Federal Court of Australia

Commissioner of Taxation v Glencore Investment Pty Ltd [2020] FCAFC 187

ORDERS

THE COURT ORDERS THAT:

1.    The parties are to confer and, if agreement can be reached, provide the Court with orders for final relief within seven days hereof, or failing that, each party shall file written submissions on the issue of the form of final relief limited to 10 pages in length.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

MIDDLETON AND STEWARD JJ.

1    For the years of income ended 31 December 2007 to 31 December 2009 in lieu of the years of income ended 30 June 2008 to 30 June 2010 (the “2007 to 2009 years”), Cobar Management Pty Ltd (“C.M.P.L.”), a resident of the Commonwealth of Australia, sold copper concentrate to its ultimate parent, Glencore International A.G. (“G.I.A.G.”), a resident of the Swiss Confederation. It did so pursuant to amended terms agreed in February 2007 (the “C.M.P.L.-G.I.A.G. agreement”). The Commissioner of Taxation (the “Commissioner”) contends that the consideration G.I.A.G. paid C.M.P.L. for its copper concentrate was not, to use a generic expression, an arm’s length price. He submits that it was too low. Relying on Australia’s transfer pricing laws — contained for the purposes of the 2007 to 2009 years in Div. 13 of Pt. III of the Income Tax Assessment Act 1936 (Cth.) (the “1936 Act”) and in Subdiv. 815-A of the Income Tax Assessment Act 1997 (Cth.) (the “1997 Act”) — the Commissioner issued amended assessments to Glencore Investment Pty Ltd (the “taxpayer”), who at the time was the provisional head company of a multiple entry consolidated group. C.M.P.L. was a subsidiary member of that group. The amended assessments included in the assessable income of the taxpayer what the Commissioner contended, and still contends, was the correct arm’s length consideration for the sale by C.M.P.L. of copper concentrate to G.I.A.G. The learned primary judge disagreed with that contention. Her Honour was satisfied that C.M.P.L. had been paid arm’s length consideration by G.I.A.G., and accordingly that the amended assessments were wholly excessive. The Commissioner appeals that decision to this Court.

Legislation and Swiss Treaty

2    It is necessary to set out the applicable provisions of Div. 13 of the 1936 Act and Subdiv. 815-A of the 1997 Act. Div. 13 has been repealed, but relevantly continues to apply to the 2007 to 2009 years.

3    Here, in applying Div. 13, the Commissioner made determinations pursuant to both ss. 136AD(1) and (4) for the 2007 to 2009 years. Sections 136AD(1) and (4) provided:

Arm’s length consideration deemed to be received or given

(1)    Where:

   (a)    a taxpayer has supplied property under an international agreement;

(b)    the Commissioner, having regard to any connection between any 2 or more of the parties to the agreement or to any other relevant circumstances, is satisfied that the parties to the agreement, or any 2 or more of those parties, were not dealing at arm’s length with each other in relation to the supply;

(c)    consideration was received or receivable by the taxpayer in respect of the supply but the amount of that consideration was less than the arm’s length consideration in respect of the supply; and

(d)    the Commissioner determines that this subsection should apply in relation to the taxpayer in relation to the supply;

then, for all purposes of the application of this Act in relation to the taxpayer, consideration equal to the arm’s length consideration in respect of the supply shall be deemed to be the consideration received or receivable by the taxpayer in respect of the supply.

...

(4)    For the purposes of this section, where, for any reason (including an insufficiency of information available to the Commissioner), it is not possible or not practicable for the Commissioner to ascertain the arm’s length consideration in respect of the supply or acquisition of property, the arm’s length consideration in respect of the supply or acquisition shall be deemed to be such amount as the Commissioner determines.

4    It was accepted that the C.M.P.L.-G.I.A.G. agreement was an “international agreement” for the purposes of Div. 13. The taxpayer did not otherwise dispute that G.I.A.G. and C.M.P.L. did not relevantly deal with each other at arm’s length (although it claims that C.M.P.L. was nonetheless paid arm’s length consideration). It also did not dispute that the Commissioner had been lawfully satisfied that C.M.P.L. and G.I.A.G. were not relevantly dealing with each other at arm’s length. Nor did the taxpayer challenge the legal efficacy of the determinations the Commissioner had made.

5    Section 136AA(3) of the 1936 Act provided as follows in respect of the definition of arm’s length consideration and related matters:

Interpretation

…

(3)    In this Division, unless the contrary intention appears:

…

(b)    a reference to consideration includes a reference to property supplied or acquired as consideration and a reference to the amount of any such consideration is a reference to the value of the property;

(c)    a reference to the arm’s length consideration in respect of the supply of property is a reference to the consideration that might reasonably be expected to have been received or receivable as consideration in respect of the supply if the property had been supplied under an agreement between independent parties dealing at arm’s length with each other in relation to the supply;

...

(e)    a reference to the supply or acquisition of property under an agreement includes a reference to the supply or acquisition of property in connection with an agreement.

6    Subdivision 815-A was inserted into the 1997 Act in 2012, but is relevantly expressed to apply retrospectively to the 2007 to 2009 years by reason of s. 815-1 of the Income Tax (Transitional Provisions) Act 1997 (Cth.) (the “Transitional Provisions Act”). The taxpayer did not challenge the validity of Subdiv. 815-A’s retrospective application.

7    The object of Subdiv. 815-A of the 1997 Act is set out in s. 815-5, which relevantly provides:

Object

The object of this Subdivision is to ensure the following amounts are appropriately brought to tax in Australia, consistent with the arm’s length principle:

(a)    profits which would have accrued to an Australian entity if it had been dealing at *arm’s length, but, by reason of non-arm’s length conditions operating between the entity and its foreign associated entities, have not so accrued;

8    The operative provision is s. 815-10, which relevantly provides:

Transfer pricing benefit may be negated

(1)    The Commissioner may make a determination mentioned in subsection 815-30(1), in writing, for the purpose of negating a *transfer pricing benefit an entity gets.

Treaty requirement

(2)    However, this section only applies to an entity if:

(a)    the entity gets the *transfer pricing benefit under subsection 815-15(1) at a time when an *international tax agreement containing an *associated enterprises article applies to the entity; or

(b)    the entity gets the transfer pricing benefit under subsection 815-15(2) at a time when an international tax agreement containing a *business profits article applies to the entity.

9    The Commissioner made determinations here pursuant to this provision for the 2007 to 2009 years in addition to the determinations made under Div. 13 of the 1936 Act.

10    Section 815-15 delineates when a taxpayer gets a transfer pricing benefit for the purposes of s. 815-10. The section relevantly provides:

When an entity gets a transfer pricing benefit

Transfer pricing benefit—associated enterprises

(1)    An entity gets a transfer pricing benefit if:

  (a)    the entity is an Australian resident; and

(b)    the requirements in the *associated enterprises article for the application of that article to the entity are met; and

(c)    an amount of profits which, but for the conditions mentioned in the article, might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued; and

  (d)    had that amount of profits so accrued to the entity:

(i)    the amount of the taxable income of the entity for an income year would be greater than its actual amount; or

(ii)    the amount of a tax loss of the entity for an income year would be less than its actual amount; or

(iii)    the amount of a *net capital loss of the entity for an income year would be less than its actual amount.

The amount of the transfer pricing benefit is the difference between the amounts mentioned in subparagraph (d)(i), (ii) or (iii) (as the case requires).

11    The term “associated enterprises article” is defined by s. 815-15(5) in the following way:

Meaning of associated enterprises article

(5) An associated enterprises article is:

(a) Article 9 of the United Kingdom convention (within the meaning of the International Tax Agreements Act 1953); or

(b) a corresponding provision of another *international tax agreement.

12    Relevantly here, the Commissioner may only make a determination under s. 815-10 to negate a transfer pricing benefit if an international tax agreement containing an associated enterprises article applies to the taxpayer. There was no dispute that this requirement was satisfied because of the application to the taxpayer of Art. 9 of the Agreement between Australia and Switzerland for the Avoidance of Double Taxation with Respect to Taxes on Income, and Protocol [1981] ATS 5 (the “Swiss Treaty”). The Swiss Treaty has since been replaced by the Convention between Australia and the Swiss Confederation for the Avoidance of Double Taxation with respect to Taxes on Income, with Protocol [2014] ATS 33, but that agreement only entered into force on 14 October 2014. Accordingly, Art. 9 of the Swiss Treaty is the relevant associated enterprises article for the 2007 to 2009 years. It provided:

Where –

(a)    an enterprise of one of the Contracting States participates directly or indirectly in the management, control or capital of an enterprise of the other Contracting State; or

(b)    the same persons participate directly or indirectly in the management, control or capital of an enterprise of one of the Contracting States and an enterprise of the other Contracting State,

and in either case conditions operate between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another, then any profits which, but for those conditions, might have been expected to accrue to one of the enterprises, but, by reason of those conditions, have not so accrued, may be included in the profits of that enterprise and taxed accordingly.

13    The taxpayer did not dispute that some conditions operated between C.M.P.L. and G.I.A.G. which differed from those which might be expected to have operated between independent enterprises dealing wholly independently with one another. It accepted that G.I.A.G. exercised financial and managerial control over C.M.P.L.’s mine.

14    Subdivision 815-A identifies certain extrinsic documents which must be considered when applying its provisions. A determination as to whether a taxpayer got a transfer pricing benefit must be made “consistently” with those documents “to the extent the documents are relevant.” Similarly, an international tax agreement must be construed consistently with those documents to the extent that they are relevant to that task. Section 815-20 thus provides:

Cross-border transfer pricing guidance

(1)    For the purpose of determining the effect this Subdivision has in relation to an entity:

(a)    work out whether an entity gets a *transfer pricing benefit consistently with the documents covered by this section, to the extent the documents are relevant; and

(b)    interpret a provision of an *international tax agreement consistently with those documents, to the extent they are relevant.

(2)    The documents covered by this section are as follows:

(a)    the Model Tax Convention on Income and on Capital, and its Commentaries, as adopted by the Council of the Organisation for Economic Cooperation and Development and last amended on 22 July 2010;

(b)    the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, as approved by that Council and last amended on 22 July 2010;

(c)    a document, or part of a document, prescribed by the regulations for the purposes of this paragraph.

(3)    However, a document, or a part of a document, mentioned in paragraph (2)(a) or (b) is not covered by this section if the regulations so prescribe.

(4)    Regulations made for the purposes of paragraph (2)(c) or subsection (3) may prescribe different documents or parts of documents for different circumstances.

15    By reason of s. 815-5 of the Transitional Provisions Act, the parties agreed that the relevant version of the Transfer Pricing Guidelines are those which were published by the Council of the Organisation for Economic Co-operation and Development (the “O.E.C.D.”) in 1995 (the “Transfer Pricing Guidelines”).

The Facts Found by the Learned Primary Judge

16    Nearly all of the facts found by the learned primary judge were not in dispute. Rather, and with respect, the real contest before us was as to which expert’s opinion should be preferred: that of Mr. Wilson, one of the experts called by the taxpayer; or that of Mr. Ingelbinck, one of the experts called by the Commissioner. We shall return to this issue.

17    C.M.P.L. owned and operated an underground copper mine at Cobar in central western New South Wales. The Glencore Group, if it may be so called, acquired the mine in the late 1990s. C.M.P.L. entered into its first contract to sell all of its copper concentrate to G.I.A.G. in 1999. The mine had two attributes which should be mentioned; it was relatively small and it had high operating costs. The learned primary judge described the mine (the “C.S.A. mine”) in the following terms at [111]-[112]:

The CSA mine is a high grade underground copper mine and copper concentrate plant located near the town of Cobar in New South Wales. Although in the relevant years it was the largest of the copper mines in the Cobar region, in relative terms it is a small mine. Mr Kelly deposed that by way of comparison some copper mines in South America (like Escondida) mine in a day or two what CSA mines in a year and, in 2006, the CSA mine’s annual production of concentrate represented about 0.2% of world copper concentrate production.

During the relevant years, the mine’s primary ore body, which was mined from an area known as QTS North, produced an homogenous copper concentrate of a consistently high grade. That ore body was comprised in a series of vertical lenses and was mined using the long hole open stoping method. Mr Kelly gave the following description of the steps involved in the mining operations:

(a)    mining of the underground stopes occurred at depths of about 1,400 metres to 1,500 metres;

(b)    vehicle and heavy machinery accessed the underground stopes via the decline, which was a road which descended into the mine from the surface in a corkscrew manner to a depth of about 1,400 metres to 1500 metres; regular worker and materials access to the mining operations was via a shaft that extended from the surface to a depth of 980 metres, and then via the decline to a depth of 1,400 metres to 1,500 metres;

  (c)    the stopes, once accessed, were drilled and blasted;

(d)    once blasted, the ore was collected and transported in trucks to an underground stockpile located at a depth of about 980 metres;

(e)    from the stockpile, the ore was transported via the ore pass to a primary crusher and then to the surface via the vertical shaft;

(f)    once the ore reached the surface, the ore was transferred to mills where it was ground to a slurry;

(g)    the slurry was then pumped to a flotation tank to separate the different minerals which were skimmed from the flotation cells and then dried. This skimmed and dried material is the copper concentrate which the mine sold; and

(h)    copper concentrate produced at CSA was stockpiled before being transported by rail to the port at Newcastle from where it was then shipped to various ports in Asia.

18    Her Honour accepted the following explanation given by the taxpayer’s lay witness as to why the mine had such high operating costs at [114]:

Mr Kelly’s evidence was that:

(a)    the depth of the mine and the need to continue mining deeper into the ore body gave rise to significant technical and financial challenges for the mine and led to high operating costs. This was because the depth of the mine increased the complexity, time, effort and cost involved in extracting the copper ore from the mine face, getting it to the underground stockpile and then transporting it from the stockpile to the surface; the mine’s depth required expenditure on infrastructure, extensive ground support and stabilisation regimes, as well as systems to provide adequate ventilation, power and air conditioning; and there were difficulties in accessing the deep underground stopes both to bring workers and equipment to and from the underground stopes and in transporting extracted ore to the surface and gave rise to congestion in the movements, as well as long trucking distances, restricting operations. Mr Kelly deposed that from at least 2004, CSA worked on projects with a view to making mining at depth more economical, including a proposal to extend the shaft to a depth of 1,500 metres;

(b)    the remoteness of the mine made it difficult to attract and retain employees and there were significant vacancies in key roles for long periods of time. For an 18 month period leading up to October 2007, there was no mine manager and at other times the mine manager changed regularly; and

(c)    the mine faced the issue of having enough water to support the mine’s production. He explained that mining operations cannot occur without a water supply and Cobar is a dry town. Its main source of water is the Burrendong dam which caters for the greater Cobar region. The dam level fell to below 3% capacity in 2007 and there was not enough water to support the town, agriculture and mining. Priority was given to the town and agriculture and CSA was required to find alternative water sourced by drilling for water, retreating water and trucking water to the site.

19    The learned primary judge also made unchallenged findings about the copper concentrate market and the pricing of both concentrate and of copper. Relevantly they were as follows:

(a)    Copper concentrate comes from mining sulphide and oxide ore. The copper content of such concentrate is highly variable (from around 10 to 50%). It may be present with other valuable minerals, such as gold and silver, as well as with impurities. Smelters refine the concentrate to produce pure copper.

(b)    Miners may sell copper concentrate to either smelters or to traders. Traders are intermediaries who buy concentrate from miners and sell it to smelters. A miner may enter into a long term contract with a trader to sell specified quantities of its concentrate or the whole of a particular mine’s production. Such contracts are known as “offtake agreements.” In this way, a miner is able to “monetize its production.” This is what C.M.P.L. did. It sold all of its copper concentrate from the C.S.A. mine to G.I.A.G. Though it was not the subject of an express finding in her Honour’s reasons, it was not in dispute that, until August 2003, G.I.A.G. sold all of this concentrate to a smelter at Port Kembla in New South Wales. But the smelter closed. Thereafter, G.I.A.G. sold C.M.P.L.’s copper concentrate to a range of smelters located in China, Japan, Korea, the Philippines and India. A miner or a trader may also sell copper concentrate on the spot market, which accounts for about 20% of all copper concentrate trading.

(c)    Whilst the copper industry has developed a pricing structure for its contracts which follows a reasonably well-established framework, each contract is nonetheless individually negotiated and the detail and pricing of individual contracts varies greatly. Critically, her Honour accepted that such variation will depend “on the needs and risk appetites of the seller and buyer and whether the contracts are for spot sales or medium/long-term contracts” (at [72]).

(d)    The “well-established” pricing framework was as follows:

(i)    Unlike pure copper, there is no terminal market for copper concentrate where it can be traded for a given price.

(ii)    Nonetheless, the price of refined copper as quoted on a metal exchange, such as the London Metal Exchange (“L.M.E.”), is the starting point for determining the price of copper concentrate. The price of pure copper is volatile. In 2006, being the year leading up to the C.M.P.L.-G.I.A.G. agreement, the copper price reached unheard of heights. This is illustrated by the following graph, extracted from a report prepared by the taxpayer’s expert, Mr. Wilson.

(iii)    Offtake agreements specify the period, whether a day, days, weeks or months, with which the price or average price of the copper is to be ascertained. In the industry this is known as the “quotational period.” Some agreements give the buyer options about how to choose the applicable quotational periods and how often such choices can be made. We will call this “quotational period optionality.” These options are described in more detail below. The C.M.P.L.-G.I.A.G. agreement gave G.I.A.G. a number of such options. The Commissioner complains that they favoured G.I.A.G.’s position.

(iv)    A deduction is then made from the ascertained copper price notionally on account of the cost to a smelter of treating and refining the copper concentrate (called “T.C.R.C.s”). This permits the trader or smelter potentially to make a profit. The size of that deduction is a matter for negotiation and often bears little relationship to the actual costs of treating and refining the concentrate.

(v)    There are different ways of determining T.C.R.C.s. One way is to adopt benchmark T.C.R.C.s which are established annually by the major players in the copper concentrate industry. Ordinarily applicable to a given year, parties may use that benchmark or agree to negotiate a deduction for T.C.R.C.s based on that benchmark. Benchmark T.C.R.C.s are subject to significant variation from year to year. They are also not necessarily correlated with the price of refined copper. Importantly, the price of copper and the amount of the annual benchmark T.C.R.C.s can move in opposite directions. The following graph, extracted from the December 2006 Brook Hunt report, illustrates this phenomenon.

(vi)    The accuracy of the above graph was never impugned. The Brook Hunt report was an annual industry report which examined the global market for copper concentrate. It was used by players in the industry. Her Honour described the role of the Brook Hunt reports in the copper industry at [93] as follows:

It was common ground between the experts and, similarly, it was the evidence of Mr Kelly, that market participants, including miners, traders and smelters, access information in relation to the copper market and particularly the copper concentrate market from industry publications such as the Brook Hunt Reports (which Mr Wilson authored/edited for around 30 years). The Brook Hunt Report presents a global analysis of the market each year. Mr Kowal and Mr Ingelbinck [the Commissioner’s experts] both acknowledged that Brook Hunt is an authoritative resource and, whether accurate or not, influences negotiations in the copper concentrate market. Mr Kowal and Mr Ingelbinck each gave evidence that they relied upon Brook Hunt publications to support their opinions in these proceedings.

(vii)    Another method for determining the T.C.R.C.s was to use spot T.C.R.C.s, which reflect the market at any given point in time. Spot T.C.R.C.s are “very volatile and vary greatly throughout a year” (at [80]).

(viii)    Parties can use a mix of these methodologies to determine T.C.R.C.s. In the contract entered into between C.M.P.L. and G.I.A.G. for the sale of copper concentrate which was on foot just before February 2007, the parties agreed to negotiate the T.C.R.C.s annually in respect of 50% of the “Base Tonnage” based upon the Japanese benchmark T.C.R.C.s. For the other 50% of Base Tonnage, the negotiation was to be based on the spot market T.C.R.C.s. For any tonnage exceeding the Base Tonnage, the terms would be those that were “mutually agreed.” The Base Tonnage was defined to be 100,000 wet metric tonnes (“w.m.t.”).

(ix)    In the period leading up to 2007, being the first year in dispute, parties sometimes included a clause which permitted “price participation.” This permitted buyers and sellers to benefit from movements in the price of copper. For example, the parties might agree a particular L.M.E. copper price above which the refining charge would be increased by an agreed amount, and below which the refining charge would be reduced by an agreed amount. At the end of 2006, it was expected that price participation was unlikely to be used by buyers and sellers, or be set to zero.

(x)    A fundamentally different formula for determining the amount of the T.C.R.C.s was called “price sharing.” Under contracts using a price sharing formula, the T.C.R.C.s are a negotiated specified percentage of the metal exchange copper price. They are not re-negotiated annually but are agreed for a period of years. That period could be between three to 13 years in length. An advantage of this type of formula was that it eliminated a particular type of dangerous volatility arising from the use of annual benchmark T.C.R.C.s, namely an inverse movement in the price of copper and benchmark T.C.R.C.s which can occur from time to time. While the formula still exposed the miner to volatility in copper prices, when the price went down, the miner would enjoy lower T.C.R.C.s; when the price went up, the T.C.R.C.s would increase, but the miner would nevertheless benefit from the increased revenues resulting from the increase in the copper price.

(xi)    The taxpayer’s expert, Mr. Wilson, reported that in the December 2006 Brook Hunt report titled “Global Concentrate & Blister/Anode Markets to 2016”, it was stated that the “normal range” of price sharing is 21-26% of the copper price. The report went on to observe that price sharing was:

also reasonably equitable to both buyer and seller because when copper prices are low, so too are the fees received by the buyer and vice versa.

(xii)    The C.M.P.L.-G.I.A.G. agreement on foot in the 2007 to 2009 years included a term that adopted a price sharing formula at a rate of 23% of the copper price, being roughly the mid-point of the historical normal range of 21-26%. The Commissioner’s experts considered that the adoption in February 2007 of such a term was very much to the disadvantage of C.M.P.L. and was not what an independent seller would have agreed to.

(xiii)    Other terms in a typical copper concentrate sale agreement further adjusted, or had the potential to adjust, directly or indirectly, the consideration receivable by the miner. These included terms prescribing penalties for the presence of deleterious elements in the copper concentrate, and terms governing freight allowances and insurance costs. As a matter of industry practice, the seller paid, directly or indirectly, for freight and insurance. In each case, the actual amount was a matter for negotiation and the agreed result might not bear much of a relationship with the actual costs of shipping and insurance. The parties could sell/buy copper concentrate on a carriage insurance and freight (“C.I.F.”) basis, whereby the seller would be responsible for arranging and paying for freight and insurance. Alternatively, the parties could sell/buy copper concentrate on a free-on-board (“F.O.B.”) basis, whereby the buyer would be responsible for shipping, but with the seller paying a freight allowance. C.M.P.L. and G.I.A.G., in the 2007 to 2009 years, sold copper concentrate on an F.O.B. basis with the allowance negotiated annually. For the 2009 year only, the Commissioner contends that the allowance negotiated was not an arm’s length allowance.

(xiv)    The foregoing mechanisms or formulae, when applied to a given shipment, resulted in a price for copper concentrate sold.

(e)    There were other terms that parties in the copper concentrate market typically negotiated but which less directly bore upon the pricing formulae set out above. One concerned the “payable metal”, which is the negotiated percentage of the metal content of the concentrate to which the copper reference price is to be applied. Another concerned the terms of payment. It is industry practice for a seller to receive a provisional payment. If the buyer is a trader (and not a smelter) such a payment is typically made after an agreed number of days following receipt of the bill of lading and other documents. The provisional payment gives the seller a cash flow advantage, at least to an extent.

(f)    The market conditions for the price of copper in 2006 were unique. The copper price had risen by more than 80% as compared to the previous year price, reaching a point that was four times the average copper price over the 1998 to 2003 period. This was unprecedented. One of the Commissioner’s experts observed that the 80% rise in copper prices in 2006 was “a somewhat unheard of event and caused all kinds of re‑evaluations amongst industry participants as to what made sense and how to structure these agreements ... people were confused”. At the start of 2007, the taxpayer’s expert considered that there would be a material decline in copper prices. The learned primary judge found that Brook Hunt’s estimates as at December 2006 for the L.M.E. price moving forward were that, for 2007, in money of day terms, the copper price would be around US$2.70/lb, in 2008 US$2/lb, in 2009 US$1.34/lb and in 2011 less than US$1/lb.

(g)    By the end of 2006, the Japanese benchmark T.C.R.C.s for the 2007 year were already known (they had been determined earlier in the year). The treatment charge was set at US$60/dry metric tonne (“d.m.t.”) and the refining charge US6.0c/lb (or US15.4c/lb expressed as a combined T.C.R.C.) for 30% concentrate, with price participation for benchmark T.C.R.C.s having been set at zero. At the same time, the benchmark T.C.R.C.s for 2008 and 2009 were unknown. They were nonetheless respectively forecast to be about US$65 and then US$60 per d.m.t. for treatment charges and US6.5c/lb and then US6.0c/lb for refining charges (or US16.7c/lb and US15.4c/lb expressed as combined T.C.R.C.s). All experts agreed that balances for copper concentrate in the 2007 to 2009 years were likely to be “tight” and there was an expectation that a significant increase in T.C.R.C.s was a relatively low probability. The taxpayer’s expert thought that it was highly unlikely that T.C.R.C.s would go above US$70 a tonne over these years.

(h)    Spot T.C.R.C.s were unknown for 2007. All experts agreed that spot pricing was very volatile and inherently difficult to predict.

(i)    C.M.P.L.’s 2007 budget and five year plan (the “2007 Budget”) was prepared in September 2006. It predicted that C.M.P.L. would realise a significant profit in 2007. It forecast a copper price of US$2.54/lb for 2007 (Brook Hunt forecast a price of US$2.70/lb). This was noted to be a conservative forecast. The 2007 Budget stated that adequate ventilation for the mine shafts and staff retention were key issues for the C.S.A. mine. The learned primary judge also accepted that the mine was at this time facing water supply issues. Cobar was in a drought. Nonetheless, output was increasing. The production of ore had increased in 2006 by over 30%. Even though operating costs had exceeded budgeted costs in 2006 by 31%, net profit and revenue far exceeded budget estimates, due to the increase in copper prices. The 2007 Budget forecasted that cost control would be an important focus of the business in 2007. It also forecast that the amount of copper concentrate would decline consistently with an expected reduction in feed grade.

(j)    Budgeted T.C.R.C.s for the 2006 year were 31.8% of forecast revenue. They ended up being only 14.9% of actual revenue. Budgeted T.C.R.C.s for the 2007 year, assuming the use of benchmark T.C.R.C.s, were US16.7c/lb (or about 6.5% of the forecast copper price). The 2007 Budget predicted for the 2008 year benchmark T.C.R.C.s of US21.8c/lb (or about 9.6% of the forecast copper price) and for the 2009 year benchmark T.C.R.C.s of US19.2c/lb (or about 9.4% of the forecast copper price). It also predicted a copper price of US$2.27/lb for the 2008 year and US$2.04/lb for the 2009 year. These were accepted to be “reasonable figures” by the learned primary judge.

(k)    It was expensive to mine ore at the C.S.A. mine. The learned primary judge found that it was a “high cost” mine. Based on the actual costs for the 2006 year, Brook Hunt placed the mine at the 89th/90th percentile of “C1” costs, being the direct cash costs of the mine. The taxpayer’s expert gave uncontradicted evidence that the fact that the C.S.A. mine was a high cost mine made C.M.P.L. vulnerable to fluctuations in gross revenue for its copper concentrate.

The International Agreements between C.M.P.L. and G.I.A.G.

20    The terms upon which C.M.P.L. sold copper concentrate to G.I.A.G. varied over time. However, and critically, it was always agreed that G.I.A.G. would purchase all of C.M.P.L.’s production of copper concentrate for the life of the C.S.A. mine, subject to an important qualification described below. The first offtake agreement entered into in 1999 required the parties to negotiate the T.C.R.C.s based upon the Japanese benchmark T.C.R.C.s, and specified only one quotational period. The learned primary judge summarised its terms as follows at [162]:

(a)    CMPL agreed to sell and deliver, and GIAG agreed to buy and take delivery of, the entire production of copper concentrate from the CSA mine for the life of the mine starting from 1 July 1999;

(b)    GIAG was to pay for 96% of the copper content of the copper concentrate (subject to a minimum deduction) at the official London Metal Exchange cash settlement quotation for Grade A copper as published in USD in the ‘Metal Bulletin’ in London, averaged over the relevant quotational period, less the TCRCs;

(c)    a typical analysis of the quality of the copper concentrate produced was required to have a copper content of approximately 30%;

(d)    there was one quotational period for material produced after 1 July 1999, which was the month following the month of production, with this single quotational period to remain valid for the life of the mine;

(e)    the amounts to be deducted as TCRCs in respect of copper concentrate for the period between 1 July 1999 and 30 June 2000 comprised a treatment charge, a refining charge that was subject to an adjustment for “price participation” under which the charge would increase if the copper price exceeded a certain amount and would decrease if the price fell beneath that amount, and a penalty to be deducted when the bismuth content of the concentrates exceeded a certain percentage;

(f)    the amounts to be deducted as the TCRCs in respect of copper concentrate and freight allowances for each delivery year after July 2000 were to be negotiated and agreed prior to the commencement of the corresponding shipment based on the existing Japanese benchmark; and

(g)    if GIAG opted to take FOB Newcastle, then GIAG would be entitled to a freight credit as if the copper concentrate had been sold Cost Insurance and Freight Free Out (“CIFFO”) Main Japanese Port, with the credit fixed for each contractual year.

