Resolution: the overheads
73 With regard to the overheads, the applicant contends that they relate indifferently to all the applicant's activities in the course of carrying on its enterprise. It says that the activities of its enterprise relevantly included the making of the following supplies:
(1) input taxed supplies of life insurance policies in Australia;
(2) input taxed supplies of reinsurance of life insurance policies in Australia;
(3) GST-free supplies of reinsurance of life insurance policies issued in New Zealand;
(ie, the three revenue earning activities identified in [3] above), and
(4) GST-free acquisition supplies of reinsurance and retrocession of life insurance policies with Hannover Rück.
74 The applicant notes that the Commissioner accepts that the overheads are creditable acquisitions with a partly creditable purpose to the extent that they relate to the GST-free supply of reinsurance of life insurance policies issued in New Zealand, and that the appropriate method of calculating the applicant's entitlement to apportionment of the overheads in respect of those supplies is to follow its statutory expense allocation.
75 The Commissioner submits that the overheads do not relate indifferently to all the applicant's activities in the course of carrying on its enterprise. However, the Commissioner accepts that the overheads can be apportioned with respect to the New Zealand reinsurance supplies. That means that the Commissioner accepts, with reference to s 11-15(2)(a), that although there is a real and substantial relationship between the overhead acquisitions and the making of the input taxed supplies of Australian insurance and reinsurance policies, the relationship is not to a complete extent. That is to say, the Commissioner accepts that there is an apportionable extent to which the overheads do not relate, in a real and substantial way, to the making of the input taxed supplies of Australian insurance and reinsurance policies - or, to put it the other way, they relate sufficiently, to an apportionable extent, to the GST-free New Zealand reinsurance supplies.
76 That acceptance by the Commissioner undercuts the Commissioner's submission that the overheads do not relate indifferently to the various activities, or supplies, of the applicant's enterprise. Putting to one side how an appropriate apportionment might best be done, that is because there is no apparently relevant distinction between the relationship that the overhead acquisitions have to the GST-free New Zealand reinsurance supplies and the relationship that they have to the GST-free Hannover Rück reinsurance acquisition supplies. If the overheads are partly creditable in respect of the New Zealand supplies then why not also the Hannover Rück supplies?
77 The Commissioner distinguishes the New Zealand supplies and the Hannover Rück acquisition supplies by categorising the overheads into two types of acquisition:
(1) The first category consists of acquisitions that allow the applicant's employees to do their work or assist them in doing that work. For the 2014 and 2015 years, this category includes rent to acquire office premises and IT, and for later years it includes office supplies, telephone and fax expenses, office expenses and expenses for copying and archiving.
(2) The second category consists of other acquisitions. For the 2014 and 2015 years, these include audit expenses, and for later years they include consulting fees and expenses for advertising. The Commissioner identifies these acquisitions as relevant to the operation of the applicant's business, but they do not assist the applicant's employees in doing their work.
78 The Commissioner submits that in respect of the first category, the relationship between the work done by an employee to issue a New Zealand reinsurance policy and the supply of the reinsurance policy is "by and through" that GST-free supply. In contrast, the relationship between the work done by an employee and the applicant's GST-free reinsurance supply with Hannover Rück is "by and through" the applicant's input taxed supply of the Australian insurance policy issued through Greenstone. That is because, the Commissioner submits, the applicant's making of the GST-free supply with Hannover Rück is wholly reactive and responsive to it making an input taxed supply of insurance or reinsurance to an Australian insured or reinsured, both as a matter of legal analysis and practical reality.
79 I do not accept that submission. The overhead acquisitions do not arise "by and through" the input taxed supplies in the same way that the Greenstone acquisitions do, or in the same way Rio Tinto's did. The input tax credits sought to be claimed on the Greenstone acquisitions, as explained, amount to the GST paid by the applicant to Greenstone on the latter's RCTIs for work done exclusively in relation to input taxed supplies. The very nature of the input tax credits sought to be claimed on the overheads expenditure is that they do not have that character; they are acquisitions that do not have any immediacy of connection to input taxed supplies. If one takes as an example the payment of an invoice for rent for premises on which GST is paid - that being the precise identification of the relevant acquisition, it has no greater relationship or connection to one type of the applicant's supplies as any other. That is the very basis for its categorisation as an undifferentiated overhead.
