Journal of Australian Taxation
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Journal of Australian Taxation |
|
THE TRADITIONAL RATIONALE OF THE ARM’S LENGTH APPROACH TO TRANSFER PRICING – SHOULD THE SEPARATE ACCOUNTING MODEL BE MAINTAINED FOR MODERN MULTINATIONAL ENTITIES?
By Kerrie Sadiq[*]
This article examines the current transfer pricing regime to consider whether it is a sound model to be applied to modern multinational entities. The arm’s length price methodology is examined to enable a discussion of the arguments in favour of such a regime. The article then refutes these arguments concluding that, contrary to the very reason multinational entities exist, applying arm’s length rules involves a legal fiction of imagining transactions between unrelated parties. Multinational entities exist to operate in a way that independent entities would not, which the arm’s length rules fail to take into account. As such, there is clearly an air of artificiality in applying the arm’s length standard. To demonstrate this artificiality with respect to modern multinational entities, multinational banks are used as an example. The article concludes that the separate entity paradigm adopted by the traditional transfer pricing regime is incongruous with the economic theory of modern multinational enterprises.
Profit shifting by multinational entities may occur by means of transfer pricing, with the result that traditional rules fail to yield a fair interjurisdictional allocation of taxing rights. The current solution to the problem of transfer price manipulation is the substitution of an arm’s length price for related party transactions. Despite the artificial nature of the process that determines an arm’s length price, using the arm’s length approach to determine a plausible price for related party transactions has worked in the case of most multinational entities.[1] This article argues, however, that the application of the arm’s length methodologies to modern multinational entities, taking into account their unique features, results in a transfer price that fails to accord with economic reality. The premise of this article is that the arm’s length pricing paradigm is not a theoretically sound model for allocating the profits of modern multinational entities. To demonstrate, this article uses multinational banks as an example of the modern multinational entity.
The first part of this article considers the OECD’s answer to transfer pricing manipulation – the adoption of the arm’s length price methodology. It then examines the rationale for the use of the arm’s length price, along with the arguments in favour of such a regime. This article refutes these justifications on the general basis that the traditional regime fails to recognise true economic activity.
This article also considers the OECD’s discussion drafts which have dealt with the issue of multinational banks and attempt to address the issue of the application of the current regime to multinational banking transactions. Particular emphasis is also placed on the following publications of the OECD: the 1998 discussion draft, The Taxation of Global Trading of Financial Instruments (“1998 Draft”),[2] as updated by the 2003 Discussion Draft on the Attribution of Profits to Permanent Establishments: Part III (Enterprises Carrying on Global Trading of Financial Instruments) (“2003 Pt III Draft”)[3] and the 2001 Discussion Draft on the
Attribution of Profits to Permanent Establishments (“2001 Draft”)[4] as updated by the 2003 Discussion Draft on the Attribution of Profits to Permanent Establishments Part II (Banks) (“2003 Pt IlI Draft”).[5] Also analysed is the OECD’s 1984 document Transfer Pricing and Multinational Enterprises: Three Taxation Issues (“1984 Report”),[6] which examines the taxation of multinational banks as a potentially difficult area in relation to transfer pricing.
In this article it is posited that, contrary to the very reason multinational entities exist,[7] applying arm’s length rules to a multinational entity involves a legal fiction of imagining transactions between unrelated parties. Multinational entities exist to operate in a way that independent entities would not, which the arm’s length rules fail to take into account. As such, there is clearly an air of artificiality in applying the arm’s length standard. It is recognised that this artificiality applies across all multinational entities, but it is argued that it is exacerbated with modern multinational entities to the point where the arm’s length standard is an inappropriate transactional allocation method.
Profits attributed to a jurisdiction may be distorted by a multinational entity, simply by the separate but related parts of that entity manipulating the prices at which goods and services are
transferred internally.[8] This occurs because of the different motivation of the multinational entity that considers itself a whole, and the taxing authorities that recognise the separate parts of the multinational entity, rather than the whole. It is the multijurisdictional activities of the firm that provide opportunities for profit shifting through transfer pricing.[9]
It is recognised globally that multinational enterprises are one economic entity, whether they operate through a branch structure or a subsidiary structure.[10] As such, the aim of the entity as a whole is to maximise profits.[11] This results in the multinational entity having no concern about the price paid or received for internal transactions, except in so far as the overall profits of the multinational entity is maximised.[12] Consequently, the multinational entity takes advantage of different tax rates across jurisdictions by shifting profits from one jurisdiction to another through transfer price manipulation.[13] In essence, profits attributed to a jurisdiction may be distorted by a multinational entity simply by the separate but related parts of that entity manipulating the prices at which goods and services are transferred. The process of charging for internal transactions is known as transfer pricing. When the price is for a non-arm’s length consideration, it is deemed to be a breach of the transfer pricing regime.