21    This first offtake agreement was replaced with a further agreement dated 1 October 2004. Following the closure of the Port Kembla smelter, the parties agreed to give G.I.A.G. much greater flexibility in deciding how to choose the quotational period for determining the L.M.E. copper price. This was achieved in two ways. First, from January 2005 G.I.A.G. was given the option to choose one of two classes to be declared on an annual basis, with each class containing three quotational periods. Then, within each of those two classes, one of the three quotational period options was to be declared on a shipment by shipment basis by G.I.A.G. at a later time. Secondly, the 2004 agreement introduced “back pricing”. This enabled G.I.A.G. to have knowledge of the average copper prices in (at least) one of the periods from which it was to make its selection of one of three quotational periods to be declared prior to each shipment from the loading port. This 2004 agreement was not before us. But a succeeding agreement applicable for the 2006 year and following and dated 12 December 2005 was included in the appeal papers. It contained a quotational period optionality clause which gave G.I.A.G. similar options as described above as well as back pricing. It was relevantly in these terms:

9.    QUOTATIONAL PERIOD:

9.1    For copper:

The quotational period for copper shall be, in Buyer’s option, either as per item a) or as per item b) below, to be declared on an annual basis at the time of negotiation of the terms for the new Contractual Year:

a)        i)    The average of the calendar month prior to the month of shipment, or

   ii)    The average of the calendar month of shipment, or

iii)    The average of the first calendar month following the month of shipment

b)        i)    The average of the first calendar month following the month of arrival of the carrying vessel at discharge port, or

ii)    The average of the second calendar month following the month of arrival of the carrying vessel at discharge port, or

iii)    The average of the third calendar month following the month of arrival of the carrying vessel at discharge port.

In case Buyer elects alternative “a)”, then the exact month, i), ii) or iii), shall be declared latest on the last LME (London Metal Exchange) market day after closing time 6 PM London time of the month of shipment, which means latest on following working day Australian time.

In case Buyer elects alternative “b)”, then the exact month, i), ii) or iii), shall be declared latest on the last LME (London Metal Exchange) market day after closing time 6 PM London time of the second month following the month of arrival of the carrying vessel at discharge port, which means latest on following working day Australian time.

22    As foreshadowed above, in December 2005, the terms upon which C.M.P.L. sold copper concentrate to G.I.A.G. changed again. The basal methodology or formula for determining price was set out at cl. 8 as follows:

The price per dry metric ton of copper concentrates shall be the sum of the payments less the deductions as specified below:

23    The clause for determining the “payments” was found in cl. 8.1.1 as follows:

Pay 96% (ninety six percent) of the final copper content, subject to a minimum deduction of 1 (one) unit, at the official LME Cash Settlement Price for Copper Grade “A” in USD, as published in the London Metal Bulletin averaged over the Quotational Period.

24    The clause setting out the “Deductions”, or T.C.R.C.s, was cl. 8.2. New cls. 8.2.1 and 8.2.2 now provided for the negotiation of T.C.R.C.s annually based upon the Japanese benchmark T.C.R.C.s for 50% of the Base Tonnage, with the remaining 50% of Base Tonnage based upon the spot market T.C.R.C.s, and for any tonnage in excess of the Base Tonnage on terms to be mutually agreed. In the 2007 Budget, it was estimated that excess tonnage would be in the region of 45,000 tonnes for that year, 39,000 tonnes for 2008 and 39,000 tonnes for 2009. Clauses 8.2.1 and 8.2.2 were relevantly as follows:

8.2    Deductions:

8.2.1    Treatment Charge for copper

Treatment charge for each Contractual Year of shipment will be negotiated and agreed between Seller and Buyer in form of an Addendum prior to the commencement of the corresponding Contractual Year.

Both parties have agreed that the terms will be negotiated on the following basis:

-    50% of the Base tonnage basis the Japanese benchmark.

-    50% of the Base tonnage basis the spot market.

For any tonnage exceeding the Base tonnage, the terms shall be mutually agreed.

8.2.2    Refining Charges:

8.2.2.1    For copper:

Refining charge for each Contractual Year of shipment will be negotiated and agreed between Seller and Buyer in form of an Addendum prior to the commencement of the corresponding Contractual Year.

Both parties have agreed that the terms will be negotiated on the following basis:

-    50% of the Base tonnage basis the existing Japanese benchmark.

-    50% of the Base tonnage basis the spot market.

For any tonnage exceeding the Base tonnage, the terms shall be mutually agreed

25    In addition, there were terms for the calculation of penalties payable by C.M.P.L. and for the making of provisional payments by G.I.A.G.

26    From 2006, the applicable agreement also provided for delivery at G.I.A.G.’s option of either “CIF F.O. main port Japan, Korea, India or China,” or on the basis of “FOB S.T. Newcastle, Australia”, in which case a freight and insurance allowance was to be mutually agreed on an annual basis at the prevailing spot market rates to “main port Japan, Korea, India or China”. In other words, in each year the parties had to agree upon which spot market rate to use by reference to the cost of shipping to Japan, Korea, India or China in order to determine a freight and insurance allowance. The applicable clauses were relevantly as follows:

6.1    CIF F.O. main port Japan, Korea, India or China, in Buyer’s option (hereinafter called “CIF shipments”).

Discharge conditions, demurrage and despatch shall be as per Charter Party, subject to Buyer’s approval.

Vessel shall furthermore be suitable for draught survey as per conditions stated in the attached Appendix No. 2, performing a constituent part of this contract.

The copper concentrates shall be shipped in accordance with IMO Code of Safe Practice for Solid Bulk Cargoes.

6.2    Buyer shall have the option to take delivery of the concentrates on basis FOB S.T. Newcastle, Australia, (hereinafter called “FOB Shipments”).

A freight and insurance allowance, based on the outturn wmt, to cover the cost of shipment from FOB S.T. Newcastle to CIF F.O. main port Japan, Korea, India or China shall be mutually agreed on an annual basis between Buyer and Seller basis prevailing spot market rates to main port Japan, Korea, India or China and shall be credited to the Buyer, unless otherwise agreed.

100% (one hundred percent) of these freight and insurance discount will be deducted from Buyer’s provisional payment to Seller (as set out in clause 10. PAYMENT, item 10.1. Provisional Payment, below).

27    We should also observe that this agreement, and we believe the pre-existing agreements, was expressed to be one that would remain in force “until the date when production ceases permanently at the Cobar mine.” However, this was qualified by G.I.A.G.’s right to terminate the agreement with 12 months’ notice prior to the end of any “Contractual Year” (being a calendar year). Relevantly, this right did not apply to a year in which terms had already been agreed for the full year.

28    In February 2007, C.M.P.L. and G.I.A.G. agreed to amend their agreement in four relevant ways.

29    First, the parties abandoned their 50/50 reliance on the Japanese annual benchmark T.C.R.C.s and spot T.C.R.C.s for the Base Tonnage and on the mutual agreement process to determine T.C.R.C.s for any excess tonnage. They agreed to adopt instead a price sharing clause to determine the deduction to be made for each shipment, and to forgo price participation. For the 2007 to 2009 years, T.C.R.C.s were set at 23% of the copper price. The new clauses were as follows:

Sub−clause 8.2.1 Treatment Charge and

Sub−clause 8.2.2.1 Refining Charge for copper:

Only for the Contractual Year 2007, 2008 and 2009, delete this sub−clauses entirely and replace by the following:

Combined Treatment Charge and Copper Refining Charge shall be 23% (twenty three percent) of the copper price established as set out in clause 9. QUOTATIONAL PERIOD below for the payable copper content.

The calculation of the Treatment Charge and Copper Refining Charge shall be as follows:

(copper price for establishment of Treatment Charge and Copper Refining Charge) *(multiply) 23% (twenty three percent) *(multiply) (final payable copper content) = (equal) combined Treatment Charge and Copper Refining Charge per dmt.

Sub−clause 8.2.3 Price Participation:

Not applicable for Contractual Years 2007, 2008 and 2009.

30    Because G.I.A.G. and C.M.P.L. had agreed the formula for the determination of the price of copper concentrate for the 2007 to 2009 years, it was said that G.I.A.G. had, for that period, thereby lost its right of termination. This level of certainty was important to the taxpayer’s case.

31    Secondly, the clause dealing with quotational periods was amended again to give G.I.A.G. even greater optionality. G.I.A.G. could now make its election to choose either option a) or b) (as set out above) on a shipment by shipment basis as opposed to an annual basis only.

32    Thirdly, the clause dealing with the required quality of copper was replaced. Previously, the copper concentrate had to contain 30% copper. This was reduced to 28.5%.

33    Finally, it will be recalled that the freight allowance had to be agreed annually. In 2009, by a written addendum to the C.M.P.L.-G.I.A.G. agreement, it was agreed that this allowance should be fixed at US$60 “per outturn wmt.” It did not appear to be in dispute that this was the rate for shipping to India. The cost of shipping to India was considerably greater than the cost of shipping to China, Korea or Japan.

The Evidence Led by the Parties

34    The taxpayer relevantly relied on one lay witness and on the opinions of one expert witness. The Commissioner relevantly relied on the opinion of two expert witnesses, but before us really relied on only one. Other expert evidence was led by the parties, but was found by the learned primary judge to be irrelevant; that finding was not challenged on appeal.

35    Perhaps the most significant feature of the taxpayer’s lay evidence is that the taxpayer did not call any person to give evidence about the entry into the various agreements, and amended agreements, referred to above. In particular, no evidence was led about why, as a fact, G.I.A.G. became entitled to increased quotational period optionality and why C.M.P.L. and G.I.A.G. had changed their formula for determining the T.C.R.C.s in 2007 from a part benchmark and part spot formula to a price sharing formula. No one gave evidence about how the 23% rate was determined. No one gave evidence about C.M.P.L.’s appetite for risk taking in 2007, even though, as will be seen, the taxpayer’s case turned upon how the issue of risk might bear upon the ascertainment of an arm’s length price. Nor was any evidence led of any actual negotiations between G.I.A.G. and C.M.P.L., although it would be strange if such negotiations had ever meaningfully taken place. C.M.P.L., indirectly, is a wholly owned subsidiary of G.I.A.G. The relevant provisions here effectively require the payment of an arm’s length price to C.M.P.L. But they do not require the parties otherwise to pretend that they are at arm’s length.

36    Instead, the taxpayer called Mr. David Kelly. He was an accountant who, prior to July 2006, had been employed at Deloitte. From July 2006 to April 2009 he was employed by G.I.A.G. (not C.M.P.L.) as the C.S.A. mine’s asset manager. From April 2009, he traded commodities for G.I.A.G. which included C.M.P.L.’s copper concentrate. From November 2007 until October 2017 he was also a director of C.M.P.L. He gave general evidence about the C.S.A. mine; the logistics and risks for C.M.P.L. in selling its product from 2006 to 2009 if it had been independent of G.I.A.G.; and he exhibited a number of offtake agreements which the taxpayer said contained comparable terms which had been used by arm’s length counterparties.

37    The Commissioner submitted that Mr. Kelly was not independent, had overstated his evidence, and was not qualified to give evidence about a number of matters. On occasion he tended to give inadmissible opinion evidence. The learned primary judge accepted some of these criticisms, but concluded that the case did not turn upon “the contentious parts of Mr Kelly’s affidavit evidence.” As best as we can tell, no part of the taxpayer’s case on appeal appeared to rely upon these contentious bits of his evidence.

38    The taxpayer’s expert witness was Mr. Richard Wilson. He graduated in 1981 from the Royal School of Mines, Imperial College London with an honours degree in Mineral Technology. He is an Associate of the Royal School of Mines. He had worked for Brook Hunt since 1982; for many years he wrote parts of the annual Brook Hunt report; he had been a director since 1988, and then Managing Director of that firm since 1996, and was the Chairman of Metals of Wood Mackenzie at the time of the trial below (Brook Hunt is now owned by Wood Mackenzie Ltd and has been since 2008).

39    The Commissioner took issue with the reliability of the opinions Mr. Wilson gave. He had never directly negotiated a contract for the sale of copper concentrate. He had never worked for a mine; he had never traded copper concentrate; and he had never worked for a smelter. But, it would appear, he had very extensive and detailed knowledge of the copper market that was regularly relied upon by direct participants in the market. The source of that knowledge was disclosed in his first report in the following terms:

Over the past 36 years I have spoken at numerous conferences around the world on topics related to the copper market and the economics of copper supply. My views on the copper industry are also regularly quoted in the press and in other media journals. I attach some of the most recent references to my work, along with conferences that I have attended, in Appendix B of this report. I have also attached other references to Brook Hunt/Wood Mackenzie’s data from various industry publications in Appendix B.

Wood Mackenzie initially cultivated deep expertise in upstream oil and gas, producing its first research report in 1973. It has since broadened its focus to deliver the same level of detailed insight for each connected sector of the energy, chemicals, metals and mining industries. The company employs over 1000 people in over 30 offices worldwide, including three hubs in Australia. Expert analysts and consultants provide monthly, quarterly and annual insights across the value chains of metals including copper, lead, zinc, nickel, aluminium and gold. Wood Mackenzie has few direct competitors; these include Commodity Research Unit (CRU), Australian Mineral Economics (AME) and Bloomsbury Mineral Economics (BME).

Data produced by Brook Hunt/Wood Mackenzie has been and continues to be used extensively by our clients who regularly reference our information in their publications. For example, we publish so-called annual ‘Benchmark’ terms between concentrate buyers and sellers, described later in this report, and we are specifically named in some copper concentrate contracts as the reference source for this data.

In 1990 I initiated an annual research offering entitled ‘Global Outlook for Copper Concentrates and Blister/Anodes’ and was the author of this study until 2011 and the editor thereafter. Up until 2011, I was also responsible for writing the commentary on copper supply and concentrate/blister markets in Brook Hunt’s monthly copper publication entitled ‘Copper Metal Service’. Since then I have been the editor of this report. I remain in regular contact with producers, consumers and traders of copper concentrates and I am a recognised authority in this market. The aforementioned reports are available on a multi-client annual subscription basis. Clients that subscribe to the abovementioned reports include mining, smelting and trading companies, governments and financial institutions.

(Footnote omitted.)

40    G.I.A.G. has been a client of Brook Hunt; so too has the Commissioner of Taxation.

41    Mr. Wilson was asked to answer the following questions:

(a)    whether the conditions (individually or together) that operate under the 1999 Off-take agreement as amended from time to time and including as amended by the Third Replacement Concentrate Agreement (together with its Addendums and Amendment) as they applied in the 2007 to 2009 calendar years differ from those that might be expected to operate between a producer/seller of copper concentrate in the position of [CMPL] and a purchaser independent of CMPL having regard to the circumstances of CMPL in the period 1999 to February 2007; and

(b)    if they do,

   (i)    how such conditions differ; and

(ii)    what profits (if any) may have been expected to accrue to CMPL by reason of such differences?

42    Mr. Wilson examined the terms upon which C.M.P.L. sold copper concentrate to G.I.A.G. Using his extensive experience of the copper market, he formed the view that the adoption of the price sharing formula that was in fact adopted by C.M.P.L. and G.I.A.G. in February 2007 was commercial and prudent given the volatile nature of the copper market. Contrary to the Commissioner’s submission, Mr. Wilson’s opinion was not limited to issues of commerciality and prudence. He expressed his answer to the questions asked of him in his first report in the following way:

For the reasons set out later in this report, I am of the opinion that the conditions both individually and taken together that operate under the 1999 Off-take agreement as amended from time to time including as amended by the Third Replacement and Concentrate Agreement (together with its Addendums and amendment) as they applied in the 2007 to 2009 calendar years did not (with the exception of the payment terms) differ either in respect of their contractual terms or in respect of the expected financial impact of those terms from those conditions that might be expected to operate between a producer/seller of copper concentrate in the position of CMPL and a purchaser (including a purchaser having substantially the same characteristics as Glencore International AG (GIAG)) which is independent of CMPL.

(Footnote omitted.)

43    Mr. Wilson gave evidence that price sharing is a recognised way of pricing T.C.R.C.s in a market where both the price of copper and T.C.R.C.s are “highly volatile”, noting in particular that spot T.C.R.C.s were “virtually impossible to predict”. He estimated that, at its peak in 2008, the tonnage of copper concentrate that had been contracted globally under price sharing contracts was equivalent to about 9% of the seaborne trade in copper concentrates in the Asian smelting region. He gave his opinion that price sharing had advantages for both buyers and sellers. They existed whether or not a mine produced larger or smaller quantities of concentrate. They were as follows:

•    it allows for greater certainty of the projected cash flows for the mine/project as TCRCs are correlated with the income stream (i.e. the LME copper price);

•    it makes the annual negotiating process easier because conditions applied to price-sharing contracts are typically set for periods of 3-13 years and are not negotiated annually. Under some price-sharing contracts the price-sharing percentage is able to be varied in accordance with terms as set out in the original contract. Where a contract provides for such a variation it almost invariably results in an increase in the price-sharing percentage;

•    it makes the project economics easier to understand for providers of funds, if applicable, as part of the costs are determined by the copper price through the price-sharing mechanism. For the copper price, as opposed to the TCRC, there is a long history of data available for statistical analysis;

•    when a floor price on the price-sharing exists, it effectively guarantees the buyer a minimum price which can provide the impetus for the development of additional smelting capacity (which was necessary during the 1990s) to accommodate the extra concentrate producing capacity. Under the CMPL-GIAG price-sharing contract there was no floor price. This would have been a significant benefit to CMPL under a low or very low copper price scenario. Most price-sharing contracts that I am aware of have/had price-sharing floors, i.e. a minimum price for the buyer;

•    it removes volatility arising from the movement of the LME copper price and TCRCs in opposite directions.

To illustrate the volatility that is removed by price sharing, Mr. Wilson observed that in 1998, despite the approximate 25% drop in annual average copper prices, the T.C.R.C.s as a percentage of copper prices had increased from 27% to 33% for a relative increase of 21%.

44    In Mr. Wilson’s opinion there is no standard price sharing contract. He said some contracts have the following features:

(a)    a price-sharing range between defined LME price caps and/or floors;

(b)    percentages of price-sharing that vary according to the LME copper price;

(c)    a fixed percentage of price-sharing regardless of LME copper price;

(d)    annual price-sharing escalators linked to a US GNP adjustment.

45    As for the fixing of a percentage of the copper price, Mr. Wilson observed, as already mentioned, that the normal range was 21-26%. That range had not altered materially since the early 1990s. The lowest percentage he had ever seen in a price sharing contract was 20%. He then turned to consider the price sharing clause adopted by C.M.P.L. and G.I.A.G. in February 2007. In his opinion, it was reasonable and prudent for C.M.P.L. to have adopted the clause, as it was the operator of a high cost mine and the copper price was expected to undergo a steep decline after having achieved an exceptionally high price. His observation on the copper price was illustrated by the following graph, based as it was on Brook Hunt’s long term copper price forecast as at December 2006.

46    Notwithstanding the general expectation of a steep decline in prices, Mr. Wilson observed that “there was considerable uncertainty as to the extent and time over which prices would decline”. Nevertheless, using the forecast figures in C.M.P.L.’s 2007 Budget, the operation of the 23% price sharing clause meant that C.M.P.L.’s profits were still expected to be “healthy.”

47    In his subsequent report, Mr. Wilson identified all of the “unknowns” as at February 2007. These included future L.M.E. copper prices for the 2007 to 2009 years, benchmark T.C.R.C.s for 2008 and 2009 (those for 2007 were known by February 2007), spot T.C.R.C.s for the 2007 to 2009 years (Brook Hunt does not attempt to forecast spot prices as they are too volatile), and whether or not price participation would be re-introduced. In his commercial judgment, having regard to the history of the C.S.A. mine (it had previously closed down in 1998 due to poor profitability) and the mine’s high costs, an overriding consideration for a miner in the position of C.M.P.L. would have been to ensure that the C.S.A. mine remained a viable going concern.

48    Mr. Wilson opined that fixing the T.C.R.C.s at 23% eliminated C.M.P.L.’s exposure to annual volatility arising from a formula that depended on both benchmark and spot T.C.R.C.s. To illustrate the volatility, Mr. Wilson observed that in the space of just two years (2004 to 2006) the average annual L.M.E. copper price had risen by 135%, and the realised benchmark T.C.R.C.s had increased by 254%. Mr. Wilson reasoned:

The use of full price-sharing terms as set out in the CMPL-GIAG Contract in 2007 removed one of the two unknowns by eliminating the volatility of TCRCs. The absence of a price-sharing floor gave CMPL added protection as the buyer was not guaranteed a minimum share of the price. It also meant that CMPL would have known precisely at any time during this three-year contract what CMPL’s net price would have been for any given copper price. Put another way, by fixing the TCRC at 23% of the copper price CMPL would have known with a high level of certainty the breakeven copper price for the CSA mine in each of the three years through the fixing of the TCRC. In my opinion, this is particularly important for a high cost mine such as CMPL which can quickly move into a loss position as copper prices decline (as was anticipated by market commentators in late 2006 and as shown by the LME forward price curve for copper (see figure at paragraph 38 of my first report)). Although it is apparent in hindsight that copper prices did not decline significantly over the three year period, benchmark TCRCs did not increase significantly, and price participation was not reintroduced, the use of the price-sharing arrangement was, in my opinion, a conservative and reasonable approach to minimising risk. If copper prices had fallen significantly over the three year period and TCRCs had increased, CMPL could have endured serious financial difficulties.

Critically, it was Mr. Wilson’s opinion that the adoption of a price sharing clause such as that adopted by C.M.P.L. in February 2007 was ultimately “a matter of commercial judgment having regard to the particular risk appetite and cost pressures facing a particular mine.”

49    Mr. Wilson’s conclusions about commerciality and prudence were criticised by the Commissioner as falling short of the statutory tests set out in Div. 13 and Subdiv. 815-A. It followed, it was said, that the taxpayer had failed to discharge its onus of proof. We respectfully disagree with that criticism. Mr. Wilson gave his opinion which sufficiently addressed the statutory test in Subdiv. 815-A (as reflected in the question asked of him) which we have already set out above. Because the case presented by both parties did not distinguish between Div. 13 and Subdiv. 815-A, it did not matter that he was not also asked a question which more directly engaged the statutory test in Div. 13. In any event, conclusions about commerciality and prudence are, in our view, conclusions about what parties dealing at arm’s length in the copper concentrate market might be expected to agree to in a contract of sale. As Allsop C.J. observed in Chevron Australia Holdings Pty Ltd v. Federal Commissioner of Taxation (2017) 251 F.C.R. 40 at 50-51 [42]:

[T]he ultimate purpose is to determine the consideration that would have been given (that is implicitly, by the taxpayer) had there not been a lack of independence in the transaction. How one comes to that assessment and the relationship between the posited arm’s length dealing and what in fact occurred will depend on the circumstances at hand, and a judgment as to the most appropriate, rational and commercially practical way of approaching the task consistently with the words of the statutory provision, on the evidence available.

(Our emphasis.)

50    Mr. Wilson’s opinion about the use of price sharing terms in contracts entered into between arm’s length parties was, in his view, supported by other contracts in the market that he was aware of. For example, in his first report he said the following:

I am aware of other price-sharing contracts involving Australian miners in the period prior to 2007. Northparkes (owned at the time by Rio Tinto and a Sumitomo consortium) had contracts with Japanese smelters between 1995 and 2002, with price-sharing rising from 21% initially, to 23.5% in the final year. Osborne (owned then by Placer Pacific) also had contracts with Japanese smelters, between 1996 and 2000, with price-sharing rising from 22.5% to 23.8% over the life of the contract.

As regards other price-sharing contracts, a further price-sharing contract that I have seen a copy of, is between Tritton Resources Ltd (Tritton), now owned by Aeris Resources, and trading company, Sempra. This contract was between unrelated parties and is similar to the CMPL-GIAG agreement in terms of being an Australian miner-trader contract with a price-sharing mechanism.

The Tritton contract was more complex than the CMPL-GIAG arrangement, with the percentage of price accruing to the buyer varying according to copper price. The contract had a minimum price sharing of 21%, increasing up to 27.5% with specified increases in the copper price. Ultimately Tritton was unable to fully perform the Sempra contract and renegotiated the terms. I do not believe that the failure of Tritton to perform the original contract was due to the price-sharing element of the contract.

(Footnotes omitted.)

51    Mr. Wilson also considered the quotational period optionality clause in the C.M.P.L.-G.I.A.G. agreement. He said that such clauses were recognised in the copper industry where the price of copper can vary significantly from month to month. In his view, some sellers required a financial inducement for the conferral of optionality on the buyer. Importantly, in his view, not all did so. Optionality was especially attractive to traders so that they could match the periods for purchases and deliveries of copper products, or hedge differentials in the periods used in purchase and sales contracts. Mr. Wilson examined a number of contracts for the sale of copper concentrate between independent parties and summarised the different selections of the quotational periods used in them. In his opinion, had C.M.P.L. sold its copper concentrate to independent smelters or traders, it would have been exposed to multiple quotational periods. In Mr. Wilson’s opinion, the move in 2004 to multiple quotational periods was a logical choice following the closure of the Port Kembla smelter in August 2003. Following this, G.I.A.G., instead of selling all of the C.S.A. mine’s copper concentrate to one smelter, became exposed to the risk of multiple smelter contracts using different quotational periods.

52    Mr. Wilson also considered the back pricing aspect of the C.M.P.L.-G.I.A.G. agreement. Again, he referred to a number of contracts entered into between independent parties where back pricing was used. He included in his report the following table of such contracts.

53    He also stated that based on his previous discussions with traders, back pricing was “common” in contracts used by “Trafigura, Louis Dreyfus, Gerald Metals, MRI, Ocean Partners, Sempra, Transamine, Traxys etc.” Whilst both quotational period optionality and back pricing were thought to be desirable for traders, in Mr. Wilson’s view the dollar value of such a benefit was “virtually impossible to predict in advance and is simply the “price paid” by the producer for the significant benefit of having a guaranteed purchaser for all of its production.”

54    In Mr. Wilson’s opinion, one clause in the C.M.P.L.-G.I.A.G. agreement differed from that usually found in copper concentrate contracts. G.I.A.G. was obliged to pay 100% of the provisional value of any concentrate produced but not shipped 14 days after the end of each production month. This gave C.M.P.L. a benefit; it gave it reliable cash flow and interest free working capital.

55    The Commissioner’s first expert was Mr. Marc Ingelbinck, who had “retired from corporate life at the end of 2010,” and was a self-employed consultant at the time of the trial. He has had extensive experience in relation to mining and trading in non-ferrous metals. From 2002 to 2010 he had been the Vice President, Marketing and Trading for B.H.P.’s Base Metals Customer Sector Group (our references to “B.H.P.” should be read as references generally to the dual listed company arrangement that exists between B.H.P. Group Limited and B.H.P. Group p.l.c., as those entities are now known). He had overall responsibility for the marketing of all of B.H.P.’s concentrate. This included copper concentrate. B.H.P. was the single largest supplier of custom copper concentrate in the market. As such, it was often the driver of benchmark T.C.R.C.s.

56    In his first report, Mr. Ingelbinck was asked to answer the following questions:

(1)    For the [2007 to 2009 years], do the conditions that were operating between CMPL and GIAG in their commercial and financial relations, differ from the conditions which might have been expected to operate between an Independent Producer/Seller and Independent Buyer dealing wholly independently with one another? If so:

   (a)    how and to what extent do such conditions differ; and

(b)    what profits (if any) by reason of these conditions might reasonably have been expected to have accrued to CMPL but for those conditions?

(2)    For the [2007 to 2009 years], is the consideration received or receivable by CMPL for the sale of Cobar copper concentrate to GIAG different from the consideration that might reasonably have been expected to have been received or receivable for the sale of Cobar copper concentrate by an Independent Producer/Seller to an Independent Buyer dealing wholly independently with one another? If so:

   (a)    how and to what extent; and

(b)    what consideration might reasonably have been expected to have been received or receivable by an Independent Producer/Seller but for that difference?

(3)    Please answer the above questions in the alternative and assuming that the Independent Producer/Seller was the operator of the Cobar Mine in the period 1 January 1999 to 31 December 2009 and:

(a)    operated in a commercial context as a wholly owned subsidiary of a multinational global resources company in the same or similar circumstances to that of CMPL and GIAG; or

(b)    for the period between 1 January 2007 and 31 December 2009 operated as a stand‑alone producer/seller of Cobar copper concentrate.

57    He was also asked, more generally, to consider Mr. Wilson’s expert reports and to comment on them.

58    Mr. Ingelbinck reviewed the terms upon which C.M.P.L. had historically sold copper concentrate to G.I.A.G. He was of the view that the initial 1999 offtake agreement had been generally structured as a traditional benchmark type contract. His focus, however, was on the amendments made to the terms of that agreement in 2004, 2005 and then in 2007. In his view, an independent seller would not have agreed to those amendments as they uniformly favoured the position of the buyer to the detriment of the seller. His views remained constant even where the independent seller was assumed to be the operator of the C.S.A. mine, either in a commercial context as a wholly-owned subsidiary of a multinational global resources company in the same or similar circumstances to that of C.M.P.L. and G.I.A.G. (Question 3(a)), or as a stand-alone seller of copper concentrate from the C.S.A. mine (Question 3(b)).

