method, or where the VAT system is being abused.

The position in the UK is that you start with a statutory presumption that the “standard method” will produce a fair and reasonable outcome and it is open to the taxpayer, or the Revenue, to seek to adopt a different method if that is not the case.

Methods of apportionment

In GSTR 2006/4 the Commissioner discusses a number of different apportionment methods, described as direct and indirect methods (paragraphs 108-120). The Commissioner also notes that taxpayers are not required to use any of these methods, provided that whatever alternative method is used is fair and reasonable.

Direct methods – These methods seek to identify a direct measure of the use of the acquisition. Examples are distance travelled by a motor vehicle as evidenced by a log book, floor area where space is used for different activities, and time sheets.

The Commissioner is of the view that these methods will usually best reflect the extent of creditable purpose. However, the Commissioner accepts that a taxpayer could choose not to use a direct method where, for example:

• an indirect method is considered to be more appropriate;
• the cost of measuring the use of the acquisition is disproportionate to the cost of the acquisition; or
• it is impracticable to use a direct method in the context of the particular business.

Indirect methods – These methods use information that is not directly identifiable with the particular acquisition. The methods operate on the assumption that measures of inputs and outputs of an enterprise may provide a basis upon which to estimate the application of certain acquisitions which relate to making both taxable and input taxed supplies. The information operates as a “proxy” that may provide a reasonable basis for calculating an entitlement to an input tax credit entitlement. The Commissioner accepts that indirect methods may be appropriate in circumstances where there are overhead expenses or a large number of small acquisitions and it is not cost effective to try to measure the use of each acquisition.

Indirect methods can be “input based” or “output based”.

Input based methods seek to identify a proxy based on the already established use of other inputs. The Commissioner considers that these methods are only useful where a direct method has already been used in respect of a majority of the acquisitions of the enterprise, so as to provide a reliable basis for their use in calculating the creditable use of the remaining inputs.

Output based methods seek to identify a proxy based on the output of the enterprise. An example of such a method is found in Amex. The basis for the agreed formula was described by the majority of the Full Federal Court as follows (at [126]-[127]):

126 The formula employs revenue as a proxy for the relationship between acquisitions and the making of supplies based on an underlying assumption that there is a roughly proportional relationship between: (a) the relative amount of revenue an entity derives from the making of particular supplies; and (b) the proportion of the total supplies made by entity represented by those particular supplies. Absent such an assumption, the formula is untethered from the text of GST Act. Although not directly stated in the Commissioner’s rulings, it seems clear that the foundation for the assumption is the expectation that a rational profit-driven corporation will, in general, only dedicate its resources (including, relevantly, its potentially creditable acquisitions) to making particular supplies to the extent that doing so maximizes revenue…

127 To put it in terms of the present case, the formula assumes that it is fair and reasonable to expect a roughly proportional relationship between: (a) the amount of revenue Amex Intl earns from making financial supplies relative to its total revenue; and (b) the extent to which Amex Intl’s acquisitions relate to making financial supplies, relative to the acquisitions’ relationship to its supplies overall…

The “standard method” in the UK VAT regime is another example of an indirect output method.

In GSTR 2006/4 the Commissioner also takes the view that the basis of apportionment needs to make sense in the context of an enterprise and should not produce significant distortions. Accordingly, factors which may distort the results should be excluded. Such factors may include extraordinary supplies, income items or acquisitions or substantial one-off capital acquisitions.

A look at some of the cases

There have been a number of cases in Australia where the issue of apportionment has been raised. However, my research has identified only one case where the issue of whether a particular apportionment methodology was appropriate or was fair and reasonable was directly considered. In the other cases the parties have agreed on a particular apportionment methodology and the issue was the proper characterisation of items that were to be included in the formula: for example Amex and Living Choice Australia Ltd and Commissioner of Taxation [2014] AATA 168; and in others the Court did not see the need to address the question of apportionment because of the findings made by the Court: for example AXA Australia Pacific Holdings Limited v Commissioner of Taxation (2008) 173 FCR 500 (see [72]-[74])[9] and Rio Tinto. There have been a number of cases in the United Kingdom addressing the issue of whether an apportionment methodology was fair and reasonable. While recognising that caution must be had in referring to overseas authorities, and that the statutory test in the UK is different – being whether there is a “direct and immediate link” between the acquisitions and the taxable outputs of the enterprise, I consider that the decisions are instructive as to the issues that may arise here.