Jurisdictions generally adopt, in their domestic legislation, a requirement that transactions between related parties, or separate parts of the one entity, occur at an arm’s length consideration. The arm’s length standard is based on what is referred to as the separate accounting or separate entity approach. The separate parts of an entity are defined by reference to national boundaries, or what is commonly referred to as the “water’s edge”.[14] Income and expenses are then allocated to the relevant jurisdictions on a transactional basis,[15] that is, specific transactions are considered as if they were between distinct entities, with each entity reporting separate taxable income.[16]
Empirical evidence suggests that multinational entities are undertaking transfer price manipulation, as well as the fact that jurisdictions are losing revenue from such activities.[17] This evidence adds weight the theoretical argument that the traditional regime allows for manipulation which results in an outcome that fails to allocate income according to economic reality. While studies have
generally found that multinational entities are shifting profits through transfer pricing, evidence suggests that this is particularly the case with modern multinational entities such as multinational banks. It is believed that multinational banks have, even more than other multinational entities, opportunities for reducing their tax burdens in high-tax countries by way of intra-firm transfer pricing.[18] In their study of multinational banks Also Demirguc-Kunt and Harry Huizinga found that taxes paid by foreign banks were relatively low in many of the major industrialised nations. By examining the relationship between the taxes paid and the statutory tax rate, they obtained further support for their profit shifting hypothesis. Their study showed a negative relationship between taxes paid and the statutory tax rate suggesting the presence of profit shifting.[19] As additional support for the profit shifting premise Demirguc-Kunt and Huizinga, controlling for bank characteristics, found that foreign banks payed lower taxes in several developed countries. Further, on a cross-country basis, they found that taxes paid by foreign banks fall with the statutory tax as additional evidence of profit shifting by foreign banks.[20]
While it is the multinational entities that undertake such behavior, it is the Governments of the relevant jurisdictions that are concerned about revenue loss. As evidence of revenue losses by taxing authorities, Eugene Lester suggests that in the United States millions of dollars are being lost each year through profit shifting.[21] Lester relies on two sources: Senator Dorgan[22] who asserted that Internal Revenue Service studies show that the United States loses at
least two billion dollars a year through tax evasion under the arm’s length principle;[23] and James Wheeler, Professor of Accounting at the University of Michigan, who estimated that foreign multinational entity groups operating in the United States cheat the United States Treasury out of about thirty million tax dollars each year.[24] Consequently, if these studies and assertions are correct, there is a failure of the arm’s length standard to allocate profits in an optimal manner.[25]
Because of the potentially large revenue losses to taxing jurisdictions through transfer price manipulation, the OECD has devised a means of adjusting transactions internal to the multinational entity. The aim of these adjustments is to achieve a result that means the allocation of income purportedly reflects the location of the activities performed.
The solution to transfer price manipulation, devised by the OECD, is arm’s length pricing. Under the arm’s length pricing model each part of the multinational entity is treated as a separate part of the economic entity (whether it is a branch or a subsidiary) and a price is substituted that would have been used in the transaction if it had been with an unrelated third party rather than a related entity within the same multinational enterprise. This solution is designed to apply to all multinational entities, and to date, the OECD does not consider that any specific type of multinational entity fall outside the
bounds of the arm’s length pricing regime. The OECD maintains this method of transactional allocation as the theoretically superior model despite the arm’s length method being criticized for theoretical, empirical, and administrative reasons.[26]
The arm’s length concept, as the solution to transfer price manipulation, has been around for over 80 years, yet it is only in the past 20 years that the OECD has started articulating ways in which tax authorities can determine an arm’s length price.[27] The OECD released its original guidelines as to how to determine an arm’s length price in the 1979 report on transfer pricing (“1979 Report”).[28] As a response, however, to increased cross-border transactions and multinational enterprise operations, the OECD revised its international transfer pricing guidelines (“1995 Guidelines”) in 1995.[29] The current guidelines represent a consensus among 25 OECD member countries on the approach to international transfer pricing issues. Because of the limited membership of the OECD, a caveat must be placed on the reliance on the OECD to provide a solution to the problem of transfer price manipulation.[30] At present, however, the OECD is undertaking the most comprehensive investigation of this problem. As such, the OECD documents are primarily considered. To this extent, the approach of the OECD towards multinational banks is considered.