80 The Commissioner submits that subject to one exception, the making of the GST-free supply with Hannover Rück does not involve any work being done by the applicant's employees other than the work that they do to make the input taxed supply of Australian insurance or reinsurance. The exception is any identifiable work done by employees to negotiate the contractual terms of the arrangements with Hannover Rück, and work done to carry out those contractual terms, such as to prepare reports and notices required under the contractual terms to be given by the applicant to Hannover Rück. The Commissioner submits that to that extent some portion of the applicant's acquisition of "overheads" such as office premises allow those employees to do that work, and so those acquisitions might to that extent not relate to the making of input taxed supplies.
81 The problem, the Commissioner submits, is that there is insufficient evidence to enable the Court to identify in any meaningful way the extent to which the applicant's employees did work of that nature. Moreover, what evidence there is suggests that relatively minimal employee time and effort was spent doing such work.
82 The difficulty with that approach is that it accepts that some overheads can be appropriately attributed to the relevant GST-free supplies, but then on the basis that the (only) way to do that is with reference to an allocation of employees' time, concludes that there can be no allocation because the evidence of that time is insufficient or the time spent was insignificant. The correct approach, once it is accepted that there is a sufficient relationship between the relevant acquisition and a GST-free supply to make the acquisition for a creditable purpose, is to look for the proper method of apportionment. In that regard, the time spent by employees might be regarded as only one of a number of possible proxies to form the basis for such an apportionment.
83 There is no apparent reason why time spent by employees in relation to one or other type of supply should be regarded as the only appropriate proxy, or indeed why it is an appropriate proxy at all when it comes to acquisition supplies. For example, time spent as a proxy does not differentiate between work done by senior and junior employees. It may be that the work done on the reinsurance relationship with Hannover Rück is done with great time efficiency by very senior employees who, on a time-costed basis, come at a significant cost to the applicant although taking an insignificant amount of time, and the work done directly in relation to other supplies such as the relevant input taxed supplies is done by junior employees at less significant cost although taking a significant amount of time. Why should such undifferentiated time spent be the appropriate basis for a method of apportionment that is "fair and reasonable in the circumstances of [the] enterprise" (referring to GSTR 2006/4)?
84 I accept, of course, that the statutory expense allocation is done using employee hours as the relevant proxy, but that does not make it the necessary proxy for the creditable acquisition allocation when the supplies with Hannover Rück are concerned. The key difference between the statutory expense allocation exercise and the exercise of apportioning partly creditable acquisitions is that the former takes into account only the revenue earning output of the enterprise as expressed in its statutory funds, ie, the policies of insurance and reinsurance that it sells and on which it earns premium income. See s 30 of the Life Insurance Act. The latter exercise - apportioning partly creditable acquisitions - must take into account all supplies, which include the acquisition supplies from/to Hannover Rück which do not find expression in a statutory fund and which are undertaken at a net expense to the enterprise.
85 I therefore do not accept that because very little employee time is spent on the Hannover Rück GST-free supplies, or because the evidence of time spent on those supplies is scant, the first category of acquisitions is not partly creditable as having a partly creditable purpose in relation to the GST-free supplies.
86 The Commissioner submits that the second category of overhead acquisitions, as with the first, does not have any relationship with the Hannover Rück GST-free supplies other than "by and through" the input taxed supplies of insurance and reinsurance made by the applicant. The first example the Commissioner takes is that of audit expenses and consulting fees, and submits that there is no evidence to suggest that these "overheads" have anything to do with the reinsurance and retrocession arrangements with Hannover Rück, other than by reason of those arrangements applying automatically when the applicant makes an input taxed supply of insurance and reinsurance.