A Taxpayer and Commissioner of Taxation [2011] AATA 160.

The applicant was a property developer and constructed a complex comprising retail shops, a commercial car park and residential apartments. The intention was to sell the complex on completion. Some months after completion, after the applicant had been unable to secure a sale of the complex, the applicant sold the retail shops and commercial car park but could not sell the residential apartments. The applicant began to lease the apartments and the applicant needed to calculate an increasing adjustment under Division 129.

The applicant initially proposed an apportionment methodology based on the value of the complex attributable to the residential apartments ($115,315,000) divided by the rent received for the period May to 30 June 2008 ($455,487) to give a creditable percentage for that period of 98.4168%. This resulted in an increasing adjustment of $23,271 for the period, being 1.5832% of $1,469,872 (the total input tax credits claimed).

The Commissioner accepted this methodology but the applicant objected on the basis that the apportionment methodology was based on an “effective life” of the residential apartments of 40 years (or 480 months) – being the statutory rate for buildings provided in s 43 of the Income Tax Assessment Act 1997. The non-creditable use of the apartments was therefore to be calculated by reference to 2 out of 480 months.

The Commissioner contended that this apportionment methodology was not fair and reasonable. The Commissioner referred to the his views set out in GSTR 2009/4 as to why this method was not fair and reasonable. The reasons included that the method contemplated the entire life span of the premises rather than the actual use of the premises in the relevant period. Also, the effective life of the building is too remote and arbitrary to reasonably reflect the application of the residential premises, comprising both the building and the land, during the relevant period. The Tribunal agreed, observing that the applicant’s revised percentage appeared to be based on holding the apartments for sale for 40 years, which was an unreasonable contention.

St Helen’s School Northwood Ltd v Revenue & Customs [2006] EWCH 3306

The school constructed a sports complex which was to be used for its own exempt educational purposes during school hours and was to be available for third party use outside those hours. The use outside school hours was to be through a licence granted to a separate company and the licence fee would be subject to VAT. Any profit by the company would be gifted to the school.

The school claimed 54% of the VAT payable on the cost of construction as a credit and applied to use a special which produced that result. The method was based on a proportion of total projected hours of use of the complex by the company as compared to total use. The application was rejected by the Revenue.

The Court observed (at [16]) that the UK legislation envisaged the “fair and reasonable” attribution of input tax on goods and services to taxable supplies to reflect the extent of the use of those goods and services in making taxable supplies.

The school contended that the “standard method” did not produce a fair result. The school contended that the standard method included the whole of the fee income in the denominator, but only the taxable income from the use of the complex in the numerator. The Revenue contended that the special method proposed by the school would allow for over-recovery of credits.

The Court observed that the special method proposed by the school was based on the proposed physical use of the building by the company. The Court (at [50]) considered (contrary to the Tribunal below) that “the physical use of a building or a chattel by a third party may be a very good proxy for the use, for VAT purposes, of the taxable person”. However, the Court (at [75]) agreed with the Revenue that the physical use of the complex is not necessarily a fair and reasonable proxy and that the phrase “economic use” was a helpful approach to establish what the search is for. In finding for the Revenue, the Court made the following observations (at [77]-[80]):

On the facts of the present case, it seems to me that the overwhelming economic use of the sports complex by the School is in relation to the provision of educational services. In that context, I agree with Miss Simor that the source of funds and the purpose of constructing the sports complex are relevant considerations… The question is what “use” is being made of the inputs in producing the outputs. It seems to me that the purpose of the School, objectively ascertained, in constructing the sports complex is a highly relevant factor in attributing cost components between the relevant outputs…

On the evidence, it is clear that, objectively assessed, the principal purpose of the School in building the sports complex was the furtherance of its educational activities and was carried out in connection with its business of making exempt supplies of education. That conclusion is clear from the way the matter was put in the first draft of the business plan and the approach of the School to the generation of funds by out-of-school use which was designed to meet the running costs of the complex and, if possible, something over and above that. Further, the capital cost of the complex was met out of funds which were either charitable funds or derived from a fund-raising exercise and which were clearly dedicated to the educational purposes of the School. The generation of income by out-of-school use was essentially a secondary consideration, albeit that the benefit thereby produced was an aspect of the whole project from the beginning.