At the core of OECD policy on the application of the transfer pricing regime to multinational banks are two suppositions. The first supposition is that multinational banks should be treated the same as any other multinational entity.[31] The second supposition is that branches of multinational banks should be treated the same as subsidiaries of multinational banks.[32] These suppositions provide the foundation for the current application of the transfer pricing provisions and the arm’s length principle to multinational banks. Each is examined in turn.
The first supposition underpinning the transfer pricing regime and the arm’s length requirement, as it applies to multinational banks, is that multinational banks should be treated in the same manner as traditional multinational entities. The OECD believes that the traditional transfer pricing regime is suitable for multinational banks, yet there is, to a limited extent, recognition of some of the pertinent issues that arise with this modern multinational entity. The 1995 Guidelines did not identify multinational banking as a unique type of multinational entity. It did recognise, however, that some of the traits of multinational banking were a cause of the complexities for tax authorities,[33] particularly the increase in integration and technological progress.[34]
Eleven years earlier the OECD released its 1984 Report.[35] In this document, it did specifically recognise the need for special consideration in the treatment of international bank transactions.[36] This document acknowledges the unique issues that arise in relation to a multinational bank branch, as contrasted with a subsidiary, and expands on the general transfer pricing guidelines to comment on some of these unique issues. In this context, while the difference between a branch and a subsidiary is immaterial if a tax regime reflects economic reality, the current transfer price regime draws a distinction, if only in its application.
The OECD also appears to recognise, at least superficially, the distinguishing and unique features of multinational banks in the 1984 Report. While not explicitly stating that there are significant differences, the OECD recognises both the unique provision of services (that of being an intermediary) and a unique organisational structure. As to the former, the OECD states that the essential feature of banking is that the funds of customers entrusted to the bank are on loaned in a manner designed to make a profit.[37] This essentially provides the fundamental difference between a multinational bank and other traditional multinational entities – that of an intermediary service provider. The OECD, however, does not expand on this unique feature to acknowledge the inability of the transfer price regime to capture many of the transactions. As to the second unique feature, the OECD also recognises that the new business of these multinational banks is largely driven by Governments, multinational enterprises and other very large institutions or corporations,[38] thereby supporting the theory that banks become international because they are protecting their client base (defensive expansion) and internalising information asymmetry (internalisation theory). The document goes no further than to recognise these unique features, and it does not provide that it is these traits that cause the unique tax issues. It is suggested, however, that the link can be implied.
The most significant failing of the 1984 Report is that it assumes that most multinational banks operate in a decentralised manner. This assumption is based on the further postulation that the main task of a bank operating multinationally, is to meet the needs of customers in the areas in which they are located.[39] Based on these assumptions, the OECD concludes that “a high proportion of all the transactions undertaken by a multinational bank, whether concluded with third parties or with the head office or other branches or subsidiaries of the bank, would be based in the ‘arm’s length principle’“.[40] It is no longer the case, however, that the majority of multinational banks operate under a decentralised organisational structure.[41] Consequently, it is incorrect to assume that transactions are based on the arm’s length principle, especially where the organisational structure is highly integrated. Further, it is incorrect to assume that the multinational banks are primarily servicing customers within the geographical location, as this is inconsistent with the notion that banks follow existing clients abroad.
As a concession to the difficulties facing multinational banks to determine transfer prices, the OECD recognises that transactions between parts of the one entity might become complex, thereby creating considerable difficulty in the allocation process.[42] In 1984, therefore, while conceding that multinational banks are unique, the OECD did not provide for special tax considerations or the need for different international tax rules. This stance was justified on the (albeit perhaps fallacious) basis that it is possible to reach an arm’s length price for transactions undertaken within a multinational banking entity. At this point, the OECD was not prepared to recognise the truly global nature of modern multinational entities.