59    More particularly, Mr. Ingelbinck had never seen in long-term benchmark contracts a quotational period optionality clause as “liberal” as that enjoyed by G.I.A.G. An independent party would not have agreed to such an amendment as there was “zero quid pro quo granted to CMPL for the introduction of this optionality despite the fact that it materially improved the commercial position of GIAG.” Optionality, he said, equals value; the greater the optionality the greater the value. Moreover, by the end of 2006, in his view the market was aware that benchmark T.C.R.C.s were falling. The benchmark set for 2007 for 30% concentrate, which was known by the end of 2006, was US15.4c/lb. That being so, the 23% price sharing clause, based on projected copper prices, would have resulted in higher T.C.R.C.s: it would have resulted in T.C.R.C.s which were four times higher than the 2007 benchmark T.C.R.C.s (a similar observation could not be made about the 2008 and 2009 benchmark T.C.R.C.s as these were unknown in early 2007; we also note that Mr. Ingelbinck’s observation loses force when one takes into account the volatility of annual T.C.R.C. benchmarks). There was “zero justification” for such a change, which only made sense if the copper price were to average US67c/lb over the 2007 to 2009 years. But the “likelihood of this happening was essentially non-existent given the cost pressures faced by copper mines.” In contrast, Mr. Wilson was of the view that a miner in 2006 could reasonably have been somewhat sceptical about the accuracy of forecasting copper prices. We shall return to this issue.

60    Mr. Ingelbinck strongly disagreed with Mr. Wilson’s opinion that the terms upon which C.M.P.L. sold copper concentrate to G.I.A.G. were reasonable and prudent. In his first report he gave the following reasons for that opinion:

(a)    the copper price had risen significantly over the course of the year – the LME forward curve on 1 January 2007 indicated a then prevailing value of at least $6,100/MT or approximately $2.75/lb for QP’s associated with 2007 shipments – I do note that Brook Hunt’s December 2006 copper price forecast projected lower figures than the above and that LME forward curve valuations can fluctuate on a daily basis but the value of projected 2007 shipment QP’s remained high during the late 2006 period

(b)    2007 benchmark discussions were well advanced with a consensus view that the outcome would be $60/dmt and 6.0 c/lb or 15.4 c/lb for 30% concentrate

(c)    the concentrate market had tightened significantly and market observers such as Brook Hunt were forecasting such conditions would continue to exist for a number of years.

(Footnotes omitted.)

61    In Mr. Ingelbinck’s opinion, the switch to price sharing was “irresponsible” (emphasis in original). An independent seller in the position of C.M.P.L., when offered price sharing, should have answered with a simple “thanks but no, thanks.” In cross-examination, he said that if arm’s length parties in the circumstances of C.M.P.L. and G.I.A.G. had wanted to adopt price sharing in early 2007, a starting point might have been a rate of 8.1%, being the forecast average T.C.R.C.s as a percentage of the copper price for the 2007 to 2009 years (as extracted from the December 2006 Brook Hunt report). This was a “far cry” from the rate of 23%. He accepted, however, that no buyer would have agreed to purchase copper concentrate with price sharing fixed at a rate of 8.1% of the copper price.

62    In his cross-examination, Mr. Ingelbinck maintained his position that C.M.P.L. should not have agreed to the change to price sharing. His focus, as we understood it, was not upon whether the terms and conditions upon which copper concentrate was sold by C.M.P.L. to G.I.A.G. from 2007 were of an arm’s length nature, but rather, whether given the pre-existing terms and conditions it made sense to agree to the amendments made in February 2007. This can be seen in the following answer given by him in cross-examination:

And it is also a situation where you are not actually looking at establishing a contract, you are being asked to make a change from an existing contract so, under those circumstances, I would expect any commercial team to do a fairly serious analysis of, “Okay, so we can stay with what we have and what are the implications of that?” Making certain assumptions over a relatively short period of time, and then looking at the implications of making the switch to the suggested new pricing approach and evaluating what, under those same circumstances, the result is on your revenues.

63    Mr. Ingelbinck characterised price sharing as a “bet”. C.M.P.L. was betting in 2007 that in 2008 and 2009 the benchmark T.C.R.C.s would exceed 23% of the copper price. The last time this had occurred was in 2001. The force of that observation was diminished, in our view, by the sheer unpredictable volatility in copper prices at the time, and in particular following the unprecedented copper price increase in 2006 which resulted in confusion amongst copper industry participants. In the meantime, Mr. Ingelbinck observed that the benchmark T.C.R.C. rates from 2002 to 2006 had been 21.9%, 17.3%, 10%, 17.7% and 15% respectively. But comparing the use of benchmark T.C.R.C.s as against the use of a price sharing clause assumes that benchmark T.C.R.C.s would have been adopted by C.M.P.L. and G.I.A.G. in 2007 as the only means of determining T.C.R.C.s annually. In the past, this had not been the case. Prior to February 2007, benchmark T.C.R.C.s were used in the C.M.P.L.-G.I.A.G. agreement as a basis for negotiating only 50% of the Base Tonnage (and the remaining 50% used spot market T.C.R.C.s as a basis). The C.M.P.L.-G.I.A.G. agreement was silent on their use as a basis for negotiating sales in excess of the Base Tonnage. Significantly, as will be seen, the Commissioner contends that this pre-existing pricing formula was arm’s length in nature.

64    Mr. Wilson had referred in his evidence to a contract for the sale of copper concentrate entered into by B.H.P. Billiton Marketing A.G. and B.H.P. Billiton Tintaya S.A. dated 16 December 2003 (the “B.M.A.G.-Tintaya Contract”). Mr. Ingelbinck, who had worked for B.H.P., had been involved with this contract. It was not an arm’s length contract, but a contract entered into by two B.H.P. subsidiaries. Consistently with its transfer pricing obligations, B.H.P. had sought to determine an arm’s length price for this contract. It chose a price sharing formula using a rate of 25.5%. That exceeds the rate used by C.M.P.L. and G.I.A.G. Mr. Ingelbinck sought to explain why this rate had been chosen as follows:

The agreement was formalized on 16 December 2003 but had been under discussion for quite some time before that. Copper prices averaged $0.743/lb in the 2001-2003 period (a situation which had triggered a temporary curtailment of the Tintaya copper concentrate operations). We felt that given that background, a one third price sharing component at 25.5% was reasonable at that time.

65    Somewhat defensively, he then said:

I am not aware of the final outcome but I do recall that the results of the Tintaya-BMAG distribution agreement ended up being challenged by the Peruvian tax authorities.

66    In relation to the contracts relied upon by the taxpayer in support of the price sharing clause adopted by C.M.P.L. and G.I.A.G., Mr. Ingelbinck was otherwise generally of the opinion that they were not comparable either because the tonnage involved was too small or because the contracts had not been entered into contemporaneously with early 2007.

67    Mr. Ingelbinck accepted that price sharing could deliver benefits to the seller, but, as we understood his reports, these benefits were outweighed by the profits C.M.P.L. might have earned using benchmark T.C.R.C.s instead. As he said in his first report:

I guess one can argue that from a miner’s perspective:

 (a)    avoiding the need to participate in contentious annual negotiations

(b)    avoiding the potentially lethal combination of low copper prices and high TC/RC’s and

(c)    reducing the number of major budget variables one can get wrong from two to one

could be viewed as positives. Yet this very same approach could (and did in 2007/8/9) result in the mine paying significantly higher TC/RC charges than benchmark outcomes.

68    Mr. Ingelbinck also observed that a three year term was unusually short for a price sharing contract. Price sharing terms were said to be normally agreed for contracts of much greater length.

69    Mr. Ingelbinck was firmly of the view that the amendments made from 2004 to give G.I.A.G. greater optionality in choosing applicable quotational periods only favoured G.I.A.G. and would not have been agreed to if C.M.P.L. had been an arm’s length seller. He said the following:

The 1999 off-take agreement is generally structured as a traditional benchmark type contract (clause 9.2 of the 1999 off-take agreement) and a review of the contractual documents indicates the 1999 off-take agreement was indeed executed accordingly in the initial years (1999-2003). However, starting in 2004, a number of commercial term amendments were introduced which were neither triggered nor supported by the prevailing industry benchmark structures. Independent Buyers and Sellers sometimes agree to introduce modifications to their contractual agreement when both parties are comfortable that such changes further their interests or, at the very least, do not damage their interests. The 1 October 2004, 1 January 2005 and 2 February 2007 contract amendments/replacements fail this criterion as they were uniformly beneficial to GIAG and detrimental to CMPL and inconsistent with then prevailing benchmark terms to which CMPL was entitled as per the 1999 off-take agreement. The deviation from benchmark terms is shown in more detail in paragraph 28. An Independent Seller in similar circumstances to those of CMPL would in my opinion not have entertained such amendments and would have, if the Buyer continued to insist on introducing the amendments, referred the matter to an Arbitration Panel or a Court (depending on the relevant contract wording) for resolution where they would almost certainly have prevailed. In my opinion, CMPL would also have done so had it been independent. In this regard I note that the CMPL – GIAG contract does not contain a dispute resolution clause which in my experience is a common feature in contracts between Independent Market Participants. The 2004 amendment provided GIAG with broad QP optionality (including significant back-pricing privileges). No such developments took place in the broader benchmark market (which CMPL was a part of as per the 5 July 1999 contract). In 2005, the parties agreed to amend the pricing basis from 100% benchmark to 50% benchmark and 50% spot market. Again, this was not reflective of what happened in the broader benchmark market segment. And finally, […] the parties agreed to shift the (combined) TC/RC level to 23% of the prevailing copper price for 2007/8 and 9 in addition to further broadening the QP optionality introduced in 2004.

70    The only contract which contained a quotational period optionality clause which was materially identical to that adopted by C.M.P.L. and G.I.A.G. was one entered into between G.I.A.G. and Barminco Investments Pty Ltd dated 1 July 2004 (the “Barminco Contract”). Mr. Ingelbinck stated in his second report that Barminco was an “inexperienced seller.” In cross-examination, he conceded that he had no basis to make that observation. In any event, he observed that, unlike the C.M.P.L-G.I.A.G. agreement, the Barminco Contract expressly included a quid pro quo for optionality in the form of discounted T.C.R.C.s. As to the other contracts relied upon by the taxpayer, Mr. Ingelbinck summarised his views as follows:

With regard to Quotational Period optionality, most of the contracts submitted incorporate a degree of optionality but its scope and whether it incorporates back-pricing privileges and whether the grantor of the optionality receives compensation for same varies widely. From personal experience, I can say that the large long-term mine to smelter contracts that form the basis of the benchmark definition and represent a major portion of the overall custom concentrate market do NOT incorporate QP optionality or back-pricing privileges.

71    Mr. Ingelbinck made similar observations about back pricing. This should not have been introduced without the conferral of some quid pro quo that favoured C.M.P.L. In his view, back pricing was “far more likely to have a discernible negative impact on the seller’s revenue.” Those contracts which used back pricing, and which were said by the taxpayer to be comparable, he considered to be distinguishable. Again, they related to amounts of copper concentrate that were either too small or were struck at a different time.

72    Mr. Ingelbinck said that if benchmark T.C.R.C. pricing had been adhered to, he would have expected arm’s length parties to have agreed upon the following terms:

a)    refrain from introducing the QP [quotational period] optionality in 2004 and from expanding it in 2007

b)    settle the calendar 2005 TC/RC’s for all tonnage at $85.5/dmt [dry metric tonne] and 8.55 c/lb with PP [price participation] plus or minus 10% at 90 c/lb

c)    settle the calendar 2006 TC/RC’s for all tonnage at $95/dmt and 9.5 c/lb with PP plus or minus 10% at 90 c/lb

d)    settle the calendar 2007 TC/RC’s for all tonnage at $60/dmt and 6.0 c/lb, zero PP

e)    settle the calendar 2008 TC/RC’s for all tonnage at $45/dmt and 4.5 c/lb, zero PP

f)    settle the calendar 2009 TC/RC’s for all tonnage at $75/dmt and 7.5 c/lb, zero PP.

73    The foregoing represents a very different contract from the C.M.P.L.–G.I.A.G. agreement on foot just before February 2007, and which the Commissioner considered to be arm’s length.

74    More specifically with respect to C.M.P.L., Mr. Ingelbinck was of the opinion that the following should have been agreed in 2007:

If indeed CMPL management was concerned about a potential catastrophic decline in copper prices, the more appropriate course of action would have been to either

(a)    insist on benchmark terms for 2007 and separately implement a copper price protection program (sell LME forward copper positions and/or buy copper put options)

(b)    entertain the concept of a switch to price sharing but structure it based on the then prevailing ratio between copper prices and the 2007 benchmark (ie 5.6 percent using the 1 January 2007 forward curve figures) rather than on the by then totally unrealistic level of 23% and separately implement a copper price protection program).

In this regard, I do note that the 2004 introduction of extensive buyer’s QP optionality significantly impaired CMPL's ability to devise and implement fully functional hedging programs.

75    Again, this is very different from the pre-existing C.M.P.L.-G.I.A.G. agreement. Yet it is the pre-existing C.M.P.L.-G.I.A.G. agreement as it was just prior to February 2007 upon which the Commissioner’s case stands. That agreement incorporates the amendments of 2004 (and 2005) criticised by Mr. Ingelbinck.

76    Mr. Ingelbinck was also of the view that the freight allowance clause exposed C.M.P.L. to the risk of higher uncompensated freight costs and would not have been agreed to if the parties had been dealing with each other on an arm’s length basis.

77    Mr. Ingelbinck quantified the impact of the switch to price sharing, amongst other things, as being lost revenue for C.M.P.L. in excess of US$59.4 million in 2007, US$61.4 million in 2008 and US$36.4 million in 2009, though it is not clear to us how these amounts were calculated. These figures were said to be based on certain assumptions, such as the sale of a constant 150,000 d.m.t. per year, a copper price of US$2.90/lb that was derived by averaging the actual copper prices for the 2007 to 2009 years, and the use only of the actual benchmark T.C.R.C.s applicable in the 2007 to 2009 years. In other words, Mr. Ingelbinck used an assumption about production and the benefit of hindsight to calculate foregone revenue rather than forecast figures known in February 2007. As noted in Mr. Wilson’s second report, the only actual figures that C.M.P.L. and G.I.A.G. would have known as at February 2007 were the 2007 benchmark T.C.R.C.s. Mr. Ingelbinck’s supplementary report did not respond to this.

78    At this point, we should make a number of concluding observations about Mr. Ingelbinck’s reports.

79    First, we note that Mr. Ingelbinck compared the 23% price sharing clause with a formula for determining T.C.R.C.s which was wholly based on benchmark T.C.R.C.s. It was on this basis that he concluded that the switch to price sharing was “irresponsible”. But the agreement between C.M.P.L. and G.I.A.G. which existed just before February 2007, upon which the Commissioner’s case stood, was not one which only used benchmark T.C.R.C.s. Only 50% of the Base Tonnage was referable to benchmark T.C.R.C.s, and, it will be recalled, that benchmark was only to be used as a “basis” for negotiating the relevant T.C.R.C.s. T.C.R.C.s for the remaining 50% of Base Tonnage were to be set using spot market T.C.R.C.s as a “basis”, with the T.C.R.C.s for any excess over Base Tonnage to be mutually agreed without any contractually mandated reference point. As noted above, significant excess tonnage was expected in each of the 2007 to 2009 years. Thus, C.M.P.L. was significantly exposed, under this version of its agreement with G.I.A.G., to an annual re-negotiation of T.C.R.C.s.

80    In that respect, some of Mr. Ingelbinck’s comments appeared to be based upon an incorrect assumption that the C.M.P.L.-G.I.A.G. agreement as it was just before February 2007 used only benchmark T.C.R.C. pricing. For example, in his first report, Mr. Ingelbinck said:

As it relates to the specific CMPL - GIAG arrangement, the question arises for considering 1(a) whether Independent Market Participants who had been party to a traditional benchmark structure contract since 1999 were likely to have agreed in late 2006/early 2007 to abandon such structure and replace it with a 23 percent price sharing arrangement.

…

I am not aware of any other parties bound by a traditional benchmark style contract who, at the end of 2006, agreed to drop the TC/RC/PP structure in favor of a 23% price sharing approach.

Mr. Ingelbinck seemed to labour under that same misapprehension in his supplementary report:

So was it reasonable and prudent for CMPL in early 2007 to agree to shift the pricing methodology from annual benchmark to 23% price sharing for the 2007 through 2009 period and would one expect an independent seller in similar circumstances to have acted in a similar way?

But, the C.M.P.L.-G.I.A.G. agreement as it was just before February 2007 did not bind the parties to a “traditional benchmark style contract.”

81    Secondly, and again as we have already pointed out, the Commissioner submitted that the C.M.P.L.-G.I.A.G. agreement as it was just before the changes made to it in February 2007 was on arm’s length terms. In his written submissions on appeal at para. 7 he contended as follows:

[T]his Court should also find that the relevant hypothetical agreement was one with price calculated on the terms operating between GIAG and CMPL immediately before the Feb 2007 Amendment. A mutually independent hypothetical miner with the relevant characteristics of CMPL might not reasonably have been expected, as at February 2007, to agree to a 3 year price sharing term at all or 23%, nor to the Enhanced QP Optionality without any quid pro quo.

82    As a consequence of this contention, the Commissioner was, based on his own expert’s opinion, relying upon an agreement which did not reflect what arm’s length parties might be expected to have bargained for. In particular, Mr. Ingelbinck opined that an independent seller would have insisted that deliveries of copper concentrate in 2005 and 2006 be priced only on the basis of benchmark T.C.R.C.s. This perhaps explains why Mr. Ingelbinck’s analysis was based on a contract which used benchmark T.C.R.C.s only to ascertain the applicable T.C.R.C.s rather than the pricing formula said by the Commissioner to be arm’s length, which confined any reliance upon benchmark T.C.R.C.s to 50% of the Base Tonnage.

83    Thirdly, we also note that depending upon what figures (forecast or historical), what years, what periods and what assumptions are used, the quantum of T.C.R.C.s as a percentage of the copper price can be made to go up and down. For example, in contrast to Mr. Ingelbinck’s figures, Mr. Wilson determined that if one assumed the use by C.M.P.L. and G.I.A.G. of the 2006 benchmark T.C.R.C. number together with forecast price participation based on Brook Hunt’s December 2006 forecast copper price for the 2009 year, the result would be a T.C.R.C. which would have been 22% of the forecast copper price in the 2009 year.

84    Fourthly, a problem emerged when Mr. Ingelbinck was cross-examined. He said that with the introduction of back pricing, the buyer was immunised from any downside risk. However, this opinion was shown to be incorrect. In 2009, G.I.A.G. paid $4.8 million more for C.M.P.L.’s copper concentrate than it would have otherwise have paid. This is an illustration of how the various formulae for the ascertainment of T.C.R.C.s are at the mercy of volatile pricing. As the learned primary judge observed at [236]:

In cross examination Mr Ingelbinck stated that by introducing back pricing optionality, the buyer gets to take advantage of optionality with no downside risk. Whilst Mr Ingelbinck agreed that the buyer can get it wrong and choose the wrong quotational period and is not guaranteed to make a profit, he maintained that the buyer is guaranteed not to lose anything on it. His evidence was that “categorically, there is no way a qualified trader can make a mistake on having that optionality. Zero”. However, the view expressed by him was shown to be incorrect. The evidence showed that as the result of the elections made by GIAG in respect of three shipments in the 2009 year, the price that it paid to CMPL was not the lowest price and GIAG in fact paid $4.8 million more for the concentrate than it would have otherwise. Mr Ingelbinck conceded that CMPL was $4.8 million better off as a result of the quotational periods elected by GIAG.

We shall return to this issue.

85    Fifthly, Mr. Ingelbinck did not refer to the terms of any specific contract entered into by independent parties for the sale of copper concentrate. But he generally did criticise the contracts referred to by Mr. Kelly and Mr. Wilson. He did not consider them to be comparable to the C.M.P.L.-G.I.A.G. agreement. On occasion this was based on speculation. For example, in relation to the contracts relied upon by the taxpayer to show that the quotational period optionality clause adopted in February 2007 was arm’s length, Mr. Ingelbinck said:

I am confident that a more thorough review of the timing and the terms of the agreements will show a degree of compensation paid by the buyer to secure the optionality (ie the contracts have lower TC/RC’s than contemporaneous contracts that do not allow back-pricing have).

86    In relation to another contract relied upon by the taxpayer, that was entered into by P.T. Freeport Indonesia and Philippine Associated Smelting and Refining Corporation on about 30 May 2005 (the “P.T.F.I.-P.A.S.A.R.C. Contract”), Mr. Ingelbinck said:

Mr Wilson also reports a 40 kdmt/annum PTFI – Pasar contract starting in 2008 (but not reported until the 2010 version of the Brook Hunt report). The quantity is small but even so I would be quite surprised if Freeport with its seasoned marketing team would have agreed to this as a stand-alone outright sales contract. I suspect the more likely situation was that the PTFI – Pasar contract formed one leg of a swap transaction with the other leg calling for Glencore (or a Glencore related party) to sell a similar quantity at similar terms to PTFI (or a PTFI related party – Atlantic Copper’s Huelva, Spain operation comes to mind) thus enabling PTFI to save on freight costs.

For our part, we would place little weight on reasoning of this kind.

87    Finally, it transpired during the course of the cross-examination of Mr. Ingelbinck that:

(a)    he had no difficulty with the use of a price sharing formula. His complaint was with the rate used; and

(b)    he had no difficulty with the quotational period optionality with back pricing clause. Rather, his complaint was that it appeared that C.M.P.L. had received no quid pro quo for the changes made to this clause.

88    The Commissioner disputed the making of the first concession. But, a fair reading of the transcript reveals that it was made. Thus:

MR DE WIJN: I’ll come back to the range, because when we get to it, the real issue – your issue in this case is not so much that a price-sharing agreement is irrational, but your concern is really the percentage that was chosen?

MR INGELBINCK: In the – now we’re talking the particular case of CMPL at the end of 2006?

MR DE WIJN: Yes.

MR INGELBINCK: Yes.

89    The Commissioner at the trial below also relied upon the expert evidence of another expert miner, Mr. Leonard Kowal. On appeal, the Commissioner placed no reliance upon his evidence. His reports were not referenced in the Commissioner’s written submissions and we were not otherwise taken to any of them or to any of his answers given in cross-examination. That was maybe because of the following finding made about Mr. Kowal by the learned primary judge at [242]:

Mr Kowal accepted in cross examination that he had no experience of what a trader would do and had no basis upon which to assume that the trader would select one quotational period for the whole of the three years. Mr Kowal also accepted that the calculations he had done could only be done with the benefit of hindsight.

90    Then at [404], her Honour said:

There is another reason for doubt to be expressed about the reliability of Mr Kowal’s evidence. There is significant uncertainty as to the extent to which Mr Kowal himself authored his reports. He was engaged by Charles River & Associates in 2016 to assist with advice it was providing to the Commissioner in connection with the audit of Glencore. Mr Kowal said that “Charles Rivers was the contractor to the AGS, and Charles Rivers asked me to help them in certain aspects of the preparation of the report”. Mr Kowal’s report was described by him as a “report made by Charles River & Associates that [he] had input into”. Although Mr Kowal stated that “by far the majority [of his report] is mine” he frankly conceded that components of his report were prepared by Charles River & Associates and it remains wholly unclear which parts of his reports were prepared by him and which parts were not. In the circumstances I cannot conclude that the expressions of opinion in his report are based “wholly or substantially” on specialist knowledge possessed by Mr Kowal based on his training, study or experience as required by s 79 of the Evidence Act.

The Joint Report

91    Messrs. Wilson, Ingelbinck and Kowal prepared a joint report which greatly illuminated matters. Critically, they all agreed that there “is nothing in the pricing or quotational period clauses of the [C.M.P.L.-G.I.A.G. agreement] that does not also exist in contracts between independent market participants”. Mr. Ingelbinck qualified this statement with his observation that context and timing are critically important to determine what might reasonably have been agreed between independent parties in similar circumstances to those which existed between C.M.P.L. and G.I.A.G. Subject to Mr. Ingelbinck’s qualification, the three experts thus agreed as follows:

•    Treatment and refining charges (TC/RCs) can be fixed or Benchmark based

•    TC/RCs can be part Benchmark and part spot market based

•    Combined TC/RCs can be determined on a price-sharing basis

•    Quotational periods (QPs) can be fixed

•    One of the parties can be granted QP optionality

(Footnotes omitted.)

92    The three experts agreed that the original 1999 offtake agreement contained terms that were largely in line with what would have been expected in a benchmark contract entered into by independent market participants. Mr. Kowal and Mr. Wilson also agreed that the closure of the Port Kembla smelter in 2003 justified a change “in the structure” of the C.M.P.L.-G.I.A.G. agreement to increase G.I.A.G.’s options to multiple quotational periods. Mr. Ingelbinck conceded, after some hesitation, that the six options offered in 2004 were acceptable. Back pricing, however, was not. The Commissioner’s experts observed that it was very hard, if not impossible, to place a “precise value” on back pricing. The experts agreed that the impacts of quotational period optionality on a seller can be both positive and negative depending on the market circumstances, and are not directly related to the gains, if any, realised by the buyer and it is almost impossible to determine the actual value received by the seller and the buyer as in some instances the seller wins and in other instances the buyer wins. The Commissioner’s experts also agreed that the 2005 change to introduce some spot market T.C.R.C. pricing was appropriate but they differed on the extent to which C.M.P.L. should have been exposed to the spot market.

93    With respect to the February 2007 amendments, Mr. Wilson reiterated his view that it was both prudent and reasonable for C.M.P.L. to have switched from a mix of benchmark and spot T.C.R.C.s to price sharing, given the C.S.A. mine’s status as a high cost mine, the uncertainty as to benchmark T.C.R.C.s for 2008 and 2009 and as to copper prices for the 2007 to 2009 years, and the importance of ensuring the continued viability of the C.S.A. mine. The Commissioner’s experts maintained their views that the switch to 23% price sharing would not improve the profitability nor viability of the C.S.A. mine and criticised the introduction of quotational period optionality with back pricing on a shipment by shipment basis with no quid pro quo adjustment in favour of C.M.P.L.

94    Mr. Ingelbinck observed that in 2009, even though the vast majority of C.M.P.L.’s copper concentrate went to China, the parties had adopted a freight allowance based on shipping to India. This meant that C.M.P.L. paid a freight allowance of US$60 per outturn w.m.t. of copper concentrate as against US$38 per outturn w.m.t., which was the price for shipping to China, Japan and Korea. Mr. Kowal and Mr. Wilson “expressed no opinion on these observations, stating that they were not shipping experts and therefore unable to comment with any authority.”

The Contracts Said to be Comparable

95    The taxpayer relied upon a number of contracts entered into by independent parties, or by G.I.A.G. with independent parties, to demonstrate that the terms upon which C.M.P.L. sold copper concentrate in the 2007 to 2009 years to G.I.A.G. were arm’s length in nature. Some contracts were cited because of their price sharing terms; others were relied upon for the quotational period optionality clauses they contained. Only one contract was relied upon because it exhibited both features. This was a contract between Red Earth Group Pty Ltd, G.I.A.G. and Cadan Resources Corporation dated 12 January 2010 (the “Red Earth Contract”).

96    The contracts said to contain similar price sharing clauses were as follows:

(a)    The Red Earth Contract;

(b)    The contract between Redbank Mines Ltd and G.I.A.G. dated 29 November 2006;

(c)    The contract between G.I.A.G. and Jiangxi Copper Company Ltd dated 20 December 2001 (including several amendments);

(d)    The B.M.A.G.-Tintaya Contract;

(e)    Contracts between Tritton Resources Ltd and Sempra Metals & Concentrates Corp dated 25 October 2002, 11 October 2004, 12 December 2005 and 24 January 2007; and

(f)    The P.T.F.I.-P.A.S.A.R.C. Contract.

97    The contracts said to contain similar quotational period optionality clauses were as follows:

(a)    Contracts between G.I.A.G. and Peak Gold Mines Pty Ltd dated 28 May 2004 and 7 March 2005 (including amendments to the latter dated 21 June 2005, 5 August 2005 and 21 February 2007);

(b)    The Barminco Contract;

(c)    The contract between Oxiana Golden Grove Pty Ltd and G.I.A.G. dated 4 July 2008;

(d)    The contract between Oxiana Golden Grove Pty Ltd and G.I.A.G. dated 22 February 2008 (the “Oxiana Golden Grove Contract”);

(e)    The Red Earth Contract;

(f)    The contract between the Mongolian-Russian Joint Venture Erdenet and G.I.A.G. dated 24 April 2007; and

(g)    The contract between Montana Resources, L.L.P. and Glencore Ltd dated 11 January 2007.

98    Much time was spent by the Commissioner’s experts and by the Commissioner in distinguishing these contracts from the C.M.P.L.-G.I.A.G. agreement. Some were said to be not comparable because they had been entered into at different times when the copper market was behaving differently. Some involved the sale of far less tonnage compared to the quantum of concentrate sold by C.M.P.L. (Mr. Wilson thought that quantum of concentrate sold did not matter). Some were said to be limited to one shipment. Some had been entered into before the commencement of production. One contract was between related parties (the B.M.A.G.-Tintaya Contract) — in respect of this contract, Mr. Ingelbinck thought that price sharing at 25.5% was “within the range” for an arm’s length agreement. Some involved financiers. One involved a lower grade of copper concentrate. One only related to specific tonnage instead of 100% of concentrate for the life of the mine. One was for the sale of copper cement as well as copper concentrate. In relation to one contract — the Oxiana Golden Grove Contract — the Commissioner below made no submissions.