Moreover, it is also clear, I consider, that the income generated by the licence to SHEL was never intended or expected to meet a share of the capital cost proportionate to the physical use of the sports complex by SHEL. The relevance of this is that it is supportive of the view that the principal, objective, purpose of the expenditure on the sports complex was in furtherance of the School’s main function of providing education to its pupils and that the license to SHEL was secondary, simply putting to productive use that which has been acquired for a different main purpose. In terms of VAT, the provision of an exempt supply of education was the principal use of the sports complex and the taxable supply of the licence to SHEL was a secondary use.

Any method of allocation between the exempt and taxable supplies made by the School must, in my judgment, reflect that use.

This decision supports the conclusion that the relevant “use” of the acquisitions is the “economic use”.

McInroy & Wood Ltd v Revenue & Customs [2008] UKVAT V20780

The taxpayer provided investment management and advice, making both taxable and exempt supplies. A subsidiary of the taxpayer managed and administered 5 unit trusts. The subsidiary had no employees and all work was carried out by the taxpayer’s employees. The issue was the appropriate apportionment of VAT incurred on the costs of building new premises for the taxpayer’s operations. Some clients had portfolios with only direct shareholdings, some had investments in the subsidiary’s unit trusts and some had both. The clients paid an annual 1% charge on shareholdings owned by the clients, but the charge did extend to the unit holdings of the clients. Rather, a 1.5% charge was deducted by the trustee of the unit trusts. A “recharge” was paid annually by the subsidiary to the taxpayer of varying amounts.

In recent years the volume of investment in the unit trusts increased and the Group’s exempt income increased, while its taxable income had reduced. The application of the standard method therefore produced a lower recover of credits – only about 20% in recent years. The taxpayer proposed a special method which resulted in recoverability of 76%, or 57% if ignoring the building costs (this was the figure adopted by the taxpayer in final submissions). The special method was an “indirect input method” and calculated the proxy by reference to those expenses in the accounts which could be directly attributed to taxable and exempt activities respectively.

The taxpayer contended that the standard method did not produce a fair and reasonable result because there was no relationship between the resources applied by the taxpayer and the income received. The taxpayer argued that each unit trust was simply one client, and each unit trust should be viewed equally with the many individual clients. Further, the scale of each unit trust’s investments was irrelevant and the trusts required a disproportionately small amount of the staff’s work. The taxpayer also contended that the special method did produce a fair and reasonable result because the business involved intellectual effort by staff and, if staff spent more time on non-unit trust investors, expenditure on overheads should reasonably be attributed in a similar way.

The Revenue contended that the taxpayer’s special method did not provide an appropriate proxy for use, that being “economic” use. The special method did not take sufficient account of the re-charge which was paid by the subsidiary to the taxpayer. The taxpayer would have made losses without the re-charge.

The Tribunal agreed with the Revenue. The Tribunal observed that the taxpayer’s contention was that the unit trusts were viewed as five clients as against, say, 50 individuals as clients and that any extra unit trust was only one additional client irrespective of the size of its capital, using the same services, and so not increasing overheads (the acquisitions to be apportioned). The Tribunal did not find this interpretation persuasive or realistic and found that, in circumstances where the charging structure was based on the value of assets an apportionment by reference to turnover was eminently appropriate as fair and reasonable. The income and turnover of the taxpayer depended on the value of all its clients’ assets, not the number of clients. The Tribunal also considered that the special method proposed by the taxpayer was not fair and reasonable. This was because the expenses attributed to taxable expenditure included two very substantial sums which related to an outsourced matter which tended to distort a fair allocation of the un-attributed expenditure.

Camden Motors (Holdings) Ltd v Revenue & Customs [2008] UKVAT V20674

The taxpayer’s principal activity was the sale of new and used cars, including fleet sales and sales to retail customers. The fleet sales comprised over 60% of annual turnover and generated exclusively taxable income. The retail business generated taxable income through the sale of vehicles but also included acting as agent in the sale of finance and insurance products which generated exempt income. The exempt income represented less that 1% of turnover and under the standard method the taxpayer was entitled to recover 100% as the recovery could be rounded up.