Fourteen years later, the OECD, in its 1998 Draft,[43] again addressed the fact that multinational banks raise unique international tax issues. This 1998 Draft has subsequently been updated by the incorporation of parts of it into the 2003 Pt III Draft on the attribution of profits to permanent establishments.[44] Because this document has not been finalised its status is merely that of a discussion draft. It does, however, incorporate and reflect the views of some member states such as the United Kingdom and the United States of America. The 2003 Pt III Draft addresses the taxation of global trading, an activity undertaken by multinational banks.[45]
The 1998 Draft adopts the view that there is consensus that issues raised by global trading should be resolved by reference to the already existing OECD guidelines, with the proviso that the generally accepted principles may require elaboration for this “relatively new and highly specialised business”.[46] While not explicitly stated, this approach has been carried through to the 2003 updates of this discussion draft. This assumption as to consensus, however, may not be accurate, as the increase in new types of multinational entities, as contrasted with traditional ones, has made jurisdictions question the adequacy of traditional arm’s length pricing principles. For example, the European Union, in the last couple of years has made substantive moves towards an acceptance that the traditional methods may no longer work.
The 1998 Draft aims to address the challenges facing the traditional tax system and recognises that technological change, the communications revolution, and the spread of financial deregulation and liberalisation, have all contributed to these challenges.[47] As such,
it accepts that the traditional tax system is coming under increasing pressure with respect to its accurate allocation of income. Consequently, the 1998 Draft concedes that the transactions undertaken by multinational banks are unique, without conceding the need for any special tax rules. The conclusion, however, that any special circumstances can be dealt with through the application of the existing transfer pricing rules,[48] suggests that very little is achieved with this Draft.
Most recently, the OECD, in 2001 released its 2001 Draft, on the attribution of profits to permanent establishments,[49] which has also subsequently been updated.[50] While the 2001 Draft generally considers the attribution of profits to permanent establishments and establishes a working hypothesis for doing so, the 2003 Pt II Draft specifically deals with special considerations in the attribution of profits to permanent establishments of banks. This discussion draft acknowledges the earlier 1984 Report as the starting point for an analysis of the application of the working hypothesis to permanent establishments of banks, but qualifies this by recognising the changes that have occurred in the last two decades. The 2003 Pt II Draft states:
[T]here have been considerable changes in the global economy since 1984, which have affected the way multinational banks carry on business. There have also been changes in thinking about the application of the arm’s length principle, reflected most notably in the revision of the OECD Transfer Pricing Guidelines started in 1995. This report is therefore intended not only to update the issues and situations described in the 1984 report but also to deal with particular issues and situations arising from the widespread financial liberalisation and globalisation of financial markets which have been such a feature of the global economy in the late 20th Century. For example, while the risk has always been of significant concern to banks, technological developments in the late 20th Century have resulted in the willingness of banks to undertake pro-active risk management as a means of maximizing shareholder wealth and of dealing with risk-based capital adequacy requirements.[51]
This Draft is recognising not only the changes per se, due to globalisation, but also the liberation of the financial markets that occurred in Australia and the world in the last two decades. The fact that this Draft is designed to be generally applied to permanent establishments of multinational entities, but nearly half is devoted to its application to permanent establishments of multinational banks, is alone testament to the fact that the OECD sees banking as potentially one of the most problematic of industries to apply the traditional transfer pricing regime to.