99    Mr. Wilson did not accept that all of these differences mattered. In some cases, however, he accepted that the contract in question told the Court nothing about the likelihood of C.M.P.L. entering into a contract, which used the same terms as the C.M.P.L.-G.I.A.G. agreement, with an independent party, save that the independent contracts provided a “reference point.” In other words, the contracts were used as a sounding board to see whether clauses like those found in the C.M.P.L.-G.I.A.G. agreement were used in the copper concentrate market by parties who are independent of each other. Ultimately, as we understood it, the three experts agreed that this was so. Moreover, again as we understood it, the learned primary judge treated the contracts as evidence for that proposition, but no more than this. As her Honour said at [307]:

In the present case, contrary to the Commissioner’s contention, the examples of offtake agreements put forward by the taxpayer have probative value as illustrations of contracts between independent mine producers and traders for the sale of copper concentrate containing price sharing mechanisms and/or quotational period optionality clauses, including with back pricing, consistent with the joint position of the experts that such terms were standard in copper concentrate contracts in the relevant years and there was nothing in the price sharing or quotational period clauses of the February 2007 Agreement that did not exist in contracts between independent market participants.

100    Having an industry expert examine contracts entered into by other parties is, perhaps, in a transfer pricing dispute, a necessary but nonetheless unsatisfactory task. There is only so much that such an expert can legitimately say about a deal she or he had nothing to do with. She or he may lack the direct knowledge of the necessary commercial context and background to understand why certain terms were selected and why they evolved in the particular way that they have. She or he may not be able to discern accurately why differences might exist between the agreement in dispute, and the agreement said to be comparable to it. There can be a tendency for the expert, with their great knowledge, to make factual comments or observations which are really no more than speculation or which constitute severe hearsay. Both Mr. Wilson and Mr. Ingelbinck did this at times. If a party wishes to rely upon factual matters with respect to a contract said to be comparable (to perhaps explain the reason for differences between contracts, or to show that any such differences are not material), it may be preferable for such evidence to be led by a lay witness, rather than through the prism of an expert’s report. Failing that, it may be possible to have recourse to publically available business records. The expert can then take that lay evidence, or those documents, into account in preparing her or his expert report.

101    Naturally, it may be very difficult for either a taxpayer or the Commissioner to lead this type of lay evidence. Witnesses may not be willing to assist and publically available business records will only go so far. If that is the case, the utility of looking at contracts said to be comparable may be confined. It may be limited to considering such contracts as no more than reference points, or as illustrations of arm’s length terms, as was the case here.

102    Notwithstanding these difficulties, as already mentioned, the parties spent considerable time and effort in making submissions about the contracts said here to be comparable. For example, during the hearing of the appeal, the taxpayer electronically handed up a 14 page document entitled “Respondent’s Note Concerning Comparability of Contracts in Evidence.” No doubt this was presented as an aide-mémoire which might efficiently capture the taxpayer’s contentions about this issue. The Commissioner sought and was granted leave to respond. A week later he filed a 47 page document on this issue. The length and detail of what was filed reflected, with profound respect, a degree of unnecessary energy about an issue which is perhaps not as nuanced or as sophisticated as the parties might have thought.

103    In contrast to the making of lay observations about contracts said to be comparable, and the commercial conditions in which such deals are struck, an industry expert may validly give an opinion, based on her or his experience and expertise, about whether a clause in a contract specifying the consideration payable, is one which independent parties dealing with each other at arm’s length, might or might not be expected to adopt. She or he may rely on other contracts in the market place entered into by independent parties, as a reference point, sounding board or as other similar support in forming that view.

104    We shall return to consider the contracts said to be comparable to the C.M.P.L.-G.I.A.G. agreement.

The Judgment Below

105    The learned primary judge found in favour of the taxpayer on all issues relevant for this appeal. In essence, her Honour preferred the evidence of Mr. Wilson to that of Mr. Ingelbinck.

106    Her Honour explained the issue which she considered she needed to address at [172] in the following terms:

In this case, as a subsidiary of a multinational commodity trading group selling all of its copper concentrate to GIAG for the life of the mine, CMPL did not have, and had no need for, any marketing department or agency agreement in place to market its copper production, nor did it have to arrange the logistics required for transporting its copper concentrate to any overseas destination as this was all done by Glencore’s Traffic team. In those circumstances, the consideration that might reasonably be expected to be received by an independent seller of copper concentrate in the position of CMPL must therefore be determined on the hypothesis of an independent buyer of copper concentrate in the position of GIAG acquiring the whole of CMPL’s copper concentrate for the life of the mine and providing logistical and marketing support to the seller of the copper concentrate, and on the hypothesis of an independent seller of copper concentrate in the position of CMPL selling the whole of its production of copper concentrate to an independent buyer for the life of the mine, and with no need for a marketing department or logistics expertise.

107    On appeal, the Commissioner faintly criticised the learned primary judge for what was alleged to be a conflation of the issues arising under Div. 13 of the 1936 Act with those under Subdiv. 815-A of the 1997 Act. In our view, if any such conflation took place, it reflected the way the parties presented their respective cases. No one suggested that different outcomes might be possible depending upon whether one applied Div. 13 or Subdiv. 815-A. At a factual level, the submissions made by the parties were the same regardless of which set of provisions applied.

108    The learned primary judge discussed the relevant authorities and law applicable to Div. 13 and Subdiv. 815-A in some considerable detail. One issue of law was important to her Honour’s reasons. The Commissioner submitted that in applying Div. 13 or Subdiv. 815-A, neither the Commissioner nor the Court was bound by the choices made by C.M.P.L. and G.I.A.G. about the type of pricing formula to be used for the sale of copper concentrate. For example, it was open for the Court to find that arm’s length parties would not have adopted a price sharing formula, but would have instead used benchmark T.C.R.C.s. Alternatively, it was open to find that the pre-February 2007 C.M.P.L.-G.I.A.G. agreement would have been maintained because an arm’s length seller would never have agreed to the amendments made in February 2007.

109    The learned primary judge rejected the Commissioner’s submission. In her Honour’s view, based upon what this Court said in Chevron, upon the terms of Div. 13 and upon paras. 1.37 and 1.38 in the Transfer Pricing Guidelines (which address the exceptional circumstances in which a revenue authority may, when examining a controlled transaction, deviate from the controlled transaction in fact entered into by the related parties), it was not open to the Commissioner or the Court to substitute a different pricing formula for the pricing formula in fact used by C.M.P.L. and G.I.A.G. This was said to be an impermissible reconstruction of the contract actually entered into by C.M.P.L. and G.I.A.G., which was not supported by paras. 1.37 and 1.38 of the Transfer Pricing Guidelines. Relevantly, this meant, amongst other things, that the Commissioner and the Court were stuck with price sharing. As her Honour reasoned at [314]:

In my opinion, the Commissioner’s approach impermissibly restructures the actual contract entered into by the parties into a contract of a different character. The decisions in Chevron, both at first instance and on appeal, make it clear that the hypothetical should be based on the form of the actual transaction entered into between the associated enterprises but on the assumption that the parties are independent and dealing at arm’s length, in order to identify a reliable substitute arm’s length consideration for the actual consideration given or received. This is consistent with, and confirmed by, the 1995 Guidelines, referred to by Allsop CJ in Chevron at [89], which state at [C.1.36], in a section headed “Recognition of the actual transactions undertaken”, that restructuring the controlled transaction under review is generally inappropriate when making the comparison of the conditions in a controlled transaction with conditions in transactions between independent enterprises dealing at arm’s length and that the analysis “ordinarily should be based on the transaction actually undertaken by the associated enterprises as it has been structured by them”. The 1995 Guidelines also state at [C.1.36] that in the examination of whether a “controlled transaction” satisfies the arm’s length principle “in other than exceptional circumstances”, the actual transaction should not be disregarded or other transactions substituted, and that:

Restructuring of legitimate business transactions would be a wholly arbitrary exercise the inequity of which could be compounded by double taxation created where the other tax administration does not share the same views as to how the transaction should be structured.

110    Accepting the pricing formula used by C.M.P.L. and G.I.A.G. left the following issues for determination by the learned primary judge (aside from the freight allowance) (at [345]):

(a)    the ascertainment of an applicable price sharing percentage or range; and

(b)    deciding what, if, any, quid pro quo might be expected for the quotational period optionality conferred on G.I.A.G.

111    The learned primary judge was satisfied on the evidence that the 23% price sharing rate “was within an arm’s length range” for the following eight reasons (at [346]-[354]):

(1)    The rate of 23% sat in the middle of the Brook Hunt range of 21-26% as at the end of 2006. The Commissioner never challenged the accuracy of that range.

(2)    The contracts said to be comparable, but which were ultimately used instead as “reference points”, had price sharing rates between 20% and 27.5%.

(3)    As at December 2006, according to Brook Hunt, the historical long term average ratio of Japanese benchmark T.C.R.C.s to the L.M.E. copper price was 22%. Her Honour noted that Mr. Ingelbinck thought that it was more relevant to consider forecast benchmark T.C.R.C.s, but accepted that past T.C.R.C.s retained some relevance.

(4)    The B.M.A.G.-Tintaya Contract said to be formalised in December 2003 adopted a price sharing rate of 25.5%, which was 8.5% over the top of the Brook Hunt forecast T.C.R.C.s for the 2004 to 2006 years. Yet, Mr. Ingelbinck considered this to be a “fair transfer price.”

(5)    Historically, from 2000 to 2006, C.M.P.L.’s T.C.R.C.s represented on average approximately 21% of the copper price.

(6)    There was no evidence that arm’s length parties in February 2007 would only have had regard to prevailing forecasts in determining the rate of T.C.R.C.s.

(7)    Mr. Ingelbinck in cross-examination accepted that an independent party might agree to pay higher T.C.R.C.s to avoid volatility and to secure certainty.

(8)    Mr. Ingelbinck was not asked to consider what the rate of price sharing should have been and he gave no alternative rate. In these circumstances, it was speculation to consider that a rate less than 23% might be expected to have been chosen. In contrast, the Brook Hunt report was a “reliable” source of information which supported a rate of 23%.

112    As for quotational period optionality, the learned primary judge made two initial observations. First, it was agreed that this type of optionality can result in negative and positive consequences for a seller; it just depends on market conditions. Secondly, whilst clearly desirable for a buyer, it was very hard if not impossible to place a “precise value” on back pricing. Moreover, the value of quotational period optionality in and of itself could not be quantified.

113    Her Honour then observed that Mr. Kelly had given unchallenged evidence that copper concentrate agreements are negotiated “holistically”; a particular benefit is not traded or exchanged for another particular benefit. Both Mr. Wilson and Mr. Ingelbinck agreed that quotational period optionality with back pricing was part of the overall bargain struck between the parties to an agreement. The learned primary judge ultimately rejected Mr. Ingelbinck’s opinion about the need for a quid pro quo, because he had mistakenly believed that a trader cannot “lose” with quotational period optionality with back pricing. Moreover, to the extent that he had sought to value this benefit, he could only have done so with the benefit of hindsight. Because there was no other way to value this perceived benefit, it was not otherwise ascertainable as at February 2007. C.M.P.L. and G.I.A.G. at that time had no means of determining how the benefits and disadvantages of quotational period optionality could be known, or indeed estimated, for the 2007 to 2009 years. Finally, her Honour observed that Mr. Ingelbinck never suggested what the quid pro quo should have been. The learned primary judge rejected the suggestion that it would have taken the form of a discount to an agreed T.C.R.C. rate because a review of the contracts said to be comparable to the C.M.P.L.-G.I.A.G. agreement showed that only two demonstrated some obvious quid pro quo for greater optionality. The two contracts were the Barminco Contract, and a 2002 contract between Placer Pacific (Osborne) Pty Ltd and G.I.A.G. referred to in the Commissioner’s submissions before her Honour. In other words, this specific type of exchange in value was found by her Honour to be “unusual.”

114    The learned primary judge also considered the freight allowance paid in 2009 by C.M.P.L. Her Honour rejected the Commissioner’s submission that using the costs referrable to shipping to India (US$60 per outturn w.m.t.) did not result in an arm’s length allowance. This was because the Commissioner’s case on the freight allowance relied to a substantial degree upon the benefit of hindsight. Her Honour reasoned as follows at [376]-[378]:

First, the use of a FOB freight provision whereby the buyer arranges for freight, and is provided with a mutually agreed freight allowance which is deducted against the amount paid to the seller, was a standard term in contracts for the supply of copper concentrate in the relevant years.

Secondly, the Commissioner’s complaint impermissibly relies on hindsight analysis – that is, that few shipments were made to India for 2009 and the rates for shipping to China, Japan, and Korea in that year were less than the actual rate charged.

Thirdly, it does not follow from the simple fact that GIAG received a higher freight allowance in 2009 than if it had chosen the freight rates for China, Japan, or Korea that it might be expected that independent parties would not have agreed such terms. As Mr Kelly explained in cross examination, there are a number of different factors which might affect freight rates agreed between the parties.

115    For these reasons, her Honour concluded that the consideration in fact paid to C.M.P.L. “fell within the range of consideration that might reasonably be expected” to have been paid for the purposes of Div. 13 of the 1936 Act. Her Honour also concluded, for the purposes of Subdiv. 815-A of the 1997 Act, that there was no amount of profits which, but for the conditions mentioned in Art. 9 of the Swiss Treaty, might have been expected to accrue to C.M.P.L. but which, by reason of those conditions, did not so accrue.

Submissions on Appeal

116    The Commissioner filed three notices of appeal out of time. Two set out 26 grounds of appeal. One set out 27 grounds of appeal. The Commissioner was granted a sufficient extension of time on 28 November 2019.

117    It is unnecessary to set out the grounds of appeal. That is because, and putting aside for one moment the freight allowance issue, the substance of the Commissioner’s case, notwithstanding its great length and intricate detail was, if we may respectfully say so, attractively simple.

118    In essence, the Commissioner submitted that C.M.P.L. had entered into a contract with G.I.A.G. which had been on arm’s length terms, being the C.M.P.L.-G.I.A.G. agreement as it was prior to February 2007 which incorporated the amendments made in 2004 and 2005. But in February 2007, C.M.P.L. did something which, if it had been independent of G.I.A.G., it would never have agreed to. It submitted to a new pricing formula which significantly reduced its gross earnings from the sale of copper concentrate. It got nothing for this change and so had acted irresponsibly. Critically, and moreover, Mr. Wilson agreed that, based on the December 2006 Brook Hunt report and C.M.P.L.’s own budget at the time, it was reasonable to expect that C.M.P.L.’s profit was going to be worse off for the 2007 to 2009 years under the new pricing formula. No party acting independently from G.I.A.G. and at arm’s length would ever have agreed to such detrimental changes.

119    Importantly, and even accepting that a price sharing clause, for example, could confer commercial benefits on C.M.P.L. (such as those identified by Mr. Ingelbinck), the taxpayer had never demonstrated that these benefits justified giving up so much expected gross revenue.

120    In relation to the issue of the quotational period optionality with back pricing clause, the Commissioner relied upon the expert evidence of Mr. Ingelbinck that this term really favoured the buyer. Mr. Wilson did not appear to disagree with that proposition. He also agreed that such a clause is far more likely to result in a discernible negative impact on a seller’s revenue. Whilst the Commissioner did not quantify the effect of this clause on the consideration C.M.P.L. in fact received for the sale of copper concentrate, he submitted that the onus lay on the taxpayer to demonstrate that the failure to negotiate a quid pro quo for this clause did not affect the taxpayer’s case that the amount C.M.P.L. had received was an arm’s length price. The taxpayer had failed to discharge that onus.

121    Finally, on the issue of the freight allowance, the Commissioner’s case was even simpler. The Commissioner submitted that the taxpayer did not lead any expert evidence to demonstrate that the payment of US$60 per outturn w.m.t. in 2009 was an arm’s length allowance. Mr. Wilson conceded in the joint expert report that he was not a shipping expert and could not comment on Mr. Ingelbinck’s concerns about the allowance. It followed that the taxpayer had failed to discharge its onus of proof on this issue.

122    Several passages in the transcript of the trial were critical to the presentation of the Commissioner’s case on the issue of price sharing. These passages are the very marrow of the Commissioner’s case. They were relied upon to try to demonstrate that Mr. Wilson had conceded that the February 2007 amendments significantly disadvantaged C.M.P.L., and thus would not have been adopted by independent parties in the positions of C.M.P.L. and G.I.A.G. The Commissioner’s senior counsel cross-examined Mr. Wilson at length. Two of these passages are sufficient to illustrate the Commissioner’s case at its highest.

123    In the first, Mr. Wilson agreed that Brook Hunt had observed that in 2006 mining costs had exceeded budgeted forecasts “across the board.” He agreed that this was also the case for C.M.P.L. On this basis, and based on forecast copper prices, it was put to Mr. Wilson that switching to price sharing put the financial viability of the C.S.A. mine at risk. Mr. Wilson did not agree with this proposition because the costs of operating the C.S.A. mine were “all over the place” and did not support any particular “trend” in the numbers. Mr. Wilson was then asked to assume that the C.S.A. mine had an issue with escalating costs. On that basis he was asked if the adoption of price sharing at a rate of 23% exposed C.M.P.L. to the risk of financial viability “moving forward.” Mr. Wilson responded as follows:

But you’re making assumptions that, you know, you simply don’t know. In terms of cost escalation – we made it very clear at Brook Hunt that the escalation in costs, there were two components to that. One was the sticky-type of costs that were due to rising oil prices, rising steel prices etcetera. Now, of course, if copper prices were going to be falling because of a global recession, the other thing that would have fallen was oil prices. That would have impacted sea freight, that would have impacted land freight, chemical costs, you would have had steel prices falling quite substantially. So in fact, the argument of rising costs as we’re approaching the end of an economic cycle, really don’t stand up.

124    Mr. Wilson was again pressed on this issue and asked to consider whether the adoption of price sharing at a rate of 23% put the financial viability of the C.S.A. mine at risk if one assumed that there was a concern about “rising operating costs”. Mr. Wilson agreed with the proposition, but said that he was “forced” to do so because of the assumption he had been asked to make. He thus responded in the following way:

If this is your construct and I’m total constrained, then, you know – I mean, I would also have to know by how much the costs were escalating and, you know, what that meant in terms of the cost relative to prices. But if you’re tying my hand up and saying, you know, “Would this happen?” then I guess I’m forced to say yes on the basis of your construct. I don’t agree with it, but, you know, you put me in a corner where I have to say yes.

125    It was then put to Mr. Wilson that the only reason he had for disagreeing with the Commissioner on this issue was the potential issue of rising costs. Mr. Wilson responded by stating that it was the “main” reason for his disagreement with the Commissioner. In his view, the issue of cost escalation should have peaked in 2006 because the market was coming to the “end of an economic cycle.”

126    For the purposes of this cross-examination, the Commissioner relied upon certain answers given by Mr. Kelly to make good the assumption Mr. Wilson had been asked to make about “rising operating costs”. Mr. Kelly had been asked about whether the C.M.P.L. 2007 Budget identified any risks for the C.S.A. mine in the future. He referred to increasing operating costs and the risk that these might blow out even further. Mr. Kelly said when giving his evidence:

I think the management report does detail the capital forecasts required to access that ore. It outlines the increasing operating costs and the risks to those blowing out even further. It reports on falling recoveries, falling head grade in mine, falling copper concentrate grade produced. These are all things that the report specifically details. The mine plan, the budget going forward, specifically talks about all of these things.

(Our emphasis.)

127    In that respect, the learned primary judge had decided that it did not matter whether the 2007 Budget projected “a positive outlook or something less rosy.” Nonetheless, her Honour made the following finding about the financial condition of the mine in 2006 and 2007 at [120]:

I accept Mr Kelly’s evidence about the depth of the mine giving rise to significant technical and financial challenges and that the mine in 2006 and 2007 was experiencing difficulties with staffing and water supplies. In addition, there was documentary evidence showing that the mine had experienced difficulties in the past with ground instability, such as stope failures and ground collapses. However, as the contemporaneous documents also evidenced, as at late 2006 and early 2007 the mine had plans and steps in place dealing with these matters and there was nothing in the contemporaneous documents which identified any uncertainty for the mine in relation to its capacity to continue mining in 2007, 2008 or 2009 or to suggest that CMPL considered that there was any real risk that its level of production would not continue throughout those years. Indeed, the review of the 2006 year set out in the 2007 Budget noted that mine production of ore in 2006 had increased by over 30% to 810,000 tpa and concentrate production in 2006 would be “an all-time record for CSA”. It also stated that “this [had] been achieved even with high turnover in both staff and AWA positions and finishing the year with over 20% vacancies in staff position [sic] due to the extremely difficult employment market”.

128    The second critical passage of the cross-examination of Mr. Wilson commenced with him agreeing that it was “highly unlikely” that as at February 2007 the copper price would dip below US$1/lb at any time during the 2007 to 2009 years. He also agreed that as at that time, it was “highly unlikely” that the T.C.R.C.s would go above $70 a tonne over the 2007 to 2009 years. Mr. Wilson then agreed that on the basis of available market information, in particular from Brook Hunt, C.M.P.L. would have been expected to be “worse off financially” as a result of the switch to 23% price sharing. The relevant exchange was as follows:

MS STERN:    So do you not agree with me also that, on the basis of available market information, it was highly likely that CMPL would be worse off financially by agreeing to a three-year price-sharing agreement than if they remained on the terms that they had as at the start of 2007 in their agreement with GIAG?

MR WILSON:    That would be potentially right, yes.

MS STERN:    Not potentially right. It is right, isn’t it?

MR WILSON:    Well, if they followed the Brook Hunt forecast, they would say that.

MS STERN:    And also we had a look at their own forecast before lunch, if they followed their own forecast, it was highly likely, wasn’t it, that they would be worse off by going to a 23 per cent price-sharing agreement?

MR WILSON:    Correct.

129    It was then put to Mr. Wilson that it was not a “commercially rational decision” for C.M.P.L. to have agreed to switch to price sharing at the rate of 23% “bearing in mind the existing contract [C.M.P.L.] had with GIAG”. Mr. Wilson responded by stating that with the adoption of such a clause C.M.P.L. “would have remained profitable and [have] covered [its] cash production cost.”

130    It was then put to Mr. Wilson that if C.M.P.L.’s “aim was to be as profitable as possible, that the commercially rational thing to do in February 2007, would have been to stick with” the “current terms,” i.e. the terms applicable before the changes made in February 2007. Mr. Wilson agreed with that proposition. Here is the critical passage:

MS STERN:    Okay. But you agree with me if their aim was to be as profitable as possible, that the commercially rational thing to do in February 2007, would have been to stick with their current terms, which were part benchmark and part spot, would you agree with that?

MR WILSON:    Yes, I would.

131    In the foregoing passage, the Commissioner emphasised the presence of the definitive article before the phrase “commercially rational.” It was suggested that this implied that there was only one commercially rational outcome.

132    To make good the proposition that C.M.P.L.’s aim as an independent miner would have been to maximise profitability, the Commissioner emphasised that he had cross-examined Mr. Kelly about what was suggested to be the Glencore Group’s policy of maximising the profit of an asset, such as the C.S.A. mine, and of the Group as a whole. Mr. Kelly responded in cross-examination:

So in that role you were engaged or employed by Glencore to manage its mining assets; is that right?---Yes.

Right. And you did so with a view to maximising profit for the Glencore group; is that right?---Yes.

133    The exchange with Mr. Kelly relevantly continued as follows:

Okay. So your evidence is, then, that Glencore did exercise financial control over the asset, just not through you?---I – I take – you mean financial control to mean approving budgets. Yes.

Yes. And other respects of financial control, like approving capital expenditure that through you?---At certain material levels, yes.

Okay. And decisions about expansion of the mine or expansion of production, they were taken by Glencore as well. You agree?---Yes.

Okay. And all with a view to maximising the profit of the group?---Maximising the profit of the asset. Yes.

And maximising the profit of the asset for the group, Glencore, as a whole?---Yes.

134    The Commissioner submitted that the foregoing evidenced a Glencore Group policy of maximising profits from group assets. With respect, expressed in this highly generalised way, this so-called policy appeared to us to be no more than a bare commercial platitude. It could encompass many different ways of making a profit. In our view, there is no necessary antithesis between maximising profits and managing risk. A commercially rational entity might eschew the greatest possible earnings to lock in certainty and to avoid exposure to a risk or risks. Such an entity is still pursuing a policy of maximising profitability. It is just applying that policy conservatively. We shall return to this issue. It follows that whether or not this so-called Glencore Group policy must necessarily apply to the hypothetical seller in the position of C.M.P.L. for the purposes of applying Div. 13 and Subdiv. 815-A probably does not matter. It really tells one nothing about what arm’s length consideration an independent party in the position of C.M.P.L. might reasonably be expected to have received.

135    In response to the parts of the transcript we have set out above, the taxpayer submitted that it needed to be understood that when Mr. Wilson agreed with the proposition that switching to price sharing would result in C.M.P.L. being worse off financially, this was based on a number of assumptions. The first was that it assumed that the Brook Hunt forecasts would materialise, secondly that C.M.P.L. would have sold copper concentrate on only benchmark terms and thirdly that the forecasts in the 2007 Budget would also materialise. We do not think that this submission is entirely correct. What was put to Mr. Wilson was the proposition that the pre-existing contract terms “which were part benchmark and part spot” should have been maintained.

136    Perhaps more important was Mr. Wilson’s evidence about the reliability of forecasting copper prices and benchmark T.C.R.C.s. As for spot T.C.R.C.s, these were essentially impossible to predict. As Mr. Wilson said when giving his evidence:

The only other thing I would mention is that the one thing we don’t try and forecast – I know of very few examples of people trying to forecast it – is what occurs in the spot market. Now, the contract leading up to the change in February 2007 between CMPL and Glencore was the provision of a contracted tonnage which was 50,000 tonnes basis benchmark terms and 50,000 tonnes basis spot terms with any additional production to be mutually agreed in terms of the structure. So, you know, if we were back in a situation where TCRCs were back at that sort of 95 9.5 level, if we look at the spot markets, those terms move very much dramatically. But if we look at the 2005, for example, where the market was rising, we had an average spot TCRC of 37.7 cents a pound. Now, if you applied that, let’s say, on a fifty-fifty basis to the 2009 level, you’d have had a weighted average TCRC equivalent to 26 per cent of the copper price forecast. So there are various outcomes one could assume.

137    Mr. Wilson also gave evidence about the difficulty of just relying upon forecast benchmark T.C.R.C.s. When taken to a Brook Hunt report and to numbers he “personally put together” he said that a high cost producer (such as, we infer, C.M.P.L.) might be somewhat concerned about the forecast. He then said the following:

There are so many moving parts if I was a third party high-cost producer, I might be a little bit nervous – given there was only going to be 125,000 deficit in 2008 predicted in this report – of how those TCRCs may have played out. You know, the potential is that if the Chinese smelters had shut down, that could have led to, for example, quite a large surplus in the market which would have fundamentally changed those forecasts. So I just want to make it absolutely clear that, while I stick by those forecasts, there are some very big variables that we have to just make assumptions on in order to come up with those forecasts.

138    The nub of Mr. Wilson’s point was that with a forecast material reduction in copper prices it was commercially prudent in early 2007 for a high cost mine to seek the certainty of three years of price sharing, which here, based on prevailing forecasts, would keep C.M.P.L. profitable. As he said during his cross-examination:

The TCRC market was highly volatile so, for example, in 2004, the TCRC was 13 cents a pound. In 2006, it had trebled to 46 cents a pound. And for a mine that was concerned particularly on – you know, the lower price environment, with the very, very material changes – high volatility – in the benchmark TCRC market, if they could have guaranteed, on the basis of Brook Hunt’s price forecast by 2009, that they still would have been profitable, that would have been a very, very strong commercial incentive to switch to price-sharing.

139    The Commissioner sought to downplay (in our respectful view unsuccessfully) the unreliability of forecasting in the copper concentrate market and put to Mr. Wilson that his very expertise was in analysing markets and making forecasts. This exchange then occurred:

MS STERN:    Of course, Mr Wilson, but that’s your expertise, isn’t it, analysing markets?

MR WILSON:    It’s impossible. In my honest opinion, we do our very best to come up with what – you can’t forecast disruptions to mines; you simply can’t. You know, there will be some strikes – I can’t forecast when a pit wall – a pit wall will slip and fall into the pit, for example. It’s impossible. So, you know, it’s – I – we use our best professional opinion to come up with a base case, but it is only a base case, you know, assuming factors of various moving parts. So I just want to make that absolutely clear. This is not a straightforward, “Well, it’s supply minus demand and there it gives you the answer.” It’s – there are a lot of – and this is another reason why forecasting prices is so very difficult.

MS STERN:    But, Mr Wilson, your expertise is the analysis of all these factors in the market and coming up with estimates?

MR WILSON:    Correct.

MS STERN:    That is – that’s what you developed in your professional life as your particular area of expertise; do you agree with that?

MR WILSON:    I do and, with respect, I would also say there are many of these factors that are impossible to forecast accurately.