The Commissioner contended that the taxpayer was required to adopt an override method because the standard method was not fair and reasonable. The standard method did not fairly represent the extent to which the goods and services were used because the apportionment was distorted by including, in the same calculation, the selling of cars which had very high direct cost and very low margins, alongside finance commissions, which had very low direct costs and very high margins. The Commissioner contended that the cost of the cars should be excluded from the calculation – ie, only the margin should be included in the formula.

The Tribunal found that the standard method was fair and reasonable and that the Revenue was wrong to discount the capital cost of the cars. The reasons for this view included the following:

• If the taxpayer decided to stop selling finance, the effect on overheads would not be very considerable.
• The taxpayer incurred capital costs in order to sell cars, including the showroom and space for offices etc. While the showroom was also used to facilitate the sale of finance, it could not reasonably be said that the expenditure in respect of it is used more extensively for the sale of finance than for the sale of cars.
• the sale of cars was the economic driver of the business. In its car-dealing it incurred all the costs and bore all the rewards, which was not the case in relation to the sale of finance (in which it acted as agent).
• The primary cost of the exempt business was labour, which was not VAT bearing. To include these costs was to distort the picture.
• The Commissioner’s method ignored the central feature of the taxpayer’s business, which was selling cars, and failed to take account of the fact that it was more expensive for the taxpayer to earn taxable income than exempt income.
• The Commissioner improperly relied on the relevant profitability of the taxpayer’s taxable and exempt activities, which was contrary to the basic principles of VAT as a tax on turnover, not a tax on profit.

Oxfam v Revenue & Customs [2009] EWCH 3078

The taxpayer and the Revenue had agreed in 2000 on a special method of apportionment to calculate credit recovery.

In 2007 the Revenue decided to amend the agreed method in respect of costs incurred in securing unrestricted charitable donations. This was because in 2005 the High Court handed down a decision that changed the parties’ understanding of the law relating to the VAT status of unrestricted fundraising expenditure.[10] Prior to this decision, the approach of the Revenue was that the income received from voluntary donations was outside the scope of VAT and VAT incurred on unrestricted fundraising expenditure was wholly irrecoverable. When this approach was applied to the approved method, the unrestricted fundraising expenditure was included in the denominator and produced smaller recovery percentage. The effect of the decision was that the donations were outside the scope of VAT but that VAT was recoverable in part. The taxpayer then excluded the unrestricted fundraising expenditure from the denominator, leading to a higher credit recovery. The recovery was now 85-90%, as opposed to 75%.

The taxpayer agreed that the Revenue was entitled to revoke its acceptance of the agreed method prospectively – the issue in the case was whether that acceptance could be revoked retrospectively. On the question of the apportionment methodology, the Court made the following observations (at [27]):

This change in the level of recovery of input tax by Oxfam, if the Approved Method was adhered to, would be achieved even though there was no significant change in the pattern of supplies being made by Oxfam over the whole of the material period up to and after the judgment in Church of England Children’s Society and even though Oxfam had been content for some years to accept as fair (indeed, had proposed) a method of apportionment which gave a level of apportionment of only 75% of input tax to Oxfam’s own taxable supplies. These are points of significance, since the underlying object of the discussions about and adoption of the Approved Method was to achieve an appropriate and realistic apportionment of input tax between Oxfam’s taxable and non-taxable supplies, as required by Article 17(2) of the Sixth Directive and s. 24(5) of VATA. Since the balance between Oxfam’s own taxable and non-taxable supplies did not change over the period, then as a matter of principle one would not expect the appropriate apportionment of input tax between them to change either.

This decision is an illustration that a methodology may cease to be fair and reasonable if the facts change – eg, if the tax treatment of certain of the inputs or outputs change.

Vision Express (UK) Limited v Revenue & Customs [2009] EWHC 3245

The taxpayer was an optical services retailer which made taxable supplies (the sale of lenses and frames) and exempt supplies (the supply of professional eye testing services and the supply of prescriptions for lenses). The taxpayer sought to adopt a special method which was based on floor area and particular zoning within the store, those zones being partitioned into “taxable”, “exempt” and “non-attributable” spaces. The question before the High Court was whether this was a fair and reasonable method.