87 However, it is far from clear that that is so. First, there is nothing in the nature of audit expenses or consulting fees that suggests that they were incurred specifically in relation to input taxed supplies. Indeed, by their nature, they would appear to be acquisitions which could have been incurred in respect of all the different supplies done by the applicant and might not have been incurred exclusively in relation to input taxed supplies; presumably the Hannover Rück relationship was also subject to audit and to the work of consultants. Secondly, as mentioned, Mr Tallack's evidence was that the "overhead" expenses - including the audit and consulting fees - could not be directly attributed to any particular activities or product lines. Not only was Mr Tallack not challenged on that - he was not required for cross-examination - but there is no reason to doubt that evidence.
88 The second example taken by the Commissioner is the acquisition of advertising, which is included in the "overheads". The Commissioner submits that it is inherently more likely that the applicant was advertising its insurance products and, perhaps, its reinsurance products, rather than advertising anything to do with its relationship with Hannover Rück.
89 I accept that it does seem unlikely that the applicant incurred advertising expenses in respect of its reinsurance and retrocession arrangements with Hannover Rück. However, as mentioned, Mr Tallack's evidence was that these expenses cannot be attributed to any particular product, reinsurance treaty or product line. That unchallenged evidence must be accepted. Therefore, for the same reasons as given in relation to the first category at [79] above, the advertising, audit and consulting expenses cannot be said to have only related to the reinsurance or retrocession arrangements with Hannover Rück "by and through" the applicant's making of input taxed supplies of insurance and reinsurance. On the evidence, they were incurred independently of, or indifferently in respect of, all the four types of supply identified at [73] above.
90 In the circumstances, I am satisfied that the acquisitions described by Mr Tallack as "overheads" are indeed not attributable to any particular supply made by the applicant. I accept that in that sense none of the individual overheads has a more real and substantial relationship with any particular supply than with another. That is to say, the acquisition of the overheads relates in a real and substantial way to making supplies that would be input taxed, but no more than to the applicant's GST-free supplies and therefore only partially.
91 The acquisition of the overheads is therefore partly creditable, and the amount of the input tax credit on the acquisition must be calculated under the formula in s 11-30(3). That requires that the "extent of creditable purpose", being the extent to which the creditable acquisition is for a creditable purpose expressed as a percentage of the total purpose of the acquisition, must be calculated. That must be done in a way that is fair and reasonable in the circumstances of the applicant's enterprise.
92 The applicant proposes a formula for calculating the "extent of creditable purpose" (ECP), which uses a combination of the statutory expense allocation and premium revenue as a "proxy", as follows:
ECP = NZ% + (ARSF% * A/B) + (ASF% * C/D)
Where:
ECP = the extent of creditable purpose
NZ% = the percentage of expenses allocated to the OSF statutory fund under the Statutory Expense Allocation
ARSF% = the percentage of expenses allocated to the ARSF statutory fund under the Statutory Expense Allocation
A = total retrocession premiums paid by the Applicant to Hannover Rück in respect of the risk ceded from the ARSF statutory fund
B = total life insurance premiums received by the Applicant in respect of life insurance policies within the ARSF statutory fund
ASF% = the percentage of expenses allocated to the ASF statutory fund under the Statutory Expense Allocation
C = total reinsurance premiums paid by the Applicant to Hannover Rück in respect of the risk ceded from the ASF statutory fund
D = total life insurance premiums received by the Applicant in respect of life insurance policies within the ASF statutory fund.
93 The Commissioner offers three reasons why the applicant's proposed apportionment methodology should not be adopted or approved.