This exchange calls to mind the observation of the High Court in Commissioner of State Revenue v. Placer Dome Inc (2018) 265 C.L.R. 585, where it was noted (at 601 [47]) that an expert for the taxpayer in that case had acknowledged that “estimates of future gold prices could be “quite dramatically wrong”, predictions could be pretty unreliable and, as a result, [reports based on such estimates] could turn out quite inaccurate”.

140    Mr. Wilson also agreed that he could only “speculate” as to what a high cost miner might have chosen to do in early 2007. We do not construe this answer as a denial of his ability to give his opinion about what a high cost miner might have chosen to so. We construe it as an admission that he had not been a party to any actual contract. He elsewhere in his cross-examination explained the basis of his knowledge when he said that he had spoken to many traders over 36 years as to how the copper concentrate market works. Nonetheless, the Commissioner made a submission that such speculation, as he called it, was of no assistance in applying Div. 13 and Subdiv. 815-A. We shall return to this issue.

141    We will otherwise advert to the other submissions of the parties, where we need to, when giving our reasons for the disposition of the matter below.

Disposition

The Role of this Court

142    The first issue for consideration is the role of an appeal court in a transfer pricing case. The Commissioner urged us to review most of the evidence before the learned primary judge, which comprised thousands of pages and thousands of transcript lines, and form our own view about whether the taxpayer had discharged its onus for the purpose of applying Div. 13 and Subdiv. 815-A to the sales of copper concentrate under the C.M.P.L.-G.I.A.G. agreement. For that purpose, over almost two days of argument, it appeared to us that the Court was taken to every piece of evidence, and every transcript reference, that supported the Commissioner’s case. The Commissioner submitted that the following passage from the decision of this Court in Branir Pty Ltd v. Owston Nominees (No 2) Pty Ltd (2001) 117 F.C.R. 424 at 436 [25] (per Allsop J. as his Honour then was) was applicable:

The inability to identify error may arise in part from the unwillingness of the appeal court to be persuaded that it is in as good a position as the trial judge to deal with the issues, because of the kinds of considerations referred to in [24] above. Or, it may be that the nature of the issue is one such that (though not a discretion) there cannot be said to be truly one correct answer. In such cases the availability of a different view, indeed even perhaps the preference of the appeal court for a different view, may not be alone sufficient: see Zuvela v Cosmarnan Concrete Pty Ltd (1996) 71 A.L.J.R. 29 at 30-31; 140 ALR 227 at 229-230. In circumstances where, by the nature of the fact or conclusion, only one view is (at least legally) possible (for example, the proper construction of a statute or a clause in a document, where, although, as often said, minds might differ about such matters of construction, there can be but one correct meaning: see generally Enfield City Corporation v Development Assessment Commission (2000) 199 CLR 135 at 151-156) the preference of the appeal court for one view would carry with it the conclusion of error. However, other findings and conclusions may be far more easily open to legitimate differences of opinion eg valuation questions, see Fenton Nominees Pty Ltd v Valuer-General (1981) 27 SASR 258 at 259-263; 47 LGRA 71 at 73-76.

143    The Commissioner submitted that this was a case where only one view was possible about the application of Div. 13 and Subdiv. 815-A to the facts. Consequently, the formation of a different view of the facts by an appeal court, following a review of all of the evidence, would be sufficient to ground the existence of legal error.

144    The taxpayer took a different view. It referred the Court to the reasons of Gageler J. in Minister for Immigration and Border Protection v. SZVFW (2018) 264 C.L.R. 541 at 556-557 [33], where his Honour said:

Performing its obligation to conduct a “real review”, the appellate court “must, of necessity, observe the ‘natural limitations’ that exist in the case of any appellate court proceeding wholly or substantially on the record”. Limitations of that nature can include: “those occasioned by the resolution of any conflicts at trial about witness credibility based on factors such as the demeanour or impression of witnesses; any disadvantages that may derive from considerations not adequately reflected in the recorded transcript of the trial; and matters arising from the advantages that a primary judge may enjoy in the opportunity to consider, and reflect upon, the entirety of the evidence as it is received at trial and to draw conclusions from that evidence, viewed as a whole”. The appellate court needs to be conscious that “[n]o judicial reasons can ever state all of the pertinent factors; nor can they express every feature of the evidence that causes a decision-maker to prefer one factual conclusion over another”. The more prominently limitations of that nature feature in a particular appeal, the more difficult it will be for the appellate court to be satisfied that the primary judge was in error.

(Footnotes omitted.)

See also CSR Ltd v. Della Maddalena [2006] HCA 1; (2006) 80 A.L.J.R. 458 at 465 [17].

145    The taxpayer in particular emphasised the advantage of the learned primary judge here who had the opportunity to consider, and reflect upon, the entirety of the evidence as a whole. In that respect, not all of the evidence led below was before this Court. The justification for appellate constraint was also referred to by Kirby J. in State Rail Authority of New South Wales v. Earthline Constructions Pty Ltd (in liq) [1999] HCA 3; (1999) 73 A.L.J.R. 306 where his Honour said at 330 [89]-[90]:

None of the foregoing considerations requires the abandonment of the respect which appellate courts, by present legal authority, must pay to the advantages enjoyed by the trial judge. Instead, they require renewed attention to precisely what the advantages are which the trial judge has over those enjoyed by the appellate court, conducting a second look at the facts, usually with more opportunity to evaluate particular facts than is possible in the midst of a trial and with the appellate advantage of viewing such facts in the context of the record of the complete trial hearing.

The true advantages in fact-finding which the trial judge enjoys include the fact that the judge hears the evidence in its entirety whereas the appellate court is typically taken to selected passages, chosen by the parties so as to advance their respective arguments. The trial judge hears and sees all of the evidence. The evidence is generally presented in a reasonably logical context. It unfolds, usually with a measure of chronological order, as it is given in testimony or tendered in documentary or electronic form. During the trial and adjournments, the judge has the opportunity to reflect on the evidence and to weigh particular elements against the rest of the evidence whilst the latter is still fresh in mind. A busy appellate court may not have the time or opportunity to read the entire transcript and all of the exhibits. As it seems to me, these are the real reasons for caution on the part of an appellate court where it inclines to conclusions on factual matters different from those reached by the trial judge. These considerations acquire added force where, as in the present case, the trial was a very long one, the exhibits are most numerous, the issues are multiple and the oral and written submissions were detailed and protracted. In such cases, the reasons given by the trial judge, however conscientious he or she may be, may omit attention to peripheral issues. They are designed to explain conclusions to which the judge has been driven by the overall impressions and considerations, some of which may, quite properly, not be expressly specified.

(Footnotes omitted.)

146    More recently, in Jadwan Pty Ltd v. Rae & Partners (A Firm) [2020] FCAFC 62; (2020) 378 A.L.R. 193 at 307-312 [402]-[415], Bromwich, O’Callaghan and Wheelahan JJ. considered all of the applicable authorities concerning appellate review. Leeming J.A. in Council of the Law Society of New South Wales v. Zhukovska [2020] NSWCA 163 at [85] has since described Jadwan as containing a “helpful summary of principle and authority.” We respectfully agree.

147    In Jadwan an issue of causation arose which required the Court to consider the content of a hypothetical event concerning what advice a reasonable and prudent lawyer would have given one of the parties. The Full Court rejected the following submission put to it (at 311 [413]):

The fourth respondent also submitted that the possibility that one or more of the judges constituting this Court might, or might well have, formed a different view of the contested evidence or might, or might well have, reached different findings open on the evidence to those of the primary judge that were also open on the evidence, does not provide a principled basis for interfering with the findings of fact of the primary judge.

148    In answer to this submission, the Full Court referred to the following well known passage from Fox v. Percy (2003) 214 C.L.R. 118 at 126-127 [25]:

Within the constraints marked out by the nature of the appellate process, the appellate court is obliged to conduct a real review of the trial and, in cases where the trial was conducted before a judge sitting alone, of that judge’s reasons. Appellate courts are not excused from the task of “weighing conflicting evidence and drawing [their] own inferences and conclusions, though [they] should always bear in mind that [they have] neither seen nor heard the witnesses, and should make due allowance in this respect” [Dearman v Dearman (1908) 7 CLR 549 at 564, citing The Glannibanta (1876) 1 PD 283 at 287]. In Warren v Coombes [(1979) 142 CLR 531 at 551], the majority of this Court reiterated the rule that:

[I]n general an appellate court is in as good a position as the trial judge to decide on the proper inference to be drawn from facts which are undisputed or which, having been disputed, are established by the findings of the trial judge. In deciding what is the proper inference to be drawn, the appellate court will give respect and weight to the conclusion of the trial judge but, once having reached its own conclusion, will not shrink from giving effect to it.

149    The Full Court in Jadwan then said the following at 312 [414]:

What is set out in the above passage is subject to other principles of appellate review to which we have referred, including the requirement that an appellant show error in the primary judge’s decision, and the acknowledgement in the authorities of the circumstances in which the appellate court will not be in as good a position as the trial judge to decide on the proper inferences to be drawn. The weight to be given to the findings of a trial judge will vary according to the type of issue in question, and the nature and the extent of the advantage enjoyed by the trial judge. But if error is shown, a court of appeal may then be required to make its own findings of fact and to formulate its own reasoning based upon those findings if it is in a position to do so: Robinson Helicopter at [43]; and see also, Waterways Authority v Fitzgibbon [2005] HCA 57; (2005) 221 ALR 402 at [134]-[135] (Hayne J).

150    The principles expressed in the foregoing passage are, in our view, applicable to this appeal. In particular, the reference to the “position” of the trial judge referred to in the foregoing passage includes the advantage enjoyed here by the learned primary judge who had the opportunity to consider, and reflect upon, the entirety of the evidence as a whole. It follows that the task of this Court is to undertake a “real review” of the evidence, but to do so with a degree of restraint and deference when considering the findings of the learned primary judge. That is all the more so where, as here, those findings followed the review and consideration of a rather large volume of evidence.

The Capacity to Reconstruct a Transaction

151    As already mentioned, the learned primary judge decided that neither Div. 13 of the 1936 Act nor Subdiv. 815-A of the 1997 Act authorised the Commissioner or the Court to ascertain the consideration that might reasonably be expected to have been paid, or the profits that might have accrued, by reference to a pricing formula that was not price sharing and not quotational period optionality with back pricing. To do otherwise would have constituted an impermissible reconstruction of the agreement in fact entered into by C.M.P.L. and G.I.A.G. The submissions of the Commissioner and the taxpayer respectively sought to attack or support this conclusion.

152    The Transfer Pricing Guidelines applicable here refer to the need for a revenue authority to apply the “arm’s length principle” to the “transaction actually undertaken by the associated enterprises as it has been structured by them.” There are two exceptions to this basal principle which are expressed in the Guidelines as follows at para. 1.37:

The first circumstance arises where the economic substance of the transaction differs from its form.

…

The second circumstance arises where, while the form and substance of the transaction are the same, the arrangements made in relation to the transaction, viewed in their totality, differ from those which would have been adopted by independent enterprises behaving in a commercially rational manner and the actual structure practically impedes the tax administration from determining an appropriate transfer price.

153    We are hesitant to comment on the language used in these so-called exceptions. The language deployed is very highly generalised and is frustratingly opaque. Further, the discernible object of the Transfer Pricing Guidelines is that they are just that; they are only a guide as to how a revenue authority or a taxpayer might apply the “arm’s length principle”, or how an O.E.C.D. member country might enact the “arm’s length principle” into domestic law. In that respect, the various statements of abstract principle that may be found in the Transfer Pricing Guidelines may be contrasted with the much greater discipline and rigour in drafting that is usually found in domestic legislation. Of course, Subdiv. 815-A obliges the Court to work out whether an entity has got a transfer pricing benefit consistently with these Guidelines, but only to the extent they are relevant. In that respect, where a relevant principle is expressed in nebulous terms, it may not be of much real assistance to the task of applying Subdiv. 815-A to particular facts.

154    In any event, is it unnecessary to say anything more about the Transfer Pricing Guidelines because, and with very great respect to the learned primary judge, we are inclined to think the Commissioner did in fact apply Subdiv. 815-A and Div. 13 to the transaction actually undertaken by C.M.P.L. and G.I.A.G. All he sought to change was the consideration payable for the copper concentrate in fact supplied. Where parties to an agreement specify a price in dollar terms, or in the terms of another currency, Div. 13 gives the Commissioner a clear power to substitute a different price which he considers to be the “arm’s length consideration.” Upon making his determination under s. 136AD, that consideration is then “deemed” to be the relevant consideration received in respect of a given supply. Where parties do not specify an actual price, but rather agree that the consideration payable is to be determined by a formula or some other methodology, the Commissioner, in our view, is also permitted — in the sense of having a legal capacity — to substitute a different formula or a different methodology which he considers will result in the ascertainment of the arm’s length consideration. The word “consideration”, as Pagone J. observed in Chevron at 78-79 [133], “is not to be construed narrowly and includes that given by the acquiring party so as to move the agreement whether that be in money or in money’s worth”: Archibald Howie Pty Ltd v. Commissioner of Stamp Duties (NSW) (1948) 77 C.L.R. 143 at 152. We agree. It is at least wide enough to include a pricing formula or some other methodology for the determination of price.

155    Our conclusion that the Commissioner has this legal capacity is a necessary consequence of the terms of s. 136AD, which permit him to make a determination whereby the taxpayer is relevantly deemed to have received consideration equal to the arm’s length consideration. In that respect, the term “consideration” necessarily directs one to identify those clauses in an international agreement which define the price payable for the supply of goods or services. A distinction must therefore be drawn between such a clause or clauses and other clauses found within an international agreement, including those which may indirectly bear upon price. This distinction may be unsatisfactory, particularly in practice, but is nonetheless mandated by the text of Div. 13. Whether the Commissioner is authorised under Div. 13 to ignore or reframe those other clauses which do not define the price payable must be very seriously doubted. Again, that is because of the language of Div. 13. When the Commissioner makes his determinations under s. 136AD, the legal consequence is limited to the taxpayer being deemed to have, in this case received, arm’s length consideration as defined under s. 136AA. Consistently with the usual rule about deeming provisions, the statutory fiction mandated by Div. 13 should be construed strictly and only for the purpose which it serves: Federal Commissioner of Taxation v. Comber (1986) 10 F.C.R. 88 at 96 per Fisher J.; Financial Synergy Holdings Pty Ltd v. Federal Commissioner of Taxation (2016) 243 F.C.R. 250 at 259 [34] per Middleton and Davies JJ. There is thus no power or authority to substitute different terms of a contract where those terms are not seen as defining the consideration received, relevantly, for the supply of goods. Whether a term of a contract is to be characterised as such a term will need to be decided on a case by case basis.

156    In our respectful view, the same conclusion concerning the substitution of a different formula or methodology to ascertain the arm’s length consideration applies to Subdiv. 815-A. The “conditions” referred to in Art. 9 of the Swiss Treaty must include the consideration paid, and where parties have chosen to select a form of consideration which is ascertainable by the use of a formula or some other methodology, the “conditions” referred to in Art. 9 of the Swiss Treaty must include that formula or that methodology. In respect of the conditions in an agreement that only indirectly bear upon price, the extent to which the Commissioner can substitute different conditions if he considers that those conditions differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another is a question for another day.

157    In our view, the reasons of Allsop C.J. in Chevron support, at least with respect to Div. 13, the foregoing conclusion. His Honour observed at 52 [46]:

Whilst the property remains the same, what consideration would be given for it in a real world of independence may lead, depending upon the evidence, to the reasonable expectation of different behaviour on the part of the person in the position of the taxpayer in relation to the giving of consideration for the property and of behaviour by another or others in relation to the dealing, and which would reflect rational commercial behaviour in the environment of an arm’s length transaction. Such behaviour may affect the terms of the hypothetical agreement in question to the extent that they can be seen as part of the consideration.

(Our emphasis.)

158    As mentioned above, G.I.A.G. and C.M.P.L. chose to express the consideration payable for copper concentrate by reference to a methodology or formula which, for convenience, we will set out again. It is found in cl. 8 of the C.M.P.L.-G.I.A.G. agreement as follows:

The price per dry metric ton of copper concentrates shall be the sum of the payments less the deductions as specified below:

What follows in the agreement, it will be recalled, are clauses that address the quantification of the payments to be made, and a formula for ascertaining the applicable deductions or T.C.R.C.s, which for copper concentrate was the price sharing clause set out above, together with the quotational period optionality with back pricing clause also set out above. Relevantly, that was the formula or methodology chosen by G.I.A.G. and C.M.P.L. On one view, it might also include the freight allowance and the payment of any penalty, if applicable.

159    In our view, the Commissioner was entitled, under both Div. 13 and Subdiv. 815-A, to substitute a different methodology or formula for the ascertainment of the applicable deductions or T.C.R.C.s and for determining the quotational period. That is because the clauses dealing with the deductions that must be made, and which specify the means of determining the applicable quotational period, form part of the calculation of the consideration payable for the purpose of Div. 13, and are “conditions” for the purpose of Art. 9 of the Swiss Treaty.

160    We accept in accordance with the observations of the expert witnesses that, for example, price sharing is a “fundamentally” different methodology for the ascertainment of T.C.R.C.s than benchmark pricing. Indeed, as Mr. Ingelbinck observed, the use of price sharing to determine T.C.R.C.s amounted to an “approach” to pricing that was a “very different fashion” to benchmark pricing. Critically, as the expert evidence described above makes clear, each methodology involves a different way of managing or mitigating risk and is the product of differing commercial judgments about that issue and the copper market more generally. Mr. Ingelbinck agreed that this was so during his cross-examination. This can be seen in the following exchange:

MR DE WIJN:    Now, what I’m saying to you is that there is – I think you’ve already said – there’s no correlation between TCRCs and copper price?

 MR INGELBINCK:    Correct.

MR DE WIJN:    One of the advantages of a price-sharing agreement, you’ve already agreed, would be that you remove that element of volatility – that is, the lack of correlation between TCRCs and copper price?

 MR INGELBINCK:     If that is important to you, yes.

MR DE WIJN:    Well, it may be important – it will depend upon the mine whether that is or isn’t important to you?

 MR INGELBINCK:    That’s fair.

 MR DE WIJN:        It will be a commercial decision for the mine.

 MR INGELBINCK:    Correct.

MR DE WIJN:    One mine might say it doesn’t matter; the next mine might say, “It is important for me to eliminate that uncertainty”. You agree with that?

MR INGELBINCK:    It is possible. I would assume that in a lot of those circumstances there would be a little bit more evaluation as to the circumstances in which this happens.

MR DE WIJN:    I understand that, but that all goes to the percentage that’s set; doesn’t go to the desirability of a price-sharing agreement?

 MR INGELBINCK:    I’m okay with that, yes.

161    Mr. Ingelbinck had earlier accepted that he did not dispute that price sharing was a term legitimately used by parties dealing with each other at arm’s length; his real concern was the choice of a 23% rate under the C.M.P.L.-G.I.A.G. agreement.

162    Risk assumes a fundamental role in the pricing of copper concentrate. That is because, amongst other things, of the volatility of the copper price, the volatility of benchmark T.C.R.C. pricing, and the extreme volatility of spot T.C.R.C. pricing. The various methodologies for calculating T.C.R.C.s, which the experts agreed were variously used by arm’s length parties, represent different commercial choices designed to address these unknowns. It follows that the way in which a person might manage these risks is inextricably bound up with a determination of the price of copper concentrate. Different business people will make different decisions about how to manage risk. The decisions that are made may be based upon more objective matters, such as the state of a given mine, its expected production and production costs, and the involvement of a financier or financiers. It may also turn upon matters that more directly relate to the character of those controlling a business. This includes the formation of a commercial judgment about how the market will perform and a person’s appetite for risk. Some business people may be more conservative than others as to risk and even within conservative business circles people may yet still form different judgments about to manage risk. Some undoubtedly may make mistakes in predicting what is to happen that only become apparent with the benefit of hindsight. That such mistakes are made is unsurprising, and indeed is perhaps to be expected to some degree, given the difficulty, as recognised by Mr. Wilson, of accurately forecasting copper prices. For this reason, such mistakes do not of themselves necessarily reveal any unreasonableness in the formation of a judgment as to appropriate risk management. All of this can take place within a market in which parties deal with each other at arm’s length. It should not be supposed, in that respect, that an arm’s length deal is always perfect, or will always favour to the maximum extent one side of a bargain.

163    As was made clear in Chevron, Div. 13 contemplates a hypothetical transaction entered into by independent parties dealing at arm’s length with each other and Subdiv. 815-A contemplates hypothetical conditions which might be expected to operate between independent enterprises dealing wholly independently with one another. So, the decisive question for determination here, at least on the issue of price sharing, is what attitude or approach to risk taking in the copper concentrate market as at February 2007 should be attributed to these parties or enterprises.

The Personality of the Parties to the Hypothetical Contract

164    The Commissioner submitted that no evidence was led about C.M.P.L.’s or G.I.A.G.’s attitudes to risk taking as at February 2007. The taxpayer had therefore failed to show that C.M.P.L. would have wanted to secure the commercial benefits of using price sharing as a methodology, so it followed that such benefits were irrelevant. That conclusion was said to be open because both Div. 13 and Subdiv. 815-A require the hypothetical parties to the hypothesised copper concentrate transaction to be clothed with the very same attributes that C.M.P.L. and G.I.A.G. had in February 2007. In such circumstances, the Court was left with the stark evidence that in February 2007, and based on forecast information, C.M.P.L. was going to be worse off by agreeing to price sharing, and there was no other commercial reason as to why it had submitted to such an adverse pricing methodology.

165    In that respect, the presence of the pre-existing C.M.P.L.-G.I.A.G. agreement on different terms loomed large in the Commissioner’s case. His case was built upon a comparison of the benefits conferred on C.M.P.L. by that contract as compared with the reduced cash benefits which were expected to arise by reason of the amendments made in February 2007. Why, it was asked, would a party in the position of C.M.P.L. have agreed to such a debilitating change of terms? It simply was not what an independent party dealing at arm’s length with a buyer of copper concentrate would ever have agreed to.

166    With respect, we think the Commissioner has asked the wrong question. As a result, whether the C.M.P.L.-G.I.A.G. agreement fell within that range of hypothetical contracts for the sale of copper concentrate which independent parties dealing at arm’s length with each other might reasonably be expected to have entered into, did not feature in the Commissioner’s analysis. Because the Commissioner was not asking the correct question, he submitted that there was no range of possible outcomes for this particular mine, but just one outcome, namely retention of the pre-existing terms as they were just before February 2007.

167    The Commissioner’s submission suffers from a potential difficulty in applying Div. 13 of the 1936 Act. In Federal Commissioner of Taxation v. SNF (Australia) Pty Ltd (2011) 193 F.C.R. 149, this Court squarely rejected the Commissioner’s contention that the hypothetical taxpayer had to stand in the shoes of the actual taxpayer and be clothed with all of the attributes of the actual taxpayer. The Commissioner’s submission was recorded at 179-180 [96] of the judgment of the Full Court constituted by Ryan, Jessup and Perram JJ., as follows:

The Commissioner’s argument hinges, as he submitted in reply, on the proposition that one of the independent parties referred to in the definition of “arm’s length consideration” in s 136AA(3)(d) was the taxpayer. This was the case, so he submitted, because the definition provision had to “be read in the context of s 136AD, the operative provision” and that provision commenced “its inquiry with the ‘taxpayer’ and it is the ‘taxpayer’ which is the subject of the hypothetical in s 136AA.” The three propositions for which the Commissioner contends are, therefore:

(a)    the definition provision must be read in the context of the operative provision; and

 (b)    the operative provision — s 136AD(3) — begins its inquiry with the taxpayer;

(c)    therefore the hypothesis required by the definitive provision — s 136AA(3)(d) — must relate to the taxpayer so that “arm’s length” in that provision means “arm’s length from the taxpayer”.

168    The Court rejected the submission at 180 [98]-[99]:

[T]he description of a transaction as being at arm’s length is a statement about the independence of two parties from each other. The connexion thus disclosed is a relative one. Generally speaking a statement that two parties have a relative connexion of a particular kind does not carry with it any information about their absolute status. A requirement, for example, that two businesses be more than 20 km apart says nothing about where either business is situated. If one were to look at the definition provision in s 136AA(3)(d) in isolation it would be unsound to read it as requiring any more than that the two parties in question should be independent of each other; that is, the ordinary meaning is not as the Commissioner contends.

The question then is whether the ordinary meaning is somehow displaced or modified by the fact that the definition provision feeds into an operative provision — s 136AD(3) — which in turn utilises it to assess the position of the taxpayer. There is no doubt that s 136AD(3) is, as the Commissioner submits, about the taxpayer; that it requires a comparison between that which was actually paid by the taxpayer and an arm’s length consideration; and, that, in appropriate circumstances, it then substitutes one for the other. However, it does not follow from acceptance of all those features that arm’s length consideration — which does not, in general, refer to the actual position of either party — must be treated as overlaid by a further requirement that the consideration not only be at arm’s length but that the arm in question be attached to the taxpayer.

(Our emphasis.)

169    Chevron must be taken to have softened the foregoing conclusion, at least to an extent. The judgment of the Chief Justice, with respect to the application of Div. 13, recognised the great variety of arm’s length transactions that independent parties may negotiate. Each contract will be different but each will still be the product of an arm’s length dealing. As a result, Div. 13 should not be applied pedantically or inflexibly. As Allsop C.J. observed at 50-51 [42]:

There may be a free market into which disembodied independent third parties and the taxpayer alike could enter for substitutable or fungible goods; or there may be a market of sorts in which individually reached pricing will be available depending upon the precise characteristics of the party seeking to avail itself of the market. Given the great variety of commercial circumstances to which the provision may apply, it would be wrong either to approach the interpretation of the provisions pedantically or to dictate a rigid or fixed approach to the task of determining the arm’s length commercial consideration.

170    For that purpose his Honour observed that the inquiry does not necessarily require the detachment of the taxpayer as one of the independent parties to the hypothetical transaction. At 51 [43] the Chief Justice thus said:

There is no reason derived from the language of s 136AA(3)(d) why the hypothesis based on independence should, of necessity, do other than assess what the taxpayer or a person in the position of the taxpayer would be expected to give by way of consideration in respect of the acquisition of the property to a party independent from it. The independence hypothesis does not necessarily require the detachment of the taxpayer, as one of the independent parties, from the group which it inhabits or the elimination of all the commercial and financial attributes of the taxpayer being part of the circumstances that gave the commercial shape to the property the subject of the acquisition and that may be relevant to the consideration for the property.

171    After considering SNF, the Chief Justice was of the view that the “utter disembodiment of both parties” was not required by Div. 13. At 51-52 [44] his Honour said:

In SNF, the Commissioner’s submissions were directed to the asserted inadmissibility of evidence of certain allegedly comparable transactions because the parties did not have the same features as the taxpayer, being a history of making losses. Some of the language in SNF may be seen to be broader than was necessary to deal with the arguments and controversy in question. I do not take from SNF any requirement for a rigid or fixed approach to the place of the circumstances of the taxpayer or the party posited in the position of the taxpayer, in particular, here, its position in the group of which the other party to the actual transaction was a member also. Naturally, the one fixed and rigid proposition is that the parties to the dealing posited by s 136AA(3)(d) must be independent from each other — mutually independent. It does not follow that the party in the position of the taxpayer in the real transaction (here the borrower) must be disassociated in the hypothesis from its place in the group for whose interests it was borrowing. I do not read SNF as requiring the utter disembodiment of both parties from the circumstances of reality if one is seeking to understand not what a market price for goods was but the consideration that would be given for acquiring a characterised loan from an independent lender. The fundamental purpose of the hypothesis is to understand what the taxpayer, CAHPL, or a person in the position of the taxpayer and in its commercial context would have given by way of consideration in an arm’s length transaction.

172    The foregoing passages neither support a test which is the “utter disembodiment” of the actual parties from the hypothetical transaction, nor a test whereby the hypothetical party stands entirely in the shoes of the taxpayer. Critically, Allsop C.J. expressed the applicable test in the following way at 52 [45]:

The degree and extent of the depersonalisation will be dictated by what is appropriate to the task of determining an arm’s length consideration — that is one that satisfactorily replaces what the taxpayer gave by what it should be taken to have given had it been independent of its counterparty.

(Our emphasis.)

173    We shall return to this expression of the test. But there is another aspect to the hypothetical which we must emphasise at this point: the hypothetical “must be made to work”, and in order to make it work, one is required to draw upon those commercially rational practices adopted by independent parties operating in a particular market for goods and services. As Allsop C.J. said in Chevron at 53 [48]:

That the hypothesis must be made to work, if it can, is also to be taken from the commercially rational nature of the task — the property, the acquisition, the consideration are to be seen in their relationship to each other by reference to what can be reasonably expected, assuming independent commercial parties.

174    The reasons of Pagone J. are to similar effect. As his Honour observed at 73-74 [119]:

The hypothetical agreement contemplated by the definition of arm’s length consideration in s 136AA(3)(d) does not compel one of the parties necessarily to be the taxpayer (see Federal Commissioner of Taxation v SNF (Australia) Pty Ltd (2011) 193 FCR 149 at [9], [97]-[102]).