The Court (at [30]) observed that the taxpayer treated the income from the sale of each dispensed spectacles as consisting of 35.63% taxable and 64.37% exempt and for contact lenses the split was 49% to 51%. Nevertheless, the special method proposed by the taxpayer resulted in an attribution of 88% of the store related input tax to taxable supplies. The Court observed that those stark figures raised a very serious question, at the outset, as to whether the special method represented a fair and reasonable attribution of those costs in question. As observed by the Court, “the immediate impression from these simple figures is that something has gone wrong somewhere”.

The Court found that the taxpayer could not overcome the finding of facts of the Tribunal below that the stores as a whole were used for the supply and sale of dispensed spectacles and contact lenses and that it was unrealistic to divide them into retail areas and medical care areas. In these circumstances, the economic use of the inputs could not be fairly and reasonably attributed between taxable and exempt supplies by the rigid categorisation of floor area proposed by the taxpayer. The Court approved of the approach adopted by the Court in St Helen’s School and the search for the “economic use” to which the relevant inputs were directed.

HJ Banks & Company Ltd v Revenue & Customs [2010] UKFTT 33

The taxpayer purchased land and two commercial buildings with the intention of selling the commercial properties and building nine luxury apartments on the land for sale. The taxpayer was entitled to fully recover the VAT on the purchase. The taxpayer’s intention changed in respect of one of the commercial properties and it was sold as an exempt supply. This required an adjustment of the VAT claimed.

The Revenue contended that the application of the standard method did not produce a fair and reasonable result and sought to apply an override method. This was because the vast majority of the turnover of the taxpayer for the period in question was taxable.

The taxpayer contended that the most representative proxy for the use to which the land was put in making taxable supplies was the respective sale prices of the various onward transactions. This produced an apportionment of 85.28% to taxable supplies and 14.72% to exempt supplies. The Revenue contended that this approach did not give a fair and reasonable result because the two commercial properties were sold in the same state as they were purchased, but the sale price of the residential apartments included the value of the building. The Revenue contended that an apportionment based on site area was a more accurate reflection of the uses to which the land was put – this resulted in an apportionment of 80.94% to taxable supplies.

The Tribunal agreed with the Revenue that the taxpayer’s apportionment method contained an inbuilt distortion by including the fully developed residential site. The Tribunal observed that the taxpayer’s method would have carried more weight if the construction costs had been deducted. The Tribunal preferred the method proposed by the Revenue and considered that site area provided an objective and reasonable measure to represent the extent to which the land was used for taxable supplies.

Revenue & Customs v London Clubs Management Ltd [2011] EWCA Civ 1323:

The taxpayer operated casinos and made the following supplies: gaming, such as roulette and blackjack which were exempt from VAT; slot machines which were taxable until 27 April 2009 and then exempt; bar sales, catering, entertainment and venue hire, which were taxable.

Under the taxpayer’s existing approved special method, residual input tax was apportioned by applying a fraction comprising taxable turnover over total turnover, with an adjustment to take account of the fact that residual input tax attributable to food and drink for which customers were not charged was not deductible. The taxpayer proposed a special method based on floor space, being the floor area occupied to make taxable supplies over the total floor area, with an adjustment to take account of the residual costs associated with non-charged food and drink.

The Court of Appeal (at [38]) adopted the finding in St Helen’s School that physical use may reflect economic use, but does not necessarily do so, and that any allocation or special method must give a creditable result in economic terms. The Court also observed (at [42]) that an analysis of attribution is highly fact sensitive.

The Tribunal below accepted that the special method was fair and reasonable as it reflected the direct use of the premises costs. The Tribunal recognised that the catering activities were used, to an extent, to support and foster the gaming activities by means of the provision of free food and drink to certain gaming customers. The Tribunal found that the proposed special method recognised this feature because the effect of the formula was to apportion the food and beverage floor space between the chargeable and non-chargeable catering supplies, but to regard the element of non-chargeable catering use as economic use for the purpose of the exempt gaming activities. That was the effect of excluding the non-chargeable catering use from the numerator in the fraction, but including the whole of the food and beverage floor space in the denominator.

The High Court dismissed the Revenue’s appeal. So did the Court of Appeal.

