ATO loses transfer pricing appeal

12 November 2020

In brief

On 6 November 2020, the Commissioner of Taxation’s appeal in Commissioner of Taxation v Glencore Investment Pty Ltd [2020] FCAFC 187 (Glencore FFC) was unanimously dismissed by the Full Federal Court (except for a minor issue regarding freight). The Court took a practical and sensible approach in examining Australia’s complex transfer pricing laws. The majority explained that care must be taken “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.”

This judgment is significant because it provides further clarity to Australia’s transfer pricing rules and, in particular, further elucidates key aspects of previous Full Federal Court decisions. Subject to the outcome of any application for special leave and appeal, this decision is expected to have a significant impact on a broad range of outstanding transfer pricing disputes. Taxpayers who document evidence as to why the agreed price is commercially prudent will be in a better position. In addition, it continues to be best practice for taxpayers to set their internal prices using a framework that is similar to reasonable third parties comparables.

In detail

Background to recent transfer pricing disputes in Australia

The reason the Glencore Case has been so highly anticipated is explained by the following context:

1. The Australian Taxation Office (ATO) has explained that transfer pricing is a focus given its criticality to the Australian taxation system, with its key areas of focus including related party loans, intangible property, marketing “hubs” and inbound “distributor” supply chains. The ATO and taxpayers have often clashed on the scope of the Commissioner’s power to raise transfer pricing assessments based on his view of “what might reasonably be expected” as an alternative to the tested transaction.

2. The ATO has issued a number of Practical Compliance Guidelines (PCGs) which in broad terms, set out the ATO perspective on safe “green zone” arrangements. However, these risk assessment tools do not necessarily align with international transfer pricing guidelines. This lack of alignment could lead to double taxation but more importantly, additional uncertainty and therefore, risk for taxpayers.

3. The ATO’s focus on transfer pricing follows the Full Federal Court decision in the Chevron case which involved the application of Australia’s transfer pricing rules to a related party loan. A confidential settlement was reached in 2018 before the High Court heard the taxpayer’s special leave application.

Overview of the Glencore cases

The Glencore cases concerned transfer pricing aspects of the acquisition of copper concentrate under an offtake agreement (the contract) by Glencore International AG (buyer) (GIAG) from its subsidiary, Cobar Management Pty Ltd (seller) (CMPL), an Australian company which owned and operated the mine in Australia. The evidence heard by the Court was that in 2007 there was uncertainty in future copper prices and higher costs associated with the mine and in that context the pricing arrangement in the contract was changed.

The cases considered whether the Commissioner was able to take a “flexible” approach to identifying the substitute hypothesis for arm’s length purposes, including inserting additional clauses and changing the structure of the pricing arrangement between the parties. In particular, the Commissioner sought to replace the pricing mechanism, agreed in 2007 to be set at 23% of the copper reference price on the London Metals Exchange, with the historical benchmark previously used by the parties. In effect, the Commissioner’s approach sought to alter the arrangement to what the Commissioner considered independent parties would have agreed rather than price the actual arrangement entered into. The Commissioner issued amended assessments to reflect the mechanism previously used and increased the consideration paid under the contract by AUD241 million over the 2007 and 2009 years.

Glencore - first instance decision (Davies J)

In Glencore Investment Pty Ltd v Commissioner of Taxation of the Commonwealth of Australia [2019] FCA 1432 (Glencore FC), Davies J allowed the taxpayer’s appeal in full (with costs). The 127-page judgment set out a number of key principles regarding the Australian transfer pricing provisions (former Division 13 of the Income Tax Assessment Act 1936 (Cth) and Subdivision 815-A of the Income Tax Assessment Act 1997 (Cth)) and the Chevron decision. Our summary of her Honour’s key findings can be found here.

Glencore - the appeal (Middleton, Steward and Thawley JJ)

The Commissioner filed his notices of appeal out of time but was subsequently granted an extension. On appeal, it was noted that nearly all of the facts found by Davies J at trial were not in dispute. Rather, the real contest on appeal was whether the taxpayer had discharged its burden of proving that the assessments were excessive, in the context of competing expert evidence from both parties on the underlying issues.

Despite the heavy focus in Glencore FC on whether the Commissioner was allowed to reconstruct the taxpayer’s transaction, the issue was not ultimately decisive on appeal. In a joint judgment, Middleton and Steward JJ held that the Commissioner’s case did not involve a reconstruction, but an alteration in the pricing mechanism of the contractual arrangements to supply copper concentrate. The pricing mechanism was determined to fall within the scope of consideration for Division 13 and conditions for Subdivision 815-A, which the Commissioner was able to alter or substitute for the purposes of the respective provisions.

Instead, the taxpayer succeeded in having the appeal dismissed because of its evidence.

The Commissioner’s and the taxpayer’s experts disagreed over the pricing with which independent parties would be more likely to agree and the Court accepted aspects of both experts. It considered the matter to be “a case where reasonable minds have reasonably differed within a range of commercially acceptable arm’s length outcomes” and therefore that the ATO’s expert’s view was “another possible position that arm’s length parties might be reasonably expected to have adopted.” This reconfirms the approach taken in the SNF and Allied Pastoral cases - that is, the taxpayer does not have to demonstrate its pricing is more likely than that put by the ATO or “to lead evidence to negate positive claims put up by the Commissioner”, they just have to show through the evidence that its position is within a range of “commercially acceptable arm’s length outcomes”.