Over a two-decade period these three documents, only one of which has OECD status, all concede that unique issues arise in relation to the international taxation of multinational banks. The overall theme, however, of these three documents is that the traditional arm’s length pricing requirement can be applied to such banks, whether operating in subsidiary or branch form. In doing so, the OECD is attempting to maintain the stance that the traditional methods of determining the arm’s length price should be applied to any related party transactions. To this extent, it may be interpreted that the OECD is attempting to embrace the notion that the transfer pricing problem has been adequately dealt with.[52] There is empirical and anecdotal evidence to suggest the contrary, and as Michael Graetz points out, this lack of controversy is “no doubt only a temporary lull until the next round of transfer pricing abuses captures the attention of ... policymakers”.[53]
The second supposition underpinning the transfer pricing regime and the arm’s length requirement, as it applies to multinational banks, is that branches and subsidiaries should be treated the same. The 2001 Draft, using art 9 (Associated Enterprises) to extrapolate principles to apply to the application of art 7 (Business Profits), was clear that branches and subsidiaries should be treated the same.[54] To this extent, the 2001 Draft provides that the requirement of art 7(2) of the OECD Model Tax Convention on Income and on Capital, with its origins dating back to the “separate accounting” approach adopted by the League of Nations in 1933, is considered the statement of arm’s length principle in the context of permanent establishments, corresponding with the arm’s length principle in art 9.[55]
The OECD’s 1998 Draft was a little more equivocal.[56] It recognised that some countries believe that operations that are economically identical should produce similar results, whether conducted in branch or subsidiary form.[57] It also recognised that other countries agree with this general principle but also believe that the legal form of the trading operations does affect economic substance, and this should be reflected in the taxation of those entities.[58] Further, it acknowledged that legal form may affect the economic substance in a number of ways. The 1998 Draft provides:
For example, if a branch enters into a derivative contract with a customer, the customer is able to rely for payment, not just on the capital resources allocated to the branch but on the capital resources of the whole enterprise. However, if a customer enters into the same contract with a subsidiary, it is only the capital resources of the subsidiary which are available to meet any payments due, unless there are specific legal arrangements such as guarantees which allow access to the capital resources of the parent. The difference may affect the price charged to the customer.[59]
Under the current transfer pricing regime the question is whether a multinational bank branch should be treated the same as a multinational bank subsidiary. This article proposes that, rather than treat a branch and subsidiary the same, there is no place in the tax rules governing jurisdiction to tax and allocation of income of multinational banks to recognise the distinction to begin with. While the difference may appear to be a subtle one, the ramifications are significant. Under the OECD model, which recognises the difference, it is then necessary to postulate what a branch and subsidiary would look like if they were the same.
The next Section considers the legislative approach of substituting an arm’s length price under the traditional regime as applied to multinational banks – specifically the operation of the arm’s length standard. This is done to determine whether the problem of transfer price manipulation by multinational banks is dealt with under the traditional regime in a manner which reflects economic activity. It is contended that the arm’s length pricing requirement of the traditional transfer pricing regime is not optimal for taxing multinational banks and, as such, the focus of this article is to disprove the view that the arm’s length requirement is a satisfactory standard to apply.
It has been established that the OECD believes that the arm’s length requirement of the traditional transfer pricing regime should be applied to multinational banks. There is, therefore, no special regime to examine in the application of transfer pricing rules to multinational banks, rather the general regime must be considered. This is done to demonstrate the less than optimal results.
In most jurisdictions, the OECD requirement for goods and services to be transferred at an arm’s length price has been adopted and has its foundation in two sources: the domestic tax law, imposed by the taxing jurisdiction unilaterally; and the double tax agreements to which the jurisdiction is a party.[60] The domestic law generally grants the relevant taxing authority the power to substitute a consideration that is more or less than arm’s length consideration (or if there is no consideration at all) with one that is equal to an arm’s length price.[61] The Associated Enterprises Articles[62] and the Business Profits Articles[63] of a jurisdiction’s double tax agreements achieve a similar result for those countries that have entered into agreements.
Australia is an example of a comprehensive acceptance and adoption of the OECD views to transfer pricing and the arm’s length pricing methodologies. The requirement of an arm’s length price for related party transactions is contained both in Australia’s Income Tax Assessment Act 1936 (Cth) (“ITAA36”) and the double tax agreements to which it is a party. Australia’s treatment of related party transactions has evolved over the last 20 years to a level of relative sophistication, consistent with OECD guidelines.[64] To this extent, Australia has not disagreed with the overall OECD approach to transfer price manipulation, although it could be argued Australia has participated disproportionately in the development of OECD transfer-pricing guidelines.[65]
Australia’s current transfer pricing regime is contained in Div 13, Pt III of the ITAA36[66]and the relevant double tax agreements.[67]This legislation has been supplemented by numerous Taxation Rulings,[68] which in turn have, in general, relied on the work of the OECD. As such, the Australian approach is used in this article to demonstrate the inadequacies of the traditional arm’s length concept when applied to multinational banks. While this analysis aids an understanding of the operation and associated problems with the traditional regime, the caveat that Australia’s approach to transfer pricing may be viewed as heavily procedural and has very little to do with the interpretation of the law, needs to be placed on any discussion.[69] The transfer pricing regime applies to all related party transactions, whether for goods or services, and whether it is between separate legal entities or within the one legal entity. A distinction, however, is drawn between inter-entity transactions and intra-entity transactions. Different provisions apply to a subsidiary as compared to a branch (permanent establishment) both under domestic legislation and the double tax agreements. This different application is one which has resulted in significant discussion in relation to multinational banks. Evidence of this is found in the fact that the OECD has concentrated on these differences in the three documents discussed earlier, with the most recent document, the 2001 Draft, as updated by the 2003 Pt II Draft, specifically dealing with the application of the Business Profits Article to branches of multinational banks.