Determining the “taxable value” of supplies which are both taxable and GST-free and/or input taxed

Section 9-80 provides the statutory machinery to determine the value of taxable supplies that are partly GST-free or partly input taxed. The section provides as follows:

(1) If a supply (the actual supply) is:

(a)partly a *taxable supply; and

(b) partly a supply that is *GST-free or *input taxed;

the value of the part of the actual supply that is a taxable supply is the proportion of the value of the actual supply that the taxable supply represents.

(2) The value of the actual supply, for the purposes of subsection (1), is as follows:

*price of the actual supply * 10

10 + Taxable proportion

where:

taxable proportion is the proportion of the value of the actual supply that represents the value of the *taxable supply (expressed as a number between 0 and 1).

Food Supplier and Commissioner of Taxation

The operation of the section was considered by the President of the Tribunal in Food Supplier and Commissioner of Taxation (2007) 66 ATR 938.

That case involved the sale by the applicant of “free” promotional items sold with certain food products, like instant coffee sold with a “free” coffee cup. The combined packages were sold for the same price as the food product alone. The question was whether the promotional items attracted GST. The Tribunal found that it did, but as the food component was GST-free, the taxed component was confined to the promotional item.

The Tribunal agreed that having determined that the supplies in question gave rise to a liability for GST, s 9-80 provides the means by which the necessary calculation for GST must be made. The taxpayer contended that s 9-80 did not authorise an apportionment calculation and that the Commissioner may only apportion when the transactions themselves disclose apportionment figures. The Tribunal did not agree and observed as follows (at [23]):

Before s 9-80 can relevantly apply there must be a prior determination that there is a supply which is partly taxable and partly GST-free. Once that determination is made, s 9-80 requires an apportionment. The section may be difficult to apply but that is not a reason for declining the task or treating the supply as not falling within its terms. I note that the section refers to “value” rather than “price” or “consideration”. As Hely J said in Kmart Australia Ltd v Commissioner of Taxation (2001) 11 FCR 353, referring to a not dissimilar section in the Sales Tax Assessment Act 1991 (Cth) (s 45) “the Act requires a valuation exercise to be conducted. That exercise is an artificial one… [which] must be undertaken as a matter of practical common sense”.

The Commissioner applied s 9-80(1) and took the cost of the promotional item to the applicant and calculated the amount of GST from the proportion that amount bore to the price for which the packaged product was sold. The Tribunal’s tentative opinion was that the calculation made by the Commissioner was reasonable, but thought it best to leave the final calculation to the parties.

Commissioner of Taxation v Luxottica Retail Australia Pty Ltd

The observation of the President in Food Supplier that s 9-80 “may be difficult to apply” was borne out in the decision of the Full Federal Court in Commissioner of Taxation v Luxottica Retail Australia Pty Ltd (2011) 191 FCR 561.

The taxpayer was a retailer of spectacles which comprised prescription lenses fitted into frames for glasses and sunglasses. The taxpayer ran promotions where spectacle frames were offered at a discount from the normal selling price but on condition that the customer acquired not only the frames but also lenses for the frames. No discount was offered for the lenses.

At issue was how the GST should be calculated on the discounted price. The Commissioner submitted that the discount should be deducted from the combined purchase price of the spectacles whereas the taxpayer contended that the discount should be deducted from the frame price only.

The Full Court considered the operation of s 9-80 and observed that it was designed to ascertain the value of a taxable supply in order to calculate the GST payable in respect of that supply. However, the Court found that the formula set out in s 9-80(2) simply did not work, noting the comments of the Tribunal below that the formula was “impenetrably circular”. Accordingly, the value of the taxable supply could not be determined by the application of that formula.

Given that the statutory formula did not work, the Court found that the decision maker must, taking into account the relevant circumstances of the particular case, reach a conclusion as to value and the relationship it has to the price of the supply in question. The Court found that the Tribunal, in adopting an approach which regarded value as commensurate with price, made a considered decision as to the value of the taxable supply based on findings of fact it was entitled to make. As a matter of fact the Commissioner would prefer that the price of the frames sold as a separate item be the basis for determining value. There was no error of law in the Tribunal adopting a different approach to this factual question.