175    Justice Pagone was also of the view that in general, the hypothetical transaction for the purposes of Div. 13 had to remain “close” to the actual transaction and that the actual characteristics of the taxpayer must “serve as a basis” in the comparable agreement. As Pagone J. observed at 76-77 [128]:

The need to posit a hypothetical acquisition under an agreement for the purpose of evaluating it by reference to the standard of reasonable expectation requires a consideration of the evidence to determine a reliably comparable agreement to that which was actually entered into. That, as his Honour said at [499] required the hypothetical to remain close to the actual loan. The function of the hypothesis is to identify a reliable substitute consideration for the actual consideration which was given or agreed to be given, and the reliability of the substitute consideration depends upon the hypothetical agreement being sufficiently like the actual agreement. Thus, as his Honour held, the purchaser in this case need not be a hypothetical standalone company (see [79]) and was to be an oil and gas exploration and production subsidiary (see [80]). The characteristics of the purchaser must be such as meaningfully to inform an inquiry into whether the consideration actually given under the agreement exceeded the arm’s length consideration under the hypothetical agreement; or, to use the words of the learned trial judge at [80], in the hypothesis the independent parties are to have the characteristics relevant to the pricing of the loan “to enable the hypothesis to work”. The actual characteristics of the taxpayer must, therefore, ordinarily serve as the basis in the comparable agreement. That does not mean that all of the taxpayer’s characteristics are necessarily to be taken into account. The decision in SNF is an illustration of a feature of the taxpayer (namely that of having a history of incurring losses) being held not to be relevant to determining the arm’s length price of an arm’s length acquisition. In some cases the consideration that might reasonably be expected to be given in an agreement in which the parties were independent and dealing at arm’s length may be found in comparable dealings in an open market. What may readily be ascertained as the consideration in an open market for the property in question may supply the answer to the question but in each case the inquiry called for is a factual inquiry into the consideration that might reasonably be expected to be given in an agreement which did not lack independence between the parties and in which they dealt with each other at arm’s length.

176    The extent of depersonalisation here depends, to use the language of Allsop C.J., upon what is appropriate to the task of determining an arm’s length consideration. A number of considerations arise.

177    First, one should commence with the simple and uncontroversial proposition that it is only those attributes or features which can affect the consideration which is receivable which should clothe the hypothetical seller of copper concentrate.

178    Secondly, it is only the objective attributes or features which should be included. In SNF at first instance (SNF (Australia) Pty Ltd v. Federal Commissioner of Taxation [2010] FCA 635; (2010) 79 A.T.R. 193) Middleton J. observed at 201 [44]:

Just as in a valuation, the focus is not on the subjective or special factors of the parties involved in the transaction (for example whether they were financially sound or not), but is on the transaction itself and the consideration paid. In this sense, the task is not dissimilar to that undertaken in a valuation: see, for example, Boland v Yates Property Corporation Pty Ltd (1999) 74 ALJR 209 at 225-226 [82]-[83]; 167 ALR 575 at 596-597 [82]-[83] and Spencer v The Commonwealth (1907) 5 CLR 418.

179    Nothing said by this Court in SNF or in Chevron has led us to doubt the general correctness of this observation which we adopt. It means that one should include all of the objective circumstances of the actual C.S.A. mine, such as the means of production, the levels of production, the costs of production, the size of the mine, its location, and any problems arising from the location (e.g. water supply issues), and so on. It would include the objective circumstances of the copper concentrate market as at February 2007, including what was being reported by Brook Hunt, and what C.M.P.L. had budgeted and forecasted about the market. It would also include being a wholly owned subsidiary of a multinational natural resources group, but that group would not necessarily need to be the Glencore Group.

180    Thirdly, we think that it would be appropriate to exclude any considerations that are the product of C.M.P.L.’s non-arm’s length relationship with G.I.A.G. and the broader Glencore Group. In our view, that would include whatever attitude or policy C.M.P.L. had formed about the issue of risk when selling to G.I.A.G. Of necessity, any such attitude or risk would have been distorted by C.M.P.L.’s lack of independence from G.I.A.G. Inferentially, as a separate entity it is unlikely to have considered the issue of risk when selling to G.I.A.G., save for its attempt to comply with Div. 13 in the 2007 to 2009 years. It follows that the taxpayer’s failure to lead evidence about C.M.P.L.’s appetite for risk taking is not fatal. Nor is the failure to lead evidence about the Glencore Group’s policy about risk taking (if any). Whilst such a policy, if it existed, might have been relevant, it was also, for the reasons given below, open to the taxpayer to discharge its onus on this issue through the opinions of Mr. Wilson.

181    Fourthly, and in any event, because the issue of risk taking is so bound up here with the method or formula for determining the consideration payable for the sale of copper concentrate, the taxpayer was entitled to support the appropriateness of the particular formula chosen in February 2007 by reference to what an independent party in the position of C.M.P.L. might have done to address risk in the objective circumstances of the copper concentrate market at that time in selling either to an independent trader or smelter. For that purpose, it could legitimately adopt a more conservative approach to risk so long as it was commercially rational to do so, and it is what an independent party dealing at arm’s length might reasonably be expected to have done. The Commissioner was entitled to do likewise in support of a different pricing formula. Such a conclusion is consistent with Div. 13 imposing an objective test: W.R. Carpenter Holdings Pty Ltd v. Federal Commissioner of Taxation (2007) 161 F.C.R. 1 at 8 [27].

182    Importantly for this case, in our view it was open to the taxpayer to form its own commercial judgment about how risk was to be assessed as at February 2007 in the hypothetical transaction between independent parties. It was also open for Mr. Wilson to form such a commercial judgment arising from the fact that the C.S.A. mine was a high cost venture, so long as that judgment was the expression of what an independent party acting at arm’s length might reasonably be expected to have adopted. On behalf of the Commissioner, it was also open for Mr. Ingelbinck to perform a similar exercise. It follows that choices which are open to be made about risk may affect the determination of the arm’s length consideration. It also follows that there is likely to be more than one price which is an arm’s length price. In that respect, a taxpayer is under no obligation to choose a pricing methodology which pursues profitability in Australia at the expense of prudence. There is no obligation to “maximise” profitability at the expense of all else.

183    Fifthly, the possibility of a range of arm’s length outcomes, each of which would be sufficient to answer the statutory test, is supported by authority. As this Court said at 187-188 [125] in SNF:

But that does not mean, more generally, that there is only one arm’s length consideration. Often enough, for example, goods will change hands at prices which are different to the market value for perfectly legitimate reasons such as a need to secure long term or large volume arrangements or with traded securities, a premium for control and so on. No doubt, it was for similar reasons that Dr Becker, the Commissioner’s own witness, in response to the question ‘[a]nd you accept that, typically, there’s not one arm’s length price for a particular product?’ answered ‘[t]hat’s correct, yes’.

184    Sixthly, what controls the range of acceptable arm’s length outcomes is the concept of what might reasonably be expected. As Pagone J. observed in Chevron, that concept calls for evidence which supports a “sufficiently reliable” prediction which can be seen as reasonable. As his Honour said at 76 [127]:

The standard of reasonable expectation found in the words “might reasonably be expected” in s 136AA(3)(d) calls for a prediction based upon evidence. In Federal Commissioner of Taxation v Peabody (1994) 181 CLR 359 the High Court said at 385:

A reasonable expectation requires more than a possibility. It involves a prediction as to events which would have taken place if the relevant scheme had not been entered into or carried out and the prediction must be sufficiently reliable for it to be regarded as reasonable.

The prediction contemplated by Division 13, like that contemplated by s 177C of the 1936 Act, involves an evaluative prediction of events and transactions that did not take place but the prediction must be based upon evidence and, where appropriate, upon admissible, probative and reliable expert opinion: see Federal Commissioner of Taxation v Futuris Corporation Ltd (2012) 205 FCR 274 at [79]-[81]; see also Peabody v Commissioner of Taxation (1993) 40 FCR 531 at [39] (Hill J).

185    The Commissioner in his written submissions appeared to contend that the hypothetical mandated by Div. 13 required a taxpayer to prove what independent parties would have agreed to be the arm’s length consideration, rather than what independent parties might reasonably be expected to have paid or received. He relied on certain sentences in Chevron which appeared, on one view, to adopt a “would” test, but clarified during the hearing before us that he accepted that Div. 13 mandates a “might” test. We respectfully agree with this clarification. We do not think that those sentences in Chevron were intending to convey a test at odds with the language of s. 136AA(3)(c) and (d) of the 1936 Act. The statutory test relevantly requires the hypothetical price to be ascertained by reference to what might reasonably be expected to have been received or paid if the property had been supplied under an agreement entered into between independent parties dealing at arm’s length with each other. In our view, Art. 9 of the Swiss Treaty mandates a relevantly analogous inquiry. It refers to what might be expected and not to what would be expected.

186    Finally, in applying the foregoing a degree of flexibility and pragmatism is required. Whilst the onus remains on the taxpayer to discharge its onus of proof of demonstrating excessiveness in the amended assessments, one should not apply Div. 13, or indeed Subdiv. 815-A, narrowly. Predicting how independent parties dealing at arm’s length with each other would price a wholly controlled transaction is a difficult and complex issue. That is especially so when one integer which here directly affects the consideration payable is the formation of a commercial judgment about risk taking. The Court should acknowledge, and take into account, the practical difficulties faced by both the taxpayer and the Commissioner in finding evidence that grounds what is sufficiently reliable, or which demonstrates that something is insufficiently reliable. The answer is not always to be found in overly lengthy and complex expert reports. Common sense is required.

187    The foregoing seven propositions are what we consider to be relevant in deciding the “appropriate” degree of “depersonalisation” here, to use the language of the Chief Justice in Chevron, given the particular facts and circumstances of this case.

188    Something should be said at this point about the focus of Mr. Ingelbinck’s opinion, and thus also of the Commissioner’s case, upon the proposition that if C.M.P.L. had been dealing independently and at arm’s length with G.I.A.G., it would never have agreed to the amendments made to the pricing formula in February 2007. In that respect, we think the Commissioner established that, from the perspective of C.M.P.L.’s earnings, it might have been expected in February 2007 for those earnings to decrease with the adoption of price sharing. But in our respectful opinion, the ultimate issue for determination is not whether an arm’s length party would have agreed to the amendments, given the pre-existing terms of trade. We agree with the submissions of the taxpayer that the framing of the issue in this way is at odds with the text and purpose of Div. 13 and Subdiv. 815-A. Rather, the relevant issue is whether the consideration received by C.M.P.L. in the 2007 to 2009 years was less than the arm’s length consideration, as that term is defined in s. 136AA(3) of the 1936 Act, for the copper concentrate in fact supplied. The answer to that question does not turn upon whether the amendments made in February 2007 were in C.M.P.L.’s interests. It may be the case that both the pre-existing C.M.P.L.-G.I.A.G. agreement, and the agreement as amended in February 2007, included pricing formulae that arm’s length parties might reasonably be expected to have adopted. Both may fall within the range of arm’s length outcomes. In making this observation, we do not think that the pre-existing terms were irrelevant to the application of Div. 13 here. They form part of the objective history which may bear upon the issue as we have defined it.

189    We turn to consider the issue of “personality” under Subdiv. 815-A and Art. 9 of the Swiss Treaty. As we understood the judgment of Chief Justice in Chevron, Subdiv. 815-A applies in an analogous way to Div. 13. Both must be applied flexibly. As his Honour observed at 62 [90]:

[T]he causal test in s 815-15(1)(c) based on Art 9 is a flexible comparative analysis that gives weight, but not irredeemable inflexibility, to the form of the transaction actually entered between the associated enterprises. A degree of flexibility is required especially if the structure and detail of the transaction has been formulated by reference to the group relationship and a “tax-effective” outcome (even if, as here, one that is not said to be illegitimate). The form of that transaction may, to a degree, be altered if it is necessary to do so to permit the transaction to be analysed through the lens of mutually independent parties.

190    To similar effect Pagone J. said at 90-91 [156]:

The comparison which Art 9 required to be undertaken is akin to that contemplated by Div 13. The object was to determine whether conditions actually prevailing between the relevant enterprises differed from those which might be expected to operate if they had been independent and had been dealing wholly independently with each other. The hypothetical in that exercise is undertaken for the purpose of determining whether the dealing which actually occurred might have been expected to occur on different terms. That will generally require that the parties in the hypothetical will generally have the characteristics and attributes of the actual enterprises in question. The comparison required by Art 9 is expected to be undertaken in a practical business setting of potential transactions able to be entered into and which are to be used as a basis for a reliable hypothesis upon probative material. Ultimately the question was that of determining whether profits might have been expected to accrue to CAHPL if the transaction it entered into with CFC had been entered into where the conditions which operated between CAHPL and CFC did not operate, that is, where CAHPL and CFC had been dealing with each other wholly independently. The hypothetical thus required hypothesising circumstances in a dealing between an enterprise like CAHPL and an enterprise like CFC where, however, the conditions operating between them were between independent enterprises dealing wholly independently with each other.

191    In our view, one of the “conditions” that operated between C.M.P.L. and G.I.A.G. was the price sharing formula adopted from February 2007. As Allsop C.J. observed in Chevron (at 60 [82]), the word “conditions” is “broad and flexible”. In our opinion, it is plainly apt to include those terms in the contract which defined the consideration C.M.P.L. was to receive. The task for the taxpayer was thus to demonstrate that the pricing formula established by those terms did not differ from those formulae which might be expected to have operated between independent enterprises dealing wholly independently with one another in the copper concentrate market at the time. In that respect, it was again entirely open for the Commissioner to contend that the pricing formula did so differ, and that as a result profits that might have been expected to accrue under a different arm’s length pricing formula had not so accrued. For analogous reasons to those set out above, because risk and the pricing formula were inextricably bound up with each other, it was open for either party to lead evidence about how independent enterprises dealing wholly independently with one another might be expected to have assessed the issue or issues of risk as at February 2007. The failure by C.M.P.L. to lead evidence about its actual risk appetite or that of G.I.A.G. or the broader Glencore Group did not foreclose C.M.P.L.’s ability to lead expert evidence more generally about, and make submissions concerning, what independent enterprises might have done to address the issue of risk.

The Comparable Contracts

192    Something briefly should be said about the contracts said to be comparable to the C.M.P.L.-G.I.A.G. agreement. As already mentioned, the Commissioner sought to render them irrelevant by pointing to differences as to the size of tonnage, the time when the agreements were entered into, the stage in production, the role of financiers and so on. The Commissioner submitted that there was no evidence of any attempt to make adjustments for these differences. The taxpayer criticised this attempt to render irrelevant the contracts it adduced, relying upon SNF. On one view, the Commissioner’s submissions were contrary to what Middleton J. at first instance in SNF had said was necessary. At 201 [44] his Honour said:

The essential task is to determine the arm’s length consideration in respect of the acquisition. One way to do this is to find truly comparable transactions involving the acquisition of the same or sufficiently similar products in the same or similar circumstances, where those transactions are undertaken at arm’s length, or if not taken at arm’s length, where suitable adjustment can be made to determine the arm’s length consideration that would have taken place if the acquisition was at arm’s length.

193    In our view, many of the differences identified by the Commissioner are relatively important. They diminish the probative value of the contracts said to be comparable. But they do not negate that value entirely. The contracts were valid “reference points” both for the purpose of considering the type of pricing formula chosen by C.M.P.L. and G.I.A.G. under the C.M.P.L.-G.I.A.G. agreement, and also, in a more general sense, both the rate of price sharing and the detail of the quotational period optionality which was selected. The contracts were a sounding board. They confirmed the joint opinion of the experts that there was nothing in the pricing formula adopted from February 2007 that did not also exist in contracts between independent market participants. They also demonstrated that price sharing of 23% was not out of the market. Because of the differences identified by the Commissioner, the contracts cannot be determinative of the application of Div. 13 or Subdiv. 815-A to the facts here. However, the matters identified above demonstrate that the contracts were relevant and admissible pursuant to ss. 55 and 56 of the Evidence Act 1995 (Cth.), notwithstanding the differences identified by the Commissioner. Although one is directed by s. 815-20 of the 1997 Act to have regard to the Transfer Pricing Guidelines for the purposes of Subdiv. 815-A, the relevant standard for admissibility prescribed by ss. 55 and 56 remains the same under that Subdivision.

Application of the Foregoing Principles to the Facts

(i) Price Sharing

194    Dealing first with price sharing, Mr. Wilson was a highly experienced market analyst with 36 years of accumulated knowledge about the copper concentrate market. That knowledge was sought after by participants in that market, including miners, smelters and traders. Mr. Wilson gave a series of reports which gave sensible reasons for selecting price sharing as a valid means of determining T.C.R.C.s. Indeed, Mr. Ingelbinck ultimately did not disagree with the selection of a price sharing clause as at February 2007 as one that might reasonably be expected to have been used by independent parties dealing with each other at arm’s length. His concern was with the rate selected. But Mr. Wilson chose that rate in a rational and commercial way. He picked the mid-point in the historical range identified by Brook Hunt. He had never seen a rate that was less than 20% and the contracts he examined used similar rates. In contrast, the Commissioner led no evidence from Mr. Ingelbinck, or from anyone else, as to what the rate should be. Mr. Ingelbinck suggested that the starting point might have been 8.1%, but he agreed that no independent party buying from C.M.P.L. would have agreed to a rate for price sharing as low as this. He nonetheless said, however, that if in 1999 C.M.P.L. and G.I.A.G. had entered into an agreement which used a price sharing clause with a 23% rate, this “would have been fine.” The Commissioner’s own Audit Position Paper recorded that the T.C.R.C.s determined between C.M.P.L. and G.I.A.G. as a percentage of sales (with sales being a direct function of the copper price) from 2000 to 2006 had varied from 11.6% to 26.8%; it recorded the median rate for this period as being 20.7%.

195    For the reasons given above, in our view it was permissible for Mr. Wilson to give evidence about how a commercially rational miner operating a high cost mine might make choices about managing risk as at February 2007. His evidence was in this respect not speculation, as the Commissioner contended, but an admissible opinion. It was also permissible for Mr. Wilson to hypothesise, as at February 2007, a more conservative miner who might be prepared to trade earnings for certainty given the extreme volatility in the copper prices in 2006, and given the potential exposure arising from the use of benchmark T.C.R.C.s which might in the future, as they had in the past, move in the opposite direction to the copper price. Mr. Ingelbinck agreed that the owner of a high cost mine might tend to be more cautious. We do not think the hypothetical miner’s membership of a hypothetical multinational group affects the validity of Mr. Wilson’s opinion. In our view, the relevance, if any, of being a member of a group of companies would be a matter for an expert to consider, but this aspect of the hypothetical did not seem to matter either to Mr. Wilson or Mr. Ingelbinck.

196    We also accept that this “certainty” included effective immunisation from G.I.A.G. exercising its right of termination over three years, although this was not a matter raised by Mr. Wilson, but rather by the taxpayer’s senior counsel. We also accept the taxpayer’s submission that this possible benefit was not considered by Mr. Ingelbinck in his evidence. But as Mr. Ingelbinck remarked, a copper concentrate producer which does not have a home to monetize its production has a serious problem — the benefit of this certainty to C.M.P.L., conferred as it was at a time of significant market volatility, was therefore potentially substantial. It does not, in our view, matter that there was no evidence that G.I.A.G. had threatened to terminate the contract. C.M.P.L. was its wholly owned subsidiary, so it is most unlikely that any such evidence would have existed. Moreover, like the issue of risk, this is a matter to be tested by reference to what a reasonable and prudent independent miner in the position of C.M.P.L. might reasonably be expected to have considered as beneficial in a contract for the sale of copper concentrate to an independent purchaser. In that respect, we reject the submission made by the Commissioner in his reply to the Court that an entity in the position of C.M.P.L. would not have traded earnings for risk because C.M.P.L. had substantial reserves and because an entity in the position of C.M.P.L. could have relied on the support of its parent. Mr. Ingelbinck never expressed that opinion. Moreover, the type of “support” was never identified.

197    In short, and in simple terms, in our view, for the reasons given by Mr. Wilson, giving up forecast earnings to secure a more prudent or conservative outcome on risk was a result that arm’s length parties might be expected to have wanted in early 2007. Our conclusion is strongly fortified by the fact that, even with the changes made to the C.M.P.L.-G.I.A.G. agreement in February 2007, based on the forecasts at the time, C.M.P.L. was still going to be earning profits described as “healthy.” Of course, we accept that other independent parties could have chosen different terms as at February 2007 to price the C.S.A. mine’s copper concentrate.

198    We agree generally with the submission of the taxpayer that Mr. Ingelbinck’s opinion should be treated as an expression of a disagreement with Mr. Wilson’s commercial judgment. In that respect, we think we should take into account that Mr. Ingelbinck’s experience was as a trader selling to smelters; indeed, he worked for the largest producer of copper concentrate in the world. In contrast, Mr. Wilson’s experience, though gained in a less direct way, was more diverse as it traversed the entire copper concentrate market. Mr. Ingelbinck expressed strong disagreement with Mr. Wilson’s commercial judgment about the attractiveness of price sharing. He formed the view that its benefits were outweighed by expected foregone earnings. He said that a miner in February 2007 would have been more concerned with forecast benchmark T.C.R.C.s and copper prices rather than with historical data, such as the range of price sharing relied upon by Mr. Wilson (although the force of this criticism is deflated by the fact that the pre-existing contract only used benchmark T.C.R.C.s for 50% of the Base Tonnage). We accept the force and validity of Mr. Ingelbinck’s views. But it does not follow from that acceptance that we should also reject Mr. Wilson’s commercial judgment. On more than one occasion, Mr. Ingelbinck accepted that miners may reach different commercial views about how to manage risk. We have set out above one passage of his cross-examination where he accepted this. On another occasion he was asked whether a negotiator would take into account the fact that price sharing delivered more certainty as a commercial factor. The exchange was as follows:

MR DE WIJN:     But, of course, that’s only the starting point because you would have to factor in as a commercial negotiator the fact that a price-sharing agreement gave you some more certainty, not complete certainty, but some 10 more certainty by eliminating the volatility between TCRCs and copper price?

MR INGELBINCK:    Well, that becomes a question of whether you really want it so badly that you’re willing to pay for it versus if I can get it and I want it, I’ll do it.

MR DE WIJN:    Again, it depends upon a business judgment as to whether you want the advantages that a price-sharing agreement brings and are prepared to pay for it?

 MR INGELBINCK:    Correct.

MR DE WIJN:    It’s a bit like insurance. You may not want to insure your house because you don’t think it’s going to burn down but most of us do because they want the protection?

MR INGELBINCK:    That’s fair, although I don’t see the analogy with price-sharing giving you insurance but - - -

MR DE WIJN:    Well, it doesn’t give you insurance but it does provide protection against the lack of correlation between TCRCs and copper price?

MR INGELBINCK:    It avoids that lack of correlation, but that could be positive or negative, so you’re giving something up that’s not necessarily negative.

(Our emphasis.)

199    In our view, and using the language of Mr. de Wijn, senior counsel for the taxpayer, Mr. Wilson validly hypothesised a miner who as a member of a hypothetical multinational group was “prepared to pay” for certainty by adopting a price sharing clause.

200    In another passage of Mr. Ingelbinck’s cross-examination he accepted that participants in the market could have different views about the benefits of a price sharing clause. Here is that passage:

MR DE WIJN:    Now, when you take into account all of those factors, I suggest to you that, if you were sitting there in 2006/2007 and you were a miner wanting the advantages of a price-sharing agreement and, for that matter, a trader wanting the advantages, you might start at 8.1 per cent but probably move that figure somewhat higher?

MR INGELBINCK:    I have difficulty in answering that question because I cannot really see myself sitting there in 2006, early 2007 even having that consideration. It was just a little too outlandish. But if we go to the theoretical exercise of 8.1 per cent, or whatever the number was, then, again, it would be a function of how strongly does the miner really want that, and I can’t address that. You see, I do not share the view that it’s such an attractive thing and the elimination of one variable doesn’t really do much, in my opinion.

MR DE WIJN:    No. That’s your opinion, but you accept that different people in the market could have a different opinion about the advantages of price-sharing?

MR INGELBINCK:    They could.

MR DE WIJN:    And your issue with price-sharing, as I think you’ve agreed, is not so much the price-sharing agreement but the percentage that was set?

 MR INGELBINCK:    In this particular case, yes.

201    Again, whilst Mr. Ingelbinck did not agree with the 23% rate chosen in 2007, he did not himself venture what rate might have been struck by arm’s length parties, although he did accept that it would be a matter that would need to be negotiated. Thus, he gave the following answers when cross-examined:

MR DE WIJN:    And you accept, do you not, that if we were wanting to establish a whole-of-mine offtake agreement for a three-year period, with a trader, that you would have to come to a deal that satisfied the trader as well as satisfied the mine?

 MR INGELBINCK:    I’m okay with that, yes.

MR DE WIJN:    And the trader is, of course, going to look after its own interests and want to push the percentage as high as it can?

 MR INGELBINCK:    Yes.

MR DE WIJN:     The miner, in this hypothetical bargain, is going to want to push the position lower than it might otherwise be?

 MR INGELBINCK:    Indeed, yes.

MR DE WIJN:    And at some point in time, if you assume the parties want to have the advantages of a price-sharing agreement, they will come together at a figure?

 MR INGELBINCK:    Yes.

MR DE WIJN:    And normally with a negotiation, the best negotiation is one that neither party is happy with?

 MR INGELBINCK:    Correct.

202    As best as we can tell from the Commissioner’s submissions, C.M.P.L. should either have stuck with the T.C.R.C. clause which existed before February 2007, or have bargained a rate of price sharing that did not trade future earnings for greater future certainty of pricing. The C.S.A. mine was a high cost mine, and the operator of such a mine would have wanted to achieve the lowest possible T.C.R.C.s. The operator would also have wanted to maximise profitability. It may readily be accepted that an operator of the C.S.A. mine might reasonably be expected to have sought such objectives. But for the reasons expressed by Mr. Wilson, they are not the only commercial objectives which an operator of the C.S.A. mine might reasonably have been expected to pursue. There will ordinarily be a range of rational commercial behaviours that an entity may legitimately wish to pursue. That range can include incurring losses in order to, for example, pursue a long term strategy of market penetration. This is what the taxpayer was doing in SNF.

203    It is, of course, a very difficult thing for this Court to decide how much revenue an independent miner operating the C.S.A. mine in early 2007 might have been expected to forego to secure the benefits conferred by price sharing. The Commissioner submitted that there was no evidence before the Court that it made sense to trade foregone revenue for the benefits of price sharing as identified by both Mr. Wilson and Mr. Ingelbinck. With respect, the only calculation of any foregone revenue was made using benchmark T.C.R.C.s, rather than the terms of the pre-existing C.M.P.L.-G.I.A.G. agreement as it was just before February 2007. The terms of that agreement, said by the Commissioner to represent an arm’s length agreement, provided that benchmark T.C.R.C.s were only to be used as a basis to price 50% of the Base Tonnage. And then there is the expert opinion of Mr. Wilson which supported the reasonableness of the 23% rate. He thought that this was a commercial and prudent rate. No other rate was put to him and the Commissioner led no expert evidence about a possible competing rate. We accept Mr. Wilson’s judgment. It was logically based on the historical Brook Hunt data. The Commissioner never challenged the correctness of that data. By its nature, transfer pricing is not an exact science. In the circumstances, picking a mid-point in the Brook Hunt data was sound. It was a sufficiently reliable choice and accorded with common sense. In that respect, the Court must take care not to make the task of compliance with Australia’s transfer pricing laws an impossible burden when a revenue authority may, years after the controlled transaction was struck, find someone, somewhere, to disagree with a taxpayer’s attempt to pay or receive arm’s length consideration.

204    Should Mr. Wilson’s opinion be rejected because of the concessions made by him when he was cross-examined by the Commissioner? On balance, and with great respect to the Commissioner’s senior and junior counsel, we think not. The concession made about the threat to the commercial viability of the C.S.A. mine was predicated upon a key factual assumption with which Mr. Wilson did not agree, namely a concern about escalating costs. In our view, that factual assumption was not demonstrated to be true. Rather, the learned primary judge made findings at [120] (set out above) that are directly inconsistent with the assumption which Mr. Wilson was asked to make. Her Honour found that as at late 2006 and early 2007, the mine had plans and steps in place to address, amongst other things, its financial challenges. Her Honour also expressly found that “there was nothing in the contemporaneous documents which identified any uncertainty for the mine in relation to its capacity to continue mining in 2007, 2008 or 2009 or to suggest that CMPL considered that there was any real risk that its level of production would not continue throughout those years.” These findings were not challenged on appeal. Ironically, the foregoing conclusions are consistent with the Commissioner’s contention below (see for example at [144]) that the C.S.A. mine was in a strong position.

205    More fundamentally, at [156] the learned primary judge said:

I accept Mr Kelly’s evidence that the switch to price sharing was one that might be expected to have “guaranteed the viability” of the mine. The taxpayer’s unchallenged calculations of the margins for the 2007, 2008 and 2009 years of the 23% price sharing based on CMPL’s forecasts in the 2007 Budget, correctly accounting for the currency difference in the line items, showed margins over total cost as a percentage of copper price and C1 cost as follows:

Budgeted margin over total cost as a % of copper price:

2007: 34.63%

2008: 25.39%

2009: 27.39%

Budgeted margin over C1 cost as a % of copper price:

2007: 39.09%

2008: 30.22%

2009: 31.51%

206    The reference to the “C1” cost is to the direct cash costs, which include mining, milling, on-site administration, smelting, refining, concentrate freight and marketing costs, amongst others, net of by-product credits (such as credits for the presence of silver).