Where the multinational bank operates through a parent-subsidiary structure, the subsidiaries of the multinational bank operate as independent legal entities as distinct from the head office and is separately chartered under the laws of the local country. With respect to the transfer pricing legislation that applies to this type of relationship, the critical section of the domestic legislation, ITAA36, is s 136AD[70] headed “Arm’s Length Consideration Deemed to Be Received or Given”. This section provides the four conditions that must be satisfied before an arm’s length price can be substituted for the consideration actually paid or received: there is the supply or acquisition of property under an international agreement; the Commissioner, having regard to any connection between any two or more of the parties to the agreement or to any other relevant circumstances, is satisfied that two or more parties to the agreement were not dealing at arm’s length with each other; the consideration paid or received by the taxpayer in respect of the acquisition or supply was greater or less than the arm’s length consideration in respect of the acquisition or supply, or no consideration was given; and the Commissioner determines that the section should apply in relation to the taxpayer.[71] Where the Commissioner determines that the section should apply, it is necessary to consider what the arm’s length price is that may be substituted.
Where there is a double tax agreement the domestic legislation is complemented by the Associated Enterprises Article in those agreements,[72] which generally follow the approach of art 9 of the OECD Model Tax Convention on Income and on Capital (“Model Tax Convention”). The Model Tax Convention provides:[73]
Where:
(a) an enterprise of a Contracting State 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 a Contracting State and an enterprise of the other Contracting State,
and in either case conditions are made or imposed between the two enterprises in their commercial or financial relations which differ from those which would be made between independent enterprises, then any profits which would, but for those conditions, have accrued 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.
The provisions of art 9 only apply where the enterprises are associated, that is, there is a parent-subsidiary relationship or the companies are under common control, and commercial or financial relations between the enterprises differ from normal market terms. As with the domestic legislation, the Commissioner may then include in the profits of the enterprise any amount that would have accrued under normal market terms. If the article is applicable, again the question arises as to the method used to determine the normal market terms or, in other words the arm’s length price.[74]
Where there is a permanent establishment (branch) of a multinational bank, the legislation also requires an arm’s length price for internal transactions. In Australia, internal transactions are generally ignored for domestic tax purposes. Although, in 1981 special rules were incorporated into the transfer pricing provisions, in the form of s 136AE of the ITAA36, permitting the Commissioner to participate in the determination of the source of income attributable to both inbound, and outbound permanent establishments.[75]
Further, due to a statutory exception, contained in Pt IIIB of the ITAA36, introduced with effect from 28 November 1994,[76] Australian branches of foreign banks are treated exactly the same as Australian subsidiaries of foreign banks.[77] This measure was introduced, at the same time legislation was passed to allow foreign bank entry into Australia in branch form. It was introduced to eliminate any perceived taxation advantages of Australian branches over subsidiaries of foreign banks. By virtue of the statutory exception, the Australian branch is treated as if it were a separate legal entity.[78] This results in any transactions between a branch and its head office being treated as true dealings for the purposes of taxation. For example, the two most frequent types of financial transactions between a branch and head office – derivative transactions[79]and foreign exchange transactions[80] – are treated for tax purposes as if they were between separate legal entities, which results in the different status between a branch and a subsidiary being removed.