GSTR 2001/8

The Commissioner outlines his views on the approach to apportionment under s 9-80 in GSTR 2001/8. The Commissioner’s view is as follows (paragraph 26):

Apportionment must be undertaken as a matter of practical commonsense. You can use any reasonable basis to apportion the consideration. Depending on the facts and circumstances of the supply, a direct or indirect method may be an appropriate basis upon which to apportion the consideration and ascertain the value of the taxable part of the supply. The basis you choose must be supportable in the particular circumstances.

To address the decision in Luxottica, the ruling was amended to include the following paragraphs:

81M. To work out the taxable proportion following the Full Federal Court decision, the value of the taxable part of the supply has to be determined by having regard to the facts and circumstances and taking a practical, commonsense approach. The question to be answered is what is a fair and reasonable measure of the value of the taxable part?

81N. The value of the taxable part of a supply may be synonymous with the selling price of that part as in Luxottica or, as in Food Supplier, where there was no market for the taxable promotion item, it may be necessary to consider other practical, commonsense means of fixing value such as cost plus a margin.

The position under s 9-80 now appears to be similar to that as in Division 11, namely that it is necessary to determine an apportionment methodology which is “fair and reasonable”.

CONCLUSION

Where apportionment is available under the GST Act, the task imposed on the taxpayer is to determine an apportionment methodology that is “fair and reasonable” in the particular circumstances. This can be difficult task and one on which the Commissioner and the taxpayer may well take a different view, particularly if the Commissioner considers that the outcome gives an inappropriate revenue outcome (either by the taxpayer claiming too high a proportion of credits on acquisitions or paying too low a proportion of GST on supplies).

While there appears to have been only one reported decision in Australia directly dealing with the question of whether an apportionment methodology was “fair and reasonable”, a review of the reported decisions in the United Kingdom (being a mature VAT regime) discloses a number of cases where the issue has arisen. A review of these cases discloses the following propositions that I consider may be equally applicable in Australia:

• In each case the question is highly fact specific;
• The underlying question involves searching for the relevant “use” of the inputs which are to be apportioned. The concept of use involves “economic use”;
• The physical use of inputs may reflect the economic use of the inputs, but that is not necessarily the case;
• In considering an apportionment methodology, one must be wary of any factors which may cause a distortion in the results and a resultant disconnect between the “use” of the inputs and the outputs of the entity; and
• If the particular circumstances of an enterprise change, this may cause an apportionment methodology to no longer be fair and reasonable.

At the end of the day, to adopt the obiter comments of Lindgren J in AXA, the ultimate search may be for the reason why a particular methodology and proxy proposed are most likely to approximate the relatedness between acquisition of the things and the use made of them.

Chris Sievers

Lonsdale Chambers

13 April 2015

[1] A New Tax System (Goods and Services Tax) Act 1999. Unless stated otherwise, all statutory references are to the GST Act.

[2] Ronpibon Tin NL v Federal Commissioner of Taxation (1949) 78 CLR 47 at 55-6.

[3] For example GSTR 2006/3 Goods and services tax: determining the extent of creditable purpose of providers of financial supplies; GSTR 2006/4 Goods and services tax: determining the extent of creditable purpose for claiming input tax credits and for making adjustments for changes in extent of creditable purpose; GSTR 2008/1 Goods and services tax: when do you acquire anything or import goods solely or partly for a creditable purpose?

[4] Assuming the remaining elements in s 11-5 are satisfied.

[5] Stone and Allsop JJ agreed with his Honour’s reasons.

[6] AXA Asia Pacific Holdings Limited v Commissioner of Taxation (2008) 173 FCR 500 at [38]; approved in Rio Tinto Services Ltd v Commissioner of Taxation [2015] FCA 94 at [24], noting that the taxpayer has filed an appeal to the Full Federal Court.

[7] Kenny and Middleton JJ, Dowsett J dissenting.

[8] As from 1 April 2011 new apportionment rules were introduced in New Zealand through the Taxation (GST and Remedial Matters) Act 2010.

[9] Lindgren J did make the observation (at [75]) that in a case of apportionment under s 11-30, ideally the evidence would specify the nature and use made of the things acquired in some detail, the attempts made to relate them directly to supplies, and the reason why the particular methodology and proxies proposed are most likely to approximate the relatedness between acquisition of the things and the use made of them.

[10] Church of England Children’s Society v Commissioners of Revenue and Customs [2005] WEWHC 1692 (Ch); [2005] STC 1644.