94 First, the Commissioner submits that the applicant's methodology fails to account for the fact that, regardless of the extent to which the applicant cedes risk to Hannover Rück, the applicant still makes the whole input taxed supply of insurance or reinsurance. To illustrate the point, the Commissioner takes as an example a case where the applicant cedes 100% of the risk to Hannover Rück and so pays 100% of the premiums to Hannover Rück, in which case the fraction A/B and C/D in its methodology is 1. The otherwise partly creditable acquisition would then be entirely attributed to the GST-free supply and none would be attributable to the input taxed supply of insurance or reinsurance. That would be so even though the applicant is still making two supplies and the acquisition is said to relate to both. This is said to illustrate that the methodology does not give a fair and reasonable result.
95 I do not accept this criticism because the proposed apportionment methodology reflects the extent to which, by reference to the reinsurer's share of 75%, the applicant and Hannover Rück share in the risk, premiums and costs with respect to the issue of the life insurance policies. If the share was greater than 75%, even 100%, the outcome of the application of the methodology would then still reflect the net revenue to the applicant from that part of its business. That is to say, there is nothing that strikes me as unfair or unreasonable in a methodology that apportions an otherwise partly creditable acquisition with reference to the revenue value to the enterprise of the respective supplies. In the 100% example, the whole business would in effect be that of Hannover Rück and the applicant would be a mere conduit. It is not obvious why in that situation it should be regarded as unfair or unreasonable that the overheads be wholly creditable.
96 In any event, it is not necessary to look to an extreme example, such as the ceding or retrocession of 100% of the risk. The proposed methodology is to apply only to the known case where 75% of the risk was ceded and 75% of the premium revenue was passed on. That revenue can be an appropriate proxy to calculate the extent of creditable purpose is illustrated by Federal Commissioner of Taxation v American Express Wholesale Currency Services Pty Ltd [2010] FCAFC 122; 187 FCR 398. Kenny and Middleton JJ, the majority, observed (at [79]) that the agreed formula in that case did not directly depend on the nature of the acquisitions themselves, but used revenue as a proxy to calculate the extent of creditable purpose. Rather than undertake the analysis individually for each of the acquisitions in question, the taxpayers used the formula based on revenue figures as a proxy for the relationship between their acquisitions (in the aggregate) and the making of particular supplies (at [79]).
97 Secondly, the Commissioner submits that the methodology overlooks the fact that the applicant makes investments of the capital that it retains, which investments are input taxed supplies. The Commissioner submits that the acquisition of the overheads relates to some extent to the making of such investments but this is not accounted for in the applicant's methodology.
98 The applicant accepts this criticism of the methodology, but says that the formula can be adjusted to have the effect of stripping out the investments. It has suggested a way of achieving this which is reflected in the document MFI-1, but the Commissioner has not responded specifically to that proposal because it was raised late in the piece. In the circumstances, I propose giving the parties the opportunity to discuss the methodology and to agree an amendment to it so as to take account of the investment-related input taxed supplies. To the extent that any disagreement still remains, I will then decide the best way of dealing with it.
99 Thirdly, the Commissioner submits that the applicant's methodology utilises "current year data" which only becomes available at the end of the tax year and is not available for each of the tax periods during the year. The Commissioner submits that that is inappropriate because it is contrary to the scheme of the GST Act and that the apportionment should be calculated using the previous year's data.
100 Whilst I appreciate the underlying relevance of this criticism to a methodology to be applied into the future on an ongoing basis, the fact of the matter is that the question before the Court is whether the assessments for the relevant tax periods, which are all well in the past, are excessive. There is no compelling reason why current year data should not be used for that purpose. I note, also, that the Commissioner does not take issue with current year data being used for the New Zealand reinsurance component.
101 I am accordingly satisfied that the applicant's proposed methodology to calculate the extent of creditable purpose in respect of its overheads in the relevant tax periods is fair and reasonable, subject to an adjustment to exclude an amount attributable to the applicant's input taxed investment supplies. I will give the parties the opportunity to seek agreement on that adjustment, failing which I will hear from them further in order to decide the point myself. The appeal in relation to the overheads otherwise succeeds.