The majority judgment also provided the following important guidance as to the application of Australian transfer pricing provisions:

• Substituting pricing methodology is permitted. The Commissioner was entitled to substitute a different methodology or formula for the ascertainment of the applicable deductions and for determining the quotational period. This was 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 Division 13, and are “conditions” for the purpose of the treaty.
• Substituting terms which do not define price is not permitted. The majority confirmed that there is no power or authority to substitute different terms of a contract where those terms are not seen as defining the consideration received. The same conclusion was said to apply to Subdivision 815-A but the scope of that limitation was left as “a question for another day”. Thawley J found that no such restriction existed, but noted that the more an hypothetical agreement departed from the international agreement, the less likely that it will be probative of the arm’s length consideration.
• Commercial prudence is sufficient. The majority disagreed with the Commissioner’s argument that the taxpayer’s expert’s conclusions about commerciality and prudence fell short of the statutory tests. Conclusions about commerciality and prudence are matters that are determined by examining what parties dealing at arm’s length might be expected to agree to. A taxpayer will succeed if it can demonstrate that its position is a “sufficiently reliable prediction” which can be seen as reasonable, even where the ATO attempts to assert that it is instead unreasonable.
• Taxpayers are not required to maximise profitability at the expense of risk mitigation. The majority held that there is no obligation to “maximise” profitability at the expense of all else. This is because there is likely to be more than one price which is an arm’s length price and a taxpayer is under no obligation to choose a pricing methodology which pursues profitability in Australia at the expense of prudence, particularly when different business people will make different decisions about how to manage risk. The majority also acknowledged that, unsurprisingly, mistakes may be made in arm’s length agreements.
• Forecasts are just that. It is not appropriate to rely on estimates or forecasts as certainties, even if in hindsight they turned out to be correct. Middleton and Steward JJ referred to Commissioner of State Revenue v Placer Dome Inc (2018) 265 CLR 585, where it was noted 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”. Forecasts may still have a role to play in understanding what might be expected to occur but the decision highlights that it will not be determinative.
• Comparables that are not identical still have probative weight. While differences between comparable contracts identified by the Commissioner may diminish the probative value of the contracts said by a taxpayer to be comparable, they do not negate that value entirely. The contracts relied upon by the taxpayer in this case were still valid “reference points” and a sounding board. The contracts were held to be relevant and admissible pursuant to sections 55 and 56 of the Evidence Act 1995 (Cth), notwithstanding the differences identified by the Commissioner.
• The OECD Guidelines do not replace the text of the statutes. The OECD Guidelines are only a guide as to how a revenue authority or a taxpayer might apply the “arm’s length principle”. Consequently, the obligation under Subdivision 815-A to work out whether an entity has obtained a transfer pricing benefit consistently with these Guidelines only applies to the extent that the Guidelines are relevant. Given that the majority observed that the language deployed by the OECD Guidelines is “very highly generalised and is frustratingly opaque”, the Guidelines will not displace the application of the text in Subdivision 815-A to particular facts.
• The hypothetical “must be made to work”. In order to do this, it is necessary to draw upon commercially rational practices adopted by independent parties operating in a particular market for goods and services using a degree of “flexibility and pragmatism”. This did not require the taxpayer to lead evidence of any actual negotiations between the parties because “it would be strange if such negotiations had ever meaningfully taken place”. 
• Prior contracts are not determinative of whether the contract in dispute was arm's length. The Commissioner argued that the taxpayer would not have amended the contract in question because the Commissioner claimed that the taxpayer was "highly likely to be worse off" under the amendment rather than remaining on the existing terms. However, the majority held that the ultimate issue for determination is whether the contract in dispute was priced at arm’s length, not whether an arm’s length party would have agreed to amendments to a prior agreement. 

The ATO will have 28 days from the date the final orders are entered to ask the High Court for special leave to hear any appeal.

While the case involved the application of Australia’s former transfer pricing rules (Division 13 and Subdivision 815-A), the principles established will also be relevant to the current transfer pricing rules (Subdivision 815-B applicable from 29 June 2013) which require the arm’s length conditions to be identified in a way that is consistent with guidelines on arm’s length principles developed by the OECD.

The takeaway

The Glencore FFC decision provides helpful clarity in relation to the operation of Australia’s historic transfer pricing rules. It has affirmed the approach from the Chevron decision that the Courts will take a rigorous approach to evidence in transfer pricing matters.

Perhaps of interest is the reduced focus on the OECD Guidelines. Here there may be a difference between the legacy Division 13 and Subdivision 815-A provisions and Subdivision 815-B.

In the days since the Glencore FFC decision, it is difficult to anticipate all the possible arguments that might be proffered as to why Subdivision 815-B might be different to the legacy Subdivision 815-A and Division 13 provisions. However the observations of the majority should be insightful on interpretive aspects under Subdivision 815-B.

Taxpayers should also challenge themselves on how they prepare transfer pricing documentation for filing positions. Perhaps a reasonableness test that extends beyond a statistical analysis of a range of comparables will now be more appealing to taxpayers as a means of feeling more certain about positions adopted. An element of evidentiary analysis may become best practice for those who want to discourage a “battle of the ranges”, means and medians.

Notwithstanding this, the transfer pricing world has not been turned on its head. Taxpayers will be pleased by the sympathy of the Court for the complexity of documenting transfer pricing positions. Applying concepts of “reasonableness” and “commercial prudence” will give taxpayers some comfort that their analysis does not need to be a perfect prediction of what would occur. The underlying threshold remains “what might reasonably be expected” between independent parties and there may very well be a range of acceptable outcomes.

We recommend taxpayers use these concepts by bolstering their existing transfer pricing positions with evidence of what might reasonably be expected to occur. Those who use this as an opportunity to “down size” their analysis may have missed the context of this “battle”.

Please keep an eye out for upcoming PwC Australia events in which we will unpack this case and its implications further.