207    These findings were not specifically challenged by the Commissioner. It follows that Mr. Wilson’s concession about commercial viability is of no moment.

208    The next question is whether Mr. Wilson’s opinion should be rejected because he made the following two further concessions:

(a)    that by adopting price sharing with a rate of 23%, and based on the existing forecasts, C.M.P.L. was going to be “worse off financially”; and

(b)    that if one assumed that C.M.P.L.’s aim was to be as profitable as possible, the commercially rational thing to do in February 2007 would have been to stick with the pre-existing terms.

209    There are four difficulties with relying upon these concessions to contend that Mr. Wilson’s opinion should be rejected. First, it was not established how “worse off” C.M.P.L. was going to be as forecast. The figures Mr. Ingelbinck derived assumed a contract that relied entirely upon benchmark T.C.R.C.s to set the T.C.R.C.s; yet this was not the contract G.I.A.G. and C.M.P.L. were parties to in the period just before February 2007. Secondly, the concept of being “as profitable as possible” is, with respect, ambiguous for the reasons we have already given. It is apt to cover a range of outcomes. This diminishes greatly the utility, for the Commissioner’s case, of the answers Mr. Wilson gave. Moreover, in the context of this line of cross-examination we would construe the reference to aiming to be as “profitable as possible” as probably being to profitability at the expense of managing risk in the way Mr. Wilson considered prudent. Thirdly, we do not accept that one is required to attribute to the hypothetical miner the Glencore Group’s policy, if it can be so called, of maximising the profit of an asset for the Group, assuming that this policy has a particular meaning. Because the issue of risk is so bound up with the pricing of copper concentrate, for the reasons we have already given, it was open to Mr. Wilson to hypothesise a prudent or conservative miner. This conclusion did not depend upon the existence of lay evidence in support of it. Fourthly, the answers given assumed the correctness of the forecasts on which they were based. Yet Mr. Wilson’s view was that the owner of a high cost mine might not be prepared to make that same assumption, or would only make it in a qualified way. In cross-examination Mr. Wilson said the following about the forecasts in Brook Hunt:

I would, however – I think it would be important, in the context of these numbers, to just look at a couple of examples in here where I think, certainly a high-cost producer might be somewhat concerned about the forecast.

210    We have set out a number of passages from Mr. Wilson’s cross-examination where he emphasised how a high cost producer might have been nervous about how the forecast T.C.R.C.s might have played out, and that forecasting was not “straightforward” due to the very many variables and various “moving parts” in the market. The 2006 year had been an unprecedented year for the copper market. Even Mr. Ingelbinck thought that participants in it had been “confused.” As mentioned above, we do not characterise these observations by Mr. Wilson as “speculation.” He did not seek to give evidence about any actual particular participant in the copper market. He candidly accepted that he had never been in the position of a high cost producer. Rather, he was giving an opinion about a generic participant in the copper market operating a high cost mine. In our view, he was well qualified to do this.

211    Even Mr. Ingelbinck recognised that there could be issues with forecasting. He said that, whilst “one does as well as one can”, there are obviously a number of unexpected events that could take place that could not be forecasted. In cross-examination he agreed that at the end of 2006 there was no particular consensus about a specific forecast for either T.C.R.C.s or copper prices, with the latter being “actually even more difficult to predict” than T.C.R.C.s. That being said, he maintained his view that as at the end of 2006, T.C.R.C.s were more likely to go lower over the 2007 to 2009 years.

212    For these reasons, the foregoing two concessions do not require us to reject Mr. Wilson’s opinion. It also follows that for the foregoing reasons, the taxpayer has established that independent parties dealing at arm’s length with each other for the sale of the copper concentrate in fact sold by C.M.P.L. to G.I.A.G. from 2007 to 2009 might reasonably be expected to have agreed to a price sharing clause at the rate of 23% as part of the calculation of the T.C.R.C.s. The learned primary judge did not err in reaching this conclusion.

213    What is one then to do with Mr. Ingelbinck’s opinion? In our view, his opinion that the parties should have adopted benchmark T.C.R.C.s appears to us to represent another possible position that arm’s length parties might reasonably be expected to have adopted. But the existence of this possibility does not negate our finding that price sharing using a 23% rate was also an arm’s length outcome. It was, nonetheless, not an outcome Mr. Ingelbinck favoured. In our view, this is a case where reasonable minds have reasonably differed within a range of commercially acceptable arm’s length outcomes. That might explain why the Commissioner was prepared to accept such a significant divergence between what he considered to be an arm’s length agreement (i.e. the preceding agreement just before February 2007) and what Mr. Ingelbinck considered to be an arm’s length agreement.

214    No different outcome arises under Subdiv. 815-A. As already mentioned, it was not suggested that there could be different outcomes here depending upon which set of transfer pricing provisions applied. Both parties presented their respective cases on an indivisible factual basis that applied both to Div. 13 and to Subdiv. 815-A.

(ii)    Quotational Period Optionality

215    We turn now to consider the quotational period optionality with back pricing clause. It will be recalled that the only change made in February 2007 to this clause was to confer on G.I.A.G. the ability to choose either option a) or b) on a shipment by shipment basis. There are a number of problems with the Commissioner’s case on this clause.

216    First, the Commissioner accepted that the clause as it existed prior to February 2007 represented an arm’s length outcome. Yet in this state it already included features, such as back pricing, which Mr. Ingelbinck said were not arm’s length in the sense that such features should not have been conferred on G.I.A.G. without a quid pro quo. Indeed, the Commissioner attacked back pricing and some of the additional optionality given to G.I.A.G. as being non-arm’s length, in the face of his acceptance of these attributes as being appropriate. The contradiction was never resolved.

217    Secondly, Mr. Ingelbinck did not opine that the clause was not one which arm’s length parties might reasonably be expected to have included in an agreement for the sale of copper concentrate. Rather, his complaint was that C.M.P.L. had received no discernible quid pro quo, such as the discounted T.C.R.C.s seen in the Barminco Contract. This issue itself created multiple difficulties for the Commissioner:

(a)    The type of quid pro quo which was justified was never identified. Nor was it ever quantified — neither as a specific figure nor as a range. Instead, the Commissioner’s experts were of the view that it was very hard, “if not impossible”, to place a precise value on back pricing. In the joint expert report, the following was said:

Mr lngelbinck further stated that while back-pricing could be very profitable, there was no guarantee that a trader could benefit on a shipment-by-shipment basis because it would depend on how the market prices were evolving and also the decisions made by the trader during the pricing period. All three experts agreed that the trader could at worse make no gain from a particular shipment involving back-pricing, but would not lose out from back-pricing. The impact on the seller can be both positive as well as negative depending on market circumstances and is not directly related to the gains, if any, realized by the buyer. Mr. lngelbinck further specified that any gains realized by the buyer could not have been achieved on the same risk-free basis had it not been for the optionality granted by the seller. He walked the Registrars and the other expert witnesses through a number of examples which confirmed a wide variety of possible outcomes which ranged from breakeven to significant profits for the buyer and both positive and negative possible outcomes for the seller (see Annexure 1). For the years 2007 through 2009, Mr Wilson mentioned that had GIAG, as buyer, been limited to selecting one early or one late QP in each year prior to the actual QP execution, rather than on a shipment-by shipment basis including back-pricing, the outcome on CMPL’s net revenues over this period may or may not have been significantly different from the actual outcome depending upon the QPs selected. Mr Kowal’s and Mr Wilson’s analysis contained in their various reports illustrate an impact ranging from ~$10 MUS to ~$85 MUS dependent on the QPs selected.

The Commissioner submitted that this difficulty did not matter; the onus was on the taxpayer to show that it had received the correct quid pro quo.

(b)    Quotational period optionality and back pricing were introduced in 2004 and 2005. It is not at all clear to us whether the quid pro quo said to be owing to C.M.P.L. might have been wholly given in those years.

(c)    As already mentioned, the Commissioner’s case was that an arm’s length party would have stuck to the terms of the C.M.P.L.-G.I.A.G. agreement as they existed just prior to February 2007. The only change made then was to introduce shipment by shipment optionality. It was not clear to us whether Mr. Ingelbinck considered that an additional quid pro quo should have been received for this amendment. He does not mention it in any of his reports, although when giving oral evidence he did say that it justified the payment of greater consideration to C.M.P.L. Even then, however, it was very far from clear to us what this should have been.

218    Thirdly, compounding matters for the Commissioner was the Barminco Contract which contained a materially identical quotational period optionality clause for a similar sized mine to the C.S.A. mine. On this occasion, it was discernible from the contract terms, that depending on how the buyer was to exercise its options to determine the quotational period, a discount to the applicable T.C.R.C.s would be conferred. This was precisely the type of quid pro quo that Mr. Ingelbinck and the Commissioner contended should have been secured by C.M.P.L. But Mr. Wilson gave unchallenged evidence that the discount was, from the perspective of value, negligible or immaterial. He said:

The Barminco agreement is an example of an agreement where the agreed TCRC reduces if a particular quotational option is exercised. For the 2005 and 2006 tonnages of approximately 70,000dmt and 60,000dmt respectively, there is a $7 per tonne reduction in the annual Japanese Benchmark treatment charge and a 1c/lb reduction in the refining charge. The contract then provides for a further reduction of the treatment charge of $3 per tonne in the event that the option provided for in clause 11(b) is exercised. This $3 per tonne of concentrate reduction was equivalent to 0.5c/lb of copper, which represented 0.16% of the 2006 copper price, which averaged US$6,722.14 per tonne. This was equivalent to US$10.84 per tonne of copper which is negligible and immaterial in a copper concentrate agreement.

219    During Mr. Wilson’s cross-examination, he would not agree with the proposition that the discount was “significant.” The following exchange occurred:

MS STERN:    And you agree that a $10 discount over a TC of $60 is a quite significant discount, isn’t it?

 MR WILSON:    Relative to the TC. It’s only .15 per cent of the copper price, though.

 MS STERN:    I’m sorry?

 MR WILSON:    It’s only .15 per cent of the copper price, though.

It is apparent that the references to 0.15% should have been references to 0.16%.

220    Mr. Wilson’s evidence about this issue was accepted by the learned primary judge. It was not seriously challenged before this Court.

221    Finally, another difficulty the Commissioner faced, as mentioned above, was that the learned primary judge rejected as mistaken a significant part of Mr. Ingelbinck’s evidence. Mr. Ingelbinck had been of the view that, in applying the quotational period optionality clause adopted in 2004, 2005 and 2007 by C.M.P.L. and G.I.A.G., the buyer was not exposed to any downside risk. Yet, during his cross-examination it was found that in 2009, G.I.A.G. had paid $4.8 million more for copper concentrate than it should have. It had picked the wrong quotational period.

222    The Commissioner submitted that the learned primary judge was wrong to reject Mr. Ingelbinck’s opinion. He argued that her Honour had misunderstood Mr. Ingelbinck’s evidence. The Commissioner submitted as follows:

That finding was based on the primary judge’s misapprehension that a trader who failed to select the quotational period that gave the lowest copper price had necessarily made a “mistake” and had “lost”: J[236]. Mr Ingelbinck’s evidence was not that the Enhanced QP Optionality would always enable a trader to choose the lowest price or in fact choose the lowest price. Mr Ingelbinck recognised that there was risk the buyer could choose other than the most favourable QP and was not guaranteed to make a profit: J[236]. Mr Ingelbinck’s evidence was directed towards how traders actually sought to benefit from QP Optionality with back-pricing privileges – they used it in their trading to achieve margin enhancements from profitable QP mismatches: J[230] and [233]. It was in this sense that Mr Ingelbinck expressed the view that for a buyer QP Optionality with back-pricing carried “no downside risk”: J[231].

(Emphasis in original.)

223    With very great respect, we do not agree with the foregoing characterisation of Mr. Ingelbinck’s evidence. The evidence Mr. Ingelbinck gave in cross-examination was as follows:

MR INGELBINCK:    What I have said in the joint report is that the introduction of such type of optionality takes away any control that the - - -

MR DE WIJN:    Well - - -

MR INGELBINCK:     - - - seller had, but – and gives it to the buyer which results in a situation where the buyer can do no wrong. The buyer can – you know, a qualified buyer will elect – sometimes they will make good margins on it; sometimes they won’t. The back-pricing situation, as it exists – and we all agreed on that – can actually result in the seller receiving a benefit. It can also result in the seller receiving a significant detriment.

(Our emphasis.)

224    Mr. Ingelbinck was then pressed about this by the cross-examiner. The following exchange took place:

MR DE WIJN:    What I’m suggesting to you is that the QP back-pricing optionality, the buyer can get it wrong and choose the wrong QP? He is not guaranteed to make a profit?

MR INGELBINCK:    He is not guaranteed to make a profit. He is guaranteed not to lose anything on it.

MR DE WIJN:    Well, with respect, that’s simply not correct because at the time you exercise the option, at least one quotational period is known and one is unknown?

MR INGELBINCK:    Yes. And so – I’m sorry. I’m now dealing with something I have done for 30 plus years. I can tell you, categorically, there is no way a qualified trader can make a mistake on having that optionality. Zero.

(Our emphasis.)

225    After being confronted with the fact that G.I.A.G. did make a mistake of $4.8 million using the quotational period optionality clause the Commissioner impugns, Mr. Ingelbinck sought to excuse his own mistake in the following way:

MR INGELBINCK:    Your Honour, if I can make a comment on that, that is one of an absolute incredible number of permutations you can go through, which is what I think Mr Wilson and Mr Kowal did, and so depending on which ones you pick, and this one happens to be very one-sided one way, I am sure there is examples where you could come up – that’s the issue with – with – with a – an after-the-fact analysis, you can look at to what you would like to show when you – and – and either – either position you want to take, you can find examples that will suit that. So - - -

(Our emphasis.)

226    In our view, the finding of the learned primary judge at [236] that Mr. Ingelbinck had been shown to be incorrect, was amply supported by the foregoing passages extracted from the transcript. The last passage is especially telling. He appeared to be saying that any analysis of the copper concentrate market which used hindsight could be engineered to give the result that you wanted. That is significant, given his own use of actual as opposed to forecast data in his expert reports as noted above.

227    The learned primary judge rejected Mr. Ingelbinck’s (and Mr. Kowal’s) evidence at [360] as follows:

Mr Ingelbinck and Mr Kowal both considered that the quotational period optionality with back pricing clause in the February 2007 Agreement “provided substantial (but not necessarily quantifiable) value to [GIAG] for which there was no quid pro quo”. Mr Ingelbinck’s opinion, however, was shown to be based on the incorrect view that a trader cannot make a mistake on back pricing optionality and Mr Kowal’s opinion was shown to be based on a flawed analysis of the quotational periods in fact selected by GIAG and based on an assumption of quotational periods fixed annually which he agreed he had no proper basis to assume. Moreover, to the extent that each of them attempted to value the quotational period optionality for GIAG in the relevant years by comparing the revenues received by CMPL based on the quotational periods in fact selected by GIAG with the revenues that would have been received by CMPL had a single quotational period or combination of quotational periods been chosen in the relevant years, it is clear that such comparisons can only be done with the benefit of impermissible hindsight: cf Cameco at [810]-[812]. Mr Kowal acknowledged that the exercise he undertook could not have been done at the beginning of 2007. I accept the taxpayer’s submission that the difference between the prices that CMPL did receive and the prices it would have received had fixed quotational periods been used annually can only be known after the fact, and once the London Metal Exchange copper prices for the whole of the relevant years were known. As such, this difference in price is not one that would have been ascertainable by parties in the position of CMPL and GIAG as at early 2007, agreeing to an optionality clause of the kind in the controlled transaction. It follows that the difference between the revenues received by CMPL under the kind of back pricing optionality provision that was in fact in place and the revenues that would have been received by it had some other quotational period provision been in place provides no reliable indicator as to how independent parties – without the benefit of a crystal ball advising them of future London Metal Exchange prices and market movements – would have valued the benefit conferred upon GIAG under the February 2007 Agreement. Accordingly, I place little weight on the opinions expressed by Mr Ingelbinck and Mr Kowal that the quotational period optionality in the February 2007 Agreement had “substantial value” to GIAG or upon their analyses.

228    In the foregoing passage, the learned primary judge placed little weight on Mr. Ingelbinck’s opinion concerning the quotational period optionality clause for two reasons. First, because his opinion was based upon his view that a trader could not make a mistake using such a clause; that belief drove Mr. Ingelbinck’s opinion that such a clause would be very valuable to a trader, and thus warranted the conferral of some benefit back to the miner. Secondly, because he used hindsight to determine the revenue C.M.P.L. had foregone and thus the benefit it had lost in agreeing to such a clause. As we have already pointed out, Mr. Ingelbinck understood the unsatisfactory nature of a hindsight comparison when considering the copper concentrate market.

229    In our view, this is an occasion where we should approach the learned primary judge’s conclusion with that degree of deference which is warranted by reason of the advantage her Honour enjoyed in having the opportunity to consider, and reflect upon, the entirety of the rather large volume of evidence as it was received at trial and to draw conclusions from that evidence, viewed as a whole. We are not persuaded that her Honour erred in reaching her conclusion to place little weight on Mr. Ingelbinck’s opinion. We also respectfully agree with her conclusion that the Barminco Contract’s quid pro quo clause appeared to have been an unusual feature. Both Mr. Ingelbinck and Mr. Wilson agreed that it was the only contract they had seen which had put a specific price on the conferral of back pricing.

230    We also respectfully agree with the conclusion of the learned primary judge that the evidence did not support the proposition that, if C.M.P.L. had acted at arm’s length, it would have been able to secure substantial discounts in the T.C.R.C. formula in exchange for the quotational period optionality clause. The Commissioner could not avoid that conclusion by pointing to the taxpayer’s onus. The onus on a taxpayer is to show that an assessment issued to it is excessive. No part of that onus necessarily requires a taxpayer to lead evidence to negate positive claims put up by the Commissioner in defence of an assessment: Allied Pastoral Holdings Pty Ltd v. Federal Commissioner of Taxation [1983] 1 N.S.W.L.R. 1 at 10-11. Here, it was sufficient for the taxpayer to discharge its onus of proof by relying on the expert opinion of Mr. Wilson and by demonstrating that Mr. Ingelbinck’s opinion was not to be preferred.

231    For the reasons already given, our conclusion applies equally both to Div. 13 and to Subdiv. 815-A.

(iii)    The Freight Allowance

232    This issue was confined to the 2009 year only.

233    In argument, the Commissioner relied upon the following table, which showed the number of shipments of copper concentrate from the C.S.A. mine and the port of destination for the 2007 to 2009 years.

234    This table showed that as at the start of the 2009 year, out of a total of 38 shipments made in 2007 and 2008, only three had been to India. As it happened, in the 2009 year only one shipment to India took place. Yet C.M.P.L. and G.I.A.G. chose to price the freight allowance based upon the price of shipments to India, which at US$60 per outturn w.m.t. was easily the most expensive way of calculating that allowance.

235    The taxpayer led no lay evidence as to why India was chosen. It could, for example, have led evidence of an expectation which existed in early 2009 of a dramatic change in the destination of shipping for that year, with Indian ports predominating for the first time. It did not do this. It could, for example, have led evidence that at the start of each year C.M.P.L. had no knowledge of where its copper concentrate might be shipped in the months ahead. It did not do this. It could have led expert evidence to support the proposition that the price of US$60 per outturn w.m.t. was a price that independent parties in the position of C.M.P.L. and G.I.A.G. might reasonably be expected to have agreed in early 2009 to be an appropriate freight allowance. It did not do this. Mr. Wilson accepted that he lacked the expertise to opine upon the 2009 freight allowance.

236    In these circumstances, the Commissioner’s submission that the taxpayer had failed to discharge its onus of proof on this issue should be accepted.

237    That conclusion is not defeated by a contention that the actual clause in the C.M.P.L.-G.I.A.G. agreement which addressed the freight allowance was, by its terms, reasonable in nature and one which independent parties might reasonably be expected to have adopted. That contention does not explain why the shipping costs to India were adopted as the freight allowance in 2009. Nor is our conclusion defeated because, as the taxpayer alleges, it impermissibly relies on hindsight. It does not do so. It relies on the historical fact that very few shipments to India took place in 2007 and 2008, relative to the total number of shipments. The taxpayer could have led evidence to explain why this did not matter in 2009, when C.M.P.L. and G.I.A.G. were agreeing the freight allowance for that year. It did not do so. Nor is our conclusion defeated because freight allowances are not necessarily negotiated on the basis of anticipated shipping costs. Absent lay evidence about why India was chosen in 2009, this generalised observation is of no assistance. Finally, we reject the taxpayer’s suggestion that the history of shipping destinations over 2007 and 2008 as set out in the table above was irrelevant. In our view, it was objective evidence that was capable of affecting the freight allowance for 2009 which was properly to be attributed to the hypothetical buyer and seller.

238    Finally, for the reasons already given, the foregoing conclusion applies in relation to both Div. 13 and Subdiv. 815-A.

Conclusion

239    The taxpayer has succeeded on all issues, save for the freight allowance issue in respect of the 2009 year. It follows that the appeal should be allowed in part.

Associate:

Dated:    6 November 2020

REASONS FOR JUDGMENT

THAWLEY J:

240    I have had the advantage of reading the reasons of Middleton and Steward JJ. I agree that the appeal should be allowed and with the orders proposed by their Honours. In relation to the freight issue, which was confined to the 2009 year, I agree with the reasons of Middleton and Steward JJ and have nothing to add. I prefer to express my own reasons in relation to the operation of Div 13 of Part III of the Income Tax Assessment Act 1936 (Cth) (ITAA 1936) and Subdiv 815-A of the Income Tax Assessment Act 1997 (Cth) (ITAA 1997).

INTRODUCTION

241    On 2 February 2007, Cobar Management Pty Ltd (CMPL) and Glencore International AG (GIAG) altered the terms of their agreement for the supply of copper concentrate in two ways critical to the arguments put on appeal:

(1)    First, the parties would no longer rely on benchmark and spot market charges for treatment and refining charges (TCRCs). TCRCs, when calculated by reference to benchmark and spot market prices, continuously moved according to the market. In place of this “market-related” basis for calculating the amount GIAG would pay, the relevant TCRCs were fixed at 23% of the relevant copper price for three years. This was referred to as “price sharing”. These alterations were effected by amendments to cl 8 of the agreement.

(2)    Secondly, the parties amended cl 9 of the relevant agreement to give GIAG greater choice in selecting the quotational period. Clause 9 gave GIAG what was referred to as “quotational period optionality with back pricing”.

242    These changes were significant, involving the introduction of a new methodology for pricing. In particular, the alterations effected significant changes to the respective risks of the parties by introducing a new way of managing or mitigating risk. The reasonable expectation was that CMPL would receive less for its copper concentrate over the forthcoming three years as a result of the changes, but that risks would also be reduced.

243    The primary judge held that neither Div 13 nor Subdiv 815-A authorised the Commissioner to ascertain the consideration that might reasonably be expected to have been received by reference to a transaction or pricing formula that did not include “price sharing” or “quotational period optionality with back pricing”. The primary judge’s reason for that conclusion was that Div 13 and Subdiv 815-A both required the determination of the arm’s length consideration “for the actual transaction”: Glencore Investment Pty Ltd v Commissioner of Taxation (2019) 272 FCR 30 at [324] (hereafter referred to as “J”).

244    The primary judge understood Allsop CJ’s reasons in Chevron Australia Holdings Pty Ltd v Federal Commissioner of Taxation (2017) 251 FCR 40 to confirm, at least in respect of Subdiv 815-A, that the arm’s length price must be “based on the actual transaction as structured by the parties, save in the case of the two exceptional circumstances identified in the 1995 [OECD Transfer Pricing] Guidelines”: J[40]. Her Honour considered the same conclusion obtained under Div 13 such that, under both Div 13 and Subdiv 815-A, the arm’s length consideration had to be determined on the basis of the terms of the actual transaction unless one of the two exceptional circumstances identified in the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations published by the OECD in 1995 (1995 Guidelines) applied: J[312] to [319], in particular at [317].

245    For the reasons given below, I do not consider that Div 13 and Subdiv 815-A only permit departure from the terms of the actual transaction in the limited circumstances identified by the primary judge, namely where one of the two exceptions in the 1995 Guidelines applies.

246    The primary judge also addressed the matter on the basis that she was wrong in her conclusion that it was impermissible to depart from the terms of the actual transaction and that it was necessary for the taxpayer to establish that arm’s length parties might have entered into a transaction in the form of the transaction in fact entered into. Her Honour concluded that:

(1)    first, the form or structure of the relevant international agreement was one which might reasonably be expected between independent parties in the position of CMPL and GIAG dealing at arm’s length: J[320] to [322], [342], [344], [385] to [398]; and

(2)    secondly, the consideration for the supply of copper concentrate under the international agreement so structured was one which might reasonably be expected between independent parties dealing with each other at arm’s length: J[346] to [370], [383], [398].

247    For reasons equivalent to those given by Middleton and Steward JJ the primary judge’s conclusions in this respect should not be disturbed on appeal. The Commissioner has also not shown that, on the correct operation of Div 13 and Subdiv 815-A applied to these findings, the taxpayer failed to discharge its onus of establishing that the assessments were excessive because: (a) the consideration paid by GIAG was an arm’s length consideration for the supply of property under the international agreement (Div 13); and (b) there were not profits which, but for the conditions mentioned in Art 9 of the Agreement between Australia and Switzerland for the Avoidance of Double Taxation with Respect to Taxes on Income, and Protocol [1981] ATS 5 (Swiss Treaty), might have been expected to accrue to the taxpayer, but which, by reason of those conditions, did not so accrue (Subdiv 815-A).

248    I address Div 13 and Subdiv 815-A separately in light of the significant differences in the statutory language and structure.

DIVISION 13

249    Section 136AD(1) provided:

136AD Arm’s length consideration deemed to be received or given

(1)    Where:

   (a)    a taxpayer has supplied property under an international agreement;

(b)    the Commissioner, having regard to any connection between any 2 or more of the parties to the agreement or to any other relevant circumstances, is satisfied that the parties to the agreement, or any 2 or more of those parties, were not dealing at arm’s length with each other in relation to the supply;

(c)    consideration was received or receivable by the taxpayer in respect of the supply but the amount of that consideration was less than the arm’s length consideration in respect of the supply; and

(d)    the Commissioner determines that this subsection should apply in relation to the taxpayer in relation to the supply;

then, for all purposes of the application of this Act in relation to the taxpayer, consideration equal to the arm’s length consideration in respect of the supply shall be deemed to be the consideration received or receivable by the taxpayer in respect of the supply.

250    Central to the operation of this provision is the identification of the three matters in para (a) of s 136AD(1). First, there must be an “agreement” as defined by s 136AA(1) which is an “international agreement” as defined by s 136AC. Section 136AA(1) defined “agreement” (unless the contrary intention appears) in the following way:

(1)    In this Division, unless the contrary intention appears:

…

agreement means any agreement, arrangement, transaction, understanding or scheme, whether formal or informal, whether express or implied and whether or not enforceable, or intended to be enforceable, by legal proceedings.

251    Section 136AC defined “international agreement”:

136AC     International agreements

For the purposes of this Division, an agreement is an international agreement if:

(a)    a non-resident supplied or acquired property under the agreement otherwise than in connection with a business carried on in Australia by the non-resident at or through a permanent establishment of the non-resident in Australia; or

(b)     a resident carrying on a business outside Australia supplied or acquired property under the agreement, being property supplied or acquired in connection with that business; or

(c)     a taxpayer:

(i)     supplied or acquired property under the agreement in connection with a business; and

(ii)     carries on that business in an area covered by an international tax sharing treaty.

252    It was common ground that the relevant “international agreement” was the agreement between CMPL and GIAG as amended on 2 February 2007.

253    The second and third matters critical to the operation of s 136AD(1) are the “supply” of “property” under the international agreement. Relevantly to these two matters, s 136AA included:

(1)     In this Division, unless the contrary intention appears:

…

property includes:

   (a)     a chose in action;

(b)     any estate, interest, right or power, whether at law or in equity, in or over property;

  (c)     any right to receive income; and

  (d)     services.

…

supply includes:

(a)     supply by way of sale, exchange, lease, hire or hire-purchase; and

(b)     provide, grant or confer.

…

(3)     In this Division, unless the contrary intention appears:

(a)     a reference to the supply or acquisition of property includes a reference to agreeing to supply or acquire property …

…

(e)    a reference to the supply or acquisition of property under an agreement includes a reference to the supply or acquisition of property in connection with an agreement.

254    There was no dispute that there was a “supply” of “property” under the international agreement. CMPL supplied copper concentrate to GIAG under the agreement between CMPL and GIAG as amended on 2 February 2007. It follows that it was common ground that para (a) of s 136AD(1) was satisfied.

255    There was also no dispute that:

(1)    para (b) of s 136AD(1) was satisfied: the Commissioner was satisfied, having regard to the lack of independence between them, that CMPL and GIAG were not dealing with each other at arm’s length in relation to the supply; and

(2)    para (d) of s 136AD(1) was satisfied: the Commissioner lawfully made a determination that s 136AD(1) should apply. (The Commissioner also made determinations under s 136AD(4): J[11].)

256    The principal dispute was whether para (c) of s 136AD(1) was satisfied: whether the consideration received by CMPL in respect of the supply of copper concentrate “was less than the arm’s length consideration in respect of the supply”.