The separate entity approach is extended to Div 13 of Pt III of the ITAA36. In particular, s 160ZZW(5) ensures that s 136AD, which allows the Commissioner to substitute an arm’s length price in relation to any international dealings between associated parties, applies to transactions between a branch and head office. Normally s 136AD could not apply to any internal profit shifting. Section 160ZZW(5) provides that for the purposes of Div 13 of Pt III, the branch is taken not to be, and not to have been at any time since its establishment, a permanent establishment in Australia of the bank. This provision is necessary as s 136AD can only apply where there is an international agreement as defined. Without this provision, transactions between a permanent establishment (branch) and a non-resident parent do not fall within the definition of an international agreement for the purposes of s 136AD.
While these statutory provisions apply to Australian branches of foreign banks, they do not extend to internal dealings between an Australian bank and its foreign branches or dealings between foreign branches of an Australian bank. This fact alone, prima facie, results in an inequitable taxing of like entities. This problem is one that has been acknowledged by the Australian Government, but not addressed.[81] The Australian Treasury Department recognises that, without legislative provision for the recognition of cross-border financial arrangements, a gain corresponding to a loss in another part of the entity may permanently escape the tax net.[82] It also recognises that there is potentially a result of double taxation where a gain is recognised for tax purposes without the corresponding loss allowed as a deduction.[83]
Where a double tax agreement applies and a more favourable result ensues, a foreign bank may elect to apply not these concessionary measures of Pt IIIB in the calculation of its taxable income. Section 160ZZVB(2) provides that where there is an agreement within the meaning of the Income Tax (International Agreements) Act 1953 (Cth) that has the force of law and applies in relation to the bank, the bank may elect that Pt IIIB is not to apply in the calculation of its taxable income where one of three situations arises. The first situation is where the taxable income of a foreign bank, attributable to activities carried on by the Australian branch, is greater when calculated under Pt IIIB than when calculated without the assistance of Pt IIIB. The second situation is where, if a foreign bank would be taken not to incur a loss in a year of income in respect of activities carried on by the bank through its Australian branch that it would be taken to have incurred if Pt IIIB did not apply. The third situation is where the amount of a loss that a foreign bank incurred in a year of income in respect of activities carried on by the Australian branch under Pt IIIB is less than the amount of the loss if Pt IIIB does not apply.[84]
Where a double tax agreement is relevant, the Business Profits Article applies to attribute profits to a permanent establishment. Australia’s Business Profits Articles in the double tax agreements to which they are a party generally follow art 7 of the OECD Model Tax Convention. Paragraph 1 of that article provides:
The profits of an enterprise of a Contracting State shall be taxable only in that State unless the enterprise carries on business in the other Contracting State through a permanent establishment situated therein. If the enterprise carries on business as aforesaid, the profits of the enterprise may be taxed in the other State but only so much of them as is attributable to that permanent establishment.
Article 7(2) aims to prevent any distortion of the profits or losses of the permanent establishment by deeming the amount of the transaction to be that of an arm’s length transaction. It provides:
Subject to the provisions of paragraph 3, where an enterprise of a Contracting State carries on business in the other Contracting State through a permanent establishment situated therein, there shall in each Contracting State be attributed to that permanent establishment the profits of which it might be expected to make if it were a distinct and separate enterprise engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the enterprise of which it is a permanent establishment.
Paragraph 3 of art 7 is a continuation of this, as it allows expenses to be taken into account and, reading the paragraphs together, to be calculated at an arm’s length price. Article 7(3) provides that “in determining the profits of a permanent establishment, there shall be allowed as deductions expenses which are incurred for the purposes of the permanent establishment, including executive and general administrative expenses so incurred, whether in the State in which the permanent establishment is situated or elsewhere”.
The 2001 Draft specifically deals with the application of art 7 to permanent establishments and, in particular, Pt II, as updated by the 2003 Pt II Draft, deals with the application of art 7 to permanent establishments of multinational banks. The discussion draft concentrates on determining the profits to be allocated to a permanent establishment by virtue of a working hypothesis. The working hypothesis provides “that the profits to be attributed to a permanent establishment are the profits that the permanent establishment would have earned at arm’s length as if it were a separate enterprise performing the same functions under the same or similar conditions, determined by applying the arm’s length principle under Article 7(2)”.[85]
The application of the transfer price regime, therefore, depends on a number of factors: whether the related party transaction involves a subsidiary or a permanent establishment; and whether a double tax agreement is relevant or it is merely a case of applying the domestic laws of a jurisdiction. In essence, before the determination of an arm’s length price, an unnecessarily complicated