257    The definition of “arm’s length consideration” was contained in s 136AA(3), relevantly:

(3)    In this Division, unless the contrary intention appears:

…

(b)    a reference to consideration includes a reference to property supplied or acquired as consideration and a reference to the amount of any such consideration is a reference to the value of the property;

(c)    a reference to the arm’s length consideration in respect of the supply of property is a reference to the consideration that might reasonably be expected to have been received or receivable as consideration in respect of the supply if the property had been supplied under an agreement between independent parties dealing at arm’s length with each other in relation to the supply …

258    The evident object of s 136AD(1)(c), read with the definition of “arm’s length consideration” in s 136AA(3)(c), is to require that the consideration for a supply of property under the “international agreement” which enlivens the potential operation of Div 13 is determined objectively by reference to “an agreement” between independent parties dealing with each other at arm’s length, namely “an agreement” which is not the product of a non-arm’s length dealing between related parties. The provisions require an answer to a hypothetical question: what “consideration … might reasonably be expected to have been received or receivable as consideration in respect of the supply if the property had been supplied under an agreement between independent parties dealing at arm’s length with each other in relation to the supply”?

259    If independent parties, in the position of the parties to the international agreement, would not have agreed to supply the property with the terms affecting price on which that property was in fact supplied, then a reasonable expectation that the supplier would have received the consideration in fact received might not be established, at least on the terms of the international agreement which enlivened the potential operation of Div 13. The language of Div 13 does not indicate that it was intended to permit a taxpayer to dictate the consideration for a supply of property by implementing a transaction which included terms affecting price which might not reasonably be expected in an agreement between independent parties dealing with each other at arm’s length. Whilst related parties might be expected to enter into agreements which might not be found between independent parties dealing at arm’s length, Div 13 does not furnish parties carte blanche to structure transactions in a way which dictates uncommercial or non-arm’s length prices. As Allsop CJ stated in Chevron at [55] of the approach of interpreting Div 13 to require determining the consideration strictly in accordance with the form of the international agreement as in fact entered into:

That approach, however, almost dooms to failure the application of Div 13 if its task is to substitute commercial reality based on independence, for intra-group reality based on group control. All one would have to do would be to constrain internally the transaction to give the highest price and include or omit terms of the agreement that would never be included or omitted in an arm’s length transaction and which are not driven or dictated by commercial or operational imperatives, as the foundation for assessing an hypothesised arm’s length consideration. Such unrealistic inflexibility would undermine the sensible operation of the Division by a rigid construction of the hypothesis in a shape and form controlled by the taxpayer. The difficulty is not alleviated by giving s 136AD(4) a role to play in such circumstances.

260    If the hypothetical agreement replicates all the terms of the “international agreement”, including terms which materially affect price but which might not reasonably be expected to have been agreed between independent parties dealing at arm’s length, it would be difficult to conclude that the resulting consideration might reasonably be expected to have been received for the supply of the property under “an agreement between independent parties dealing at arm’s length with each other in relation to the supply”. That is not to confuse the reasonable expectation about consideration with what might be expected to be the terms of the hypothetical agreement. It is to recognise that a reasonable expectation about consideration might not be able to be formed on the basis of an agreement which could not reasonably be expected to exist between independent parties dealing at arm’s length. Thus s 136AD(1)(c), read with the definition of “arm’s length consideration” contained in s 136AA(3)(c), permits, and even requires in certain circumstances, the consideration to be determined by reference to “an agreement” with terms which depart from those of the “international agreement”.

261    However, nor does Div 13 enable the Commissioner to determine reliably the arm’s length consideration by arbitrarily re-writing the relevant international agreement or by ignoring terms in the international agreement which might reasonably be expected to have been agreed between independent parties dealing at arm’s length in relation to the supply. It is the “international agreement” which enlivens the potential operation of Div 13. It is the supply of property under that particular international agreement which is the underlying subject matter of the relevant pricing exercise. Division 13 requires that a hypothetical agreement be posited in order to determine whether the consideration for the supply of property under the “international agreement” was one which “might reasonably be expected to have been received or receivable as consideration in respect of the supply if the property had been supplied under an agreement between independent parties dealing at arm’s length with each other in relation to the supply” (emphasis added).

262    The more the hypothetical agreement required to be posited by s 136AA(3)(c) (“an agreement”) departs from the terms of the international agreement under which the supply of property occurred, the less the hypothetical agreement is apt to reveal “the consideration that might reasonably be expected to have been received or receivable as consideration in respect of the supply if the property had been supplied under an agreement between independent parties dealing at arm’s length with each other in relation to the supply”. Whilst the Commissioner can, as a matter of law, “substitute” terms or undertake the statutory task by reference to “an agreement” which is different from the “international agreement”, the greater the substitution of terms or the more different the hypothetical agreement is from the international agreement, the less likely it is that the hypothetical agreement is probative of the arm’s length consideration for the supply of the property under the international agreement which engaged Div 13. It is one thing to determine the arm’s length consideration by reference to “an agreement” which substitutes terms which would not be found in an international agreement between independent parties dealing with each other at arm’s length (or to add terms – such as security and covenants – which would have existed if the international agreement had been between independent parties dealing at arm’s length as in Chevron); it is another to determine the arm’s length consideration by reference to “an agreement” which ignores terms in the international agreement which are terms to which independent parties dealing at arm’s length would have agreed (this case).

263    A taxpayer might discharge the onus of establishing that the consideration for the supply of property might reasonably have been expected under an agreement between independent parties dealing with each other at arm’s length, by demonstrating that: (a) the terms of the agreement in fact entered into might reasonably have been expected between independent parties in the position of the parties to the international agreement dealing with each other at arm’s length in relation to the supply; and (b) the consideration payable in respect of the supply of the property on those terms might reasonably have been expected. That is not the only way to discharge the onus, but it is an available way.

264    In its application to the present case, if the taxpayer established that the form of the “international agreement” was one which might reasonably have been expected between independent parties dealing at arm’s length, containing both “price sharing” and “quotational period optionality with back pricing”, then it could succeed in discharging its onus if it also established that the consideration for the supply of copper concentrate on those terms was one which might reasonably have been expected between independent parties dealing at arm’s length. On the facts as found by the primary judge, the taxpayer established those matters.

265    If the conclusion had been reached that independent parties dealing at arm’s length might not reasonably have been expected to have “an agreement” (s 136AA(3)(c)) with “price sharing” or “quotational period optionality with back pricing” or those two terms in combination, then the arm’s length consideration for the supply of the copper concentrate would have needed to have been determined by reference to an agreement without those terms being one which might reasonably be expected between independent parties dealing at arm’s length in relation to the supply of the copper concentrate. As a matter of substance, that is what was decided in Chevron: the loan which was the “international agreement” would not have been made between independent parties dealing at arm’s length unless it included security and operational and financial covenants; accordingly, the consideration which might reasonably be expected could not be established by reference to “an agreement” without security and covenants.

266    Middleton and Steward JJ at [154] conclude that it is permissible, in the sense of the Commissioner having legal capacity, to substitute a different formula or methodology to that found in the international agreement being one which he considers will result in the ascertainment of the arm’s length consideration. Their Honours’ conclusion rests on the fact that those terms “define the price” in the international agreement, or are part of the formula or methodology for determining consideration, and that such clauses are to be distinguished from other clauses which affect consideration: at [155]. The former may be “substituted”, the latter may not.

267    I agree that it is permissible, in the sense of the Commissioner having legal capacity, to “substitute” a different formula or methodology to that found in the international agreement being one which he considers will result in the ascertainment of the arm’s length consideration. In my view, that conclusion rests on the fact that those clauses affect consideration, not on a potentially “unsatisfactory” classification of those terms as ones which “define price”. As explained above, Div 13 permits the arm’s length consideration to be determined by reference to terms affecting consideration which are not found in the international agreement or which are different to those terms. As noted at [259] above, it not consistent with the statutory language or object for Div 13 to be construed as allowing for parties to the international agreement to control what would be determined to be the arm’s length consideration by structuring their agreement such that non-arm’s length terms affecting consideration cannot be said to define price or are to be distinguished from clauses which are part of the methodology for determining consideration.

268    I agree with Middleton and Steward JJ that the “price sharing” and “quotational period optionality with back pricing” terms were part of the methodology for determining price, being included in cl 8 and cl 9 of the 2 February 2007 Agreement which directly defined and affected price. Clause 8 defined price and cl 9 was incorporated into cl 8 by reference. I agree that it is legally permissible for the Commissioner to undertake the statutory task under Div 13 by ignoring those terms. It is difficult to see why the Commissioner would adopt such a course if the terms were ones which might reasonably be expected between independent parties dealing at arm’s length. The explanation in the present case lies in the Commissioner disputing that those terms were ones to which independent parties dealing at arm’s length would have agreed. The primary judge accepted, however, that the terms were arm’s length ones, with the result that the Commissioner determined the consideration by reference to “an agreement” which substituted or ignored arm’s length terms affecting price and thus determined the consideration for the supply under “an agreement” which differed substantially from the “international agreement” which engaged Div 13 and which did not supply the answer to the questioned posed by s 136AD(1)(c) because the terms which had been substituted or ignored were ones which might reasonably be expected between independent parties dealing at arm’s length. The resulting assessments were excessive.

269    Most terms of an agreement affect price and many may properly be characterised as part of the consideration, whether or not the terms are found in a contractual clause specifically directed at, or defining, price – cf: Chevron at [46]. In the context of a loan for example, the provision of security directly affects price even though it is not expressed as part of the methodology for determining price. The provision of security also affects risk and cost to the lender. A customer who provides a first mortgage for a home loan might borrow from a bank at an interest rate of 5%. The same bank might only be prepared to lend to the same customer at 10% if security were more limited. If the bank would not have lent to the customer at all unless security were provided, what might reasonably be expected to be the consideration would not generally be able to be determined by reference to “an agreement” without security. At the risk of oversimplification, that is what Chevron decided: a reasonable expectation that the consideration is an arm’s length one is not proved by establishing the consideration which would be received under an agreement to which independent parties dealing at arm’s length would never have agreed. Chevron is not authority for the proposition that the arm’s length consideration for the purposes of Div 13 may reliably be determined by ignoring or substituting arm’s length terms. An assessment would likely be excessive if it were issued on the basis of “an agreement” which departs from terms in the “international agreement” which both affect consideration and might reasonably be expected between independent parties dealing with each other at arm’s length in relation to the supply. If the arm’s length consideration is determined by reference to such “an agreement”, then the issue raised by s 136AD(1)(c) would not have been reliably answered because the ignored terms of the “international agreement” which engaged the operation of Div 13 might reasonably have been expected in “an agreement” between independent parties dealing at arm’s length.

270    Each case turns on its own facts. However, generally:

(1)    If a taxpayer establishes that the terms in the “international agreement” which affect consideration were terms which might reasonably have been expected in “an agreement” between independent parties dealing at arm’s length, then the consideration which might reasonably be expected for the supply of property should be determined by reference to “an agreement” with those terms. That is so whether or not the arm’s length terms are characterised as defining price.

(2)    On the other hand, if terms in the “international agreement” affecting price were not ones which might reasonably have been expected between independent parties dealing at arm’s length, then the consideration for the supply might be established by reference to “an agreement” of sufficient similarity or probative value to that in fact entered into as to permit the statutory question to be answered, being an agreement which might reasonably be expected between independent parties in the position of the relevant parties dealing with each other at arm’s length in relation to the supply.

271    In the present case, Glencore Investment Pty Ltd (GIPL) discharged its onus of establishing that the consideration received by CMPL was one which might reasonably be expected to have been received as consideration on the assumption that the supply of copper concentrate under the international agreement had been made under an agreement between independent parties dealing at arm’s length with each other in relation to the supply. GIPL discharged its onus by showing that:

(1)    first, the terms of the “international agreement” were ones which might reasonably be expected between independent parties, in the position of CMPL and GIAG, dealing with each other at arm’s length; and

(2)    secondly, the consideration for the supply of copper concentrate under “an agreement” with those terms was within an arm’s length range.

272    The Commissioner approached the principal issue in this case on the basis that independent parties, in the position of CMPL and GIAG, dealing with each other at arm’s length, would not reasonably be expected to have agreed to a consideration for the supply of copper concentrate on the basis in fact agreed on 2 February 2007, because independent parties dealing at arm’s length would not have altered their existing agreement. Whilst facts relevant to this argument might also be relevant to the issues under Div 13, this did not correctly frame the issue under Div 13. The argument hints at a case under Pt IVA of the ITAA 1936, but no such case was made. Whilst Pt IVA and Div 13 were introduced at a similar time, Div 13 does not require an investigation or consideration of purpose or motive: W R Carpenter Holdings Pty Ltd v Commissioner of Taxation (2008) 237 CLR 198 at [38]. The real question under Div 13 is whether the consideration received by CMPL under the “international agreement” as it stood on 2 February 2007 was an arm’s length consideration within the meaning of Div 13. On the basis that, as the evidence established, the international agreement was in terms which might be expected between independent parties in the position of CMPL and GIAG, dealing with each other at arm’s length, the real issue thrown up by the facts was the arm’s length consideration for the supply of property under that agreement. This devolved into the question of whether TCRCs fixed at 23% of the copper price was arm’s length consideration. By reason of the amendments to the agreement, CMPL reduced its risk but would likely receive less consideration. The primary judge concluded, after a detailed examination of all of the evidence, that this reflected what might reasonably have been expected between independent parties in the position of CMPL and GIAG dealing with each other at arm’s length. Her Honour also concluded that the consideration was arm’s length. The reasons of Middleton and Steward JJ show why her Honour did not err in so concluding.

THE 1995 GUIDELINES

273    As to the 1995 Guidelines, I make the following observations. First, the expression “arm’s length consideration” in Div 13 is resolved by reference to its terms read in context. No serious attempt was made to show how the 1995 Guidelines could be regarded as throwing significant light on the correct interpretation of Div 13. The 1995 Guidelines were a revision of the OECD Report, Transfer Pricing and Multinational Enterprises, adopted by the Committee on Fiscal Affairs (the main tax policy body of the OECD) in January 1979 and approved for publication by the OECD Council on 16 May 1979 (1979 Report). At [15], the 1979 Report stated:

CHAPTER 1

SUMMARY OF PROBLEMS

THE ARM’S LENGTH PRINCIPLE

…

Minor adjustments and substitution of methods

15. The starting point for scrutinising transfer prices would frequently be the appearance of a discrepancy between the profits returned by an associated enterprise and those which might be expected to be made by comparable enterprises in the uncontrolled situation. Since the assessment of an arm’s length price depends very often on careful judgement and the resolution of many, perhaps conflicting, considerations by negotiation between the tax authorities and the enterprise concerned, it follows that if the prices actually paid can be substantiated by acceptable evidence as being arm’s length prices there would be no justification for seeking to make merely minor or marginal adjustments to them for tax purposes. Similar[l]y a tax authority should hesitate to disturb without good reason a pricing arrangement reasonably and consistently operated between associated enterprises if it is also reasonably and consistently operated in comparable dealings with independent parties. Moreover, as a general principle, tax authorities should base their search for an arm’s length price on actual transactions and should not substitute hypothetical transactions for them, thus seeming to substitute their own commercial judgement for that of the enterprise at the time when the transactions were concluded (though there may be some circumstances where the form of transaction has effectively to be ignored - see paragraphs 23 and 24).

274    At [23] and [24], the Report stated:

Recognition of actual payments and transactions

23. In general, the approach which is adopted in this report to the adjustment of transfer prices for tax purposes is to recognise the actual transactions as the starting point for the tax assessment and not, in other than exceptional cases, to disregard them or substitute other transactions for them. The aim in short is, for tax purposes, to adjust the price for the actual transaction to an arm’s length price. Similarly, it is considered that transactions between associated enterprises should not be treated differently for tax purposes from similar transactions between independent parties simply because the parties to the transaction are related. The report does, however, recognise that it may be important in considering, for example, what is ostensibly interest on a loan to decide whether it is an interest payment or, in reality, a dividend or other distribution of profit.

Intra-group contracts and arrangements

24. Associated enterprises are, however, able to make a much greater variety of contracts and arrangements than can unrelated enterprises because the normal conflict of interest which would exist between independent parties is often absent. Associated enterprises may and frequently do conclude arrangements of a specific nature that are not or are very rarely encountered between unrelated parties (cost contribution arrangements for research and development expenditure as discussed in Chapter III of the report would be one example). This may be done for various economic, legal or fiscal reasons dependent on the circumstances in a particular case. Moreover, contracts within an MNE could be quite easily altered, suspended, extended or terminated according to the overall strategies of the MNE as a whole and such alterations may even be made retroactively. In such instances tax authorities would have to determine what is the underlying reality behind an arrangement in considering what the appropriate arm’s length price would be.

275    Division 13 was introduced in 1982.

276    The 1995 Guidelines were approved for publication some thirteen years later, in 1995. The 1995 Guidelines set out a position reached after considering various OECD reports and competing and perhaps conflicting positions of member states and delegates. Included in the reports considered were OECD reports issued to promote acceptance of common interpretations of the various Articles in bilateral treaties, including Art 9 of the Swiss Treaty, recognising that such interpretations differed. As indicated by its full title, the 1995 Guidelines were “intended to help tax administrations (of both OECD Member countries and non-Member countries) and MNEs by indicating ways to find mutually satisfactory solutions to transfer pricing cases”: Preface at [15]. They were also intended “to govern the resolution of transfer pricing cases in mutual agreement proceedings between OECD Member countries and, where appropriate, arbitration proceedings”: Preface at [17]. The 1995 Guidelines were issued in a looseleaf format because the 1995 Guidelines focussed on the main issues of principle, work was continuing and regular reviews of the experiences of OECD Member and selected non-Member countries was anticipated: Preface at [19]. The 1995 Guidelines were not intended to guide legislators in enacting domestic laws. The 1995 Guidelines were supplemented on numerous occasions and, in 2010, a new revision of the 1979 Report was approved, namely the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (2010 Guidelines). The 2010 Guidelines continue to be supplemented, reviewed and revised.

277    The text of [C.1.36] to [C.1.38], appearing in the 1995 Guidelines, are not a sure way to understanding the correct operation of Div 13 according to its terms.

278    Secondly, contrary to the primary judge, I do not read the reasons of Allsop CJ in Chevron as stating or implying that the identification of the arm’s length consideration must be “based on the actual transaction as structured by the parties, save in the case of the two exceptional circumstances identified in the 1995 Guidelines”: J[40]; J[312] to [319].

279    Allsop CJ’s decision so far as it concerned Div 13 was based on the text of Div 13 and, more specifically, what might reasonably be expected if the consideration had been received under an agreement between independent parties dealing with each other at arm’s length in relation to the supply. His Honour concluded, in substance, that it would not reasonably be expected that the loan would have been made without security or a parent guarantee and, accordingly, the arm’s length consideration could not be reliably established by reference to “an agreement” which did not contain those features.

280    The Chief Justice’s reference to the two exceptions in the 1995 Guidelines was in the context of considering the application of Subdiv 815-A. However, his Honour was not applying the exceptions referred to in the 1995 Guidelines. There was no suggestion that the first exception applied. No conclusion was expressed that the second exception applied and nor did his Honour make findings which would have been required to engage the second exception, namely that the agreement reached: (a) differed from one “which would have been adopted by independent enterprises behaving in a commercially rational manner”; and (b) “practically impede[d] the tax administration from determining an appropriate transfer price”. Rather, the Chief Justice’s reference to the two exceptions was merely to support the conclusion otherwise reached by his Honour from the text of Subdiv 815-A that the inquiry under s 815-15(1)(c) was one which required a “comparative analysis that gives weight, but not irredeemable inflexibility, to the form of the transaction actually entered between the associated enterprises”: Chevron at [90].

281    Thirdly, notwithstanding that I differ from the primary judge in my understanding of Allsop CJ’s reasoning in Chevron, I would observe that, if the relevant agreement is one which “differ[ed] from [that] which would have been adopted by independent enterprises behaving in a commercially rational manner” ([C.1.37] of the 1995 Guidelines), then:

(1)    for the purposes of Div 13: the consideration which might reasonably be expected to have been received for the supply of property under the “international agreement” is unlikely to be reliably established by reference to “an agreement” in the same commercially irrational form;

(2)    for the purposes of Subdiv 815-A: the commercially irrational agreement might constitute a “condition” operating, or tend to indicate that “conditions operate”, “between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another” such that the question arises under s 815-15(1)(c) whether “an amount of profits which, but for the conditions … might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued”.

DIVISION 815-A

282    The object of Subdiv 815-A so far as presently relevant is set out in s 815-5(a) of the ITAA 1997:

815-5 Object

The object of this Subdivision is to ensure the following amounts are appropriately brought to tax in Australia, consistent with the arm’s length principle:

(a)    profits which would have accrued to an Australian entity if it had been dealing at *arm’s length, but, by reason of non-arm’s length conditions operating between the entity and its foreign associated entities, have not so accrued; ...

283    Section 815-10 provides for two matters which must be established in order for the Commissioner to make a determination under s 815-30(1):

815-10 Transfer pricing benefit may be negated

(1)    The Commissioner may make a determination mentioned in subsection 815-30(1), in writing, for the purpose of negating a *transfer pricing benefit an entity gets.

Treaty requirement

(2)    However, this section only applies to an entity if:

(a)    the entity gets the *transfer pricing benefit under subsection 815-15(1) at a time when an *international tax agreement containing an *associated enterprises article applies to the entity; or

(b)    the entity gets the transfer pricing benefit under subsection 815-15(2) at a time when an international tax agreement containing a *business profits article applies to the entity.

284    Relevantly to s 815-10(1), s 815-15 sets out when a taxpayer “gets a transfer pricing benefit”:

815-15 When an entity gets a transfer pricing benefit

Transfer pricing benefit—associated enterprises

(1)    An entity gets a transfer pricing benefit if:

  (a)    the entity is an Australian resident; and

(b)    the requirements in the *associated enterprises article for the application of that article to the entity are met; and

(c)    an amount of profits which, but for the conditions mentioned in the article, might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued; and

  (d)    had that amount of profits so accrued to the entity:

(i)    the amount of the taxable income of the entity for an income year would be greater than its actual amount; or

(ii)    the amount of a tax loss of the entity for an income year would be less than its actual amount; or

(iii)    the amount of a *net capital loss of the entity for an income year would be less than its actual amount.

The amount of the transfer pricing benefit is the difference between the amounts mentioned in subparagraph (d)(i), (ii) or (iii) (as the case requires).

285    Relevantly to s 815-10(2), the “treaty requirement” in the present case was Art 9 of the Swiss Treaty. Article 9 provided:

Where –

(a)    an enterprise of one of the Contracting States participates directly or indirectly in the management, control or capital of an enterprise of the other Contracting State; or

(b)    the same persons participate directly or indirectly in the management, control or capital of an enterprise of one of the Contracting States and an enterprise of the other Contracting State,

and in either case conditions operate between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another, then any profits which, but for those conditions, might have been expected to accrue to one of the enterprises, but, by reason of those conditions, have not so accrued, may be included in the profits of that enterprise and taxed accordingly.

286    Section 815-20 provides:

815-20 Cross-border transfer pricing guidance

(1)    For the purpose of determining the effect this Subdivision has in relation to an entity:

(a)    work out whether an entity gets a *transfer pricing benefit consistently with the documents covered by this section, to the extent the documents are relevant; and

(b)    interpret a provision of an *international tax agreement consistently with those documents, to the extent they are relevant.

(2)    The documents covered by this section are as follows:

(a)    the Model Tax Convention on Income and on Capital, and its Commentaries, as adopted by the Council of the Organisation for Economic Cooperation and Development and last amended on 22 July 2010;

(b)    the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, as approved by that Council and last amended on 22 July 2010;

(c)    a document, or part of a document, prescribed by the regulations for the purposes of this paragraph.

(3)    However, a document, or a part of a document, mentioned in paragraph (2)(a) or (b) is not covered by this section if the regulations so prescribe.

(4)    Regulations made for the purposes of paragraph (2)(c) or subsection (3) may prescribe different documents or parts of documents for different circumstances.

287    It was common ground that the 1995 Guidelines were “covered by” s 815-20, having regard to relevant transitional provisions.

288    The real issue in dispute between the parties was the application of paras (b) and (c) of s 815-15(1), namely whether “an amount of profits which, but for the conditions mentioned in [Art 9], might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued”.

289    The first step, required by para (b), is to identify the “conditions” in Art 9, namely what “conditions operate between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another”.

290    The second step, required by para (c), is to identify whether “but for the conditions” in Art 9 an amount of profits might have been expected to, but did not, accrue to the entity.

291    The focus of the inquiry is on whether conditions operated between enterprises in their commercial or financial relations and, if so, what effect those conditions had. As mentioned above, the primary judge approached the task under Subdiv 815-A on the basis that it was not permissible to determine consideration otherwise than by reference to the agreement in fact entered into unless one of the exceptions in the 1995 Guidelines applied. In my view, this fails to give effect to s 815-15(1)(b) and (c) which focus on whether conditions operating between the entities affected the profits which accrued.

292    Unlike the position with respect to Div 13, the 1995 Guidelines are expressly made relevant by the terms of Subdiv 815-A. If one of the exceptions in [C.1.37] of the 1995 Guidelines applied, it would be consistent with the 1995 Guidelines to determine that the effect of Subdiv 815-A in relation to the entity was that it got a transfer pricing benefit under s 815-15 which could be negated under s 815-10. However, for the reasons which follow, I would not take the step taken by the primary judge and conclude that, in all cases where the two exceptions did not apply strictly according to the terms of those exceptions as identified in the 1995 Guidelines, the relevant transaction must always be taken exactly as found.

293    Firstly, Subdiv 815-A must be applied according to its terms and one could not give effect to the 1995 Guidelines if to do so would be inconsistent with the terms of the subdivision or would fail to allow its operation according to its terms. Section 815-15(1)(c) is engaged where “an amount of profits which, but for the conditions mentioned in the article, might have been expected to accrue to the entity, has, by reason of those conditions, not so accrued”. If the conclusion were reached that the provision applied, that conclusion should be given effect even if the facts did not strictly fit within one of the two exceptions contained in [C.1.37] of the 1995 Guidelines.

294    Secondly, in any event, I do not read the 1995 Guidelines as specifying that the two situations identified are the only situations which might be regarded as “exceptional”. As noted above, I do not read the decision of the Full Court in Chevron, or the reasons of Allsop CJ specifically, as authority for such a proposition.

295    The primary judge concluded that the terms of the international agreement (and its form or structure) were ones which might reasonably be expected between independent parties, in the position of CMPL and GIAG, dealing with each other at arm’s length and that the consideration for the supply of copper concentrate under an agreement so structured was arm’s length. The application of Subdiv 815-A to those facts required the conclusion that no transfer pricing benefit was obtained. The terms under which the copper concentrate was supplied, being conditions operating between CMPL and GIAG in their commercial or financial relations, were shown not to differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another. GIPL discharged its onus of establishing that independent parties in the position of CMPL and GIAG might have been expected to agree to supply the copper concentrate on the terms it was supplied. It was consistent with the 1995 Guidelines to conclude that the effect of Subdiv 815-A was that no transfer pricing was obtained: s 815-20.

296    In the first sentence of [156], Middleton and Steward JJ state that their conclusions with respect to the operation of Div 13 concerning the “substitution” of a different formula or methodology for ascertaining the arm’s length consideration apply equally to Subdiv 815-A. Their Honours’ conclusions in relation to Div 13 were, in summary, that the Commissioner may substitute clauses which define price, but not those which do not. In my view:

(1)    If the terms in the agreement which “define price” are arm’s length terms, it would ordinarily be unlikely that those terms could be said to be “conditions” operating “between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another”. In those circumstances, it is difficult to see how they could rationally be substituted.

(2)    If the agreement contains terms which do not define price, and which would not be found in the agreement if it had been between independent enterprises dealing wholly independently with one another, it is difficult to see why those terms could not be substituted under Subdiv 815-A. Such terms may well amount to “conditions” operating “between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another”. If such terms affect the consideration payable or receivable, or profits, it is difficult to see why they should not be “substituted”.

297    In my view, the language of Subdiv 815-A does not contemplate or require a distinction between terms which define price (which Middleton and Steward JJ consider can be substituted) and those which do not (which their Honours consider may not be substituted). In this respect, I note that their Honours’ observation at [155] that “the term ‘consideration’ [in Div 13] necessarily directs one to identify those clauses in an international agreement which define the price” does not apply to the language of Subdiv 815-A.

298    That is not to say that it is irrelevant to look at the extent to which a particular clause affects consideration or profit. That is plainly and directly relevant. The point of difference is simply that what may or may not be “substituted” for the purpose of applying Subdiv 815-A does not turn on a potentially “unsatisfactory” categorisation of contractual clauses as ones which do or do not define price – see [155] of Middleton and Steward JJ’s reasons for judgment. Subdiv 815-A, in its application to the present case, turns on the question of whether conditions operated “between the two enterprises in their commercial or financial relations which differ from those which might be expected to operate between independent enterprises dealing wholly independently with one another” – see s 815-10(2); Art 9.

299    As to the final sentence of [156], this implicitly acknowledges the possibility that conditions which do not define price may be substituted, a conclusion with which I agree.

CONCLUSION

300    I agree with the orders proposed by Middleton and Steward JJ.

Associate:

Dated:    6 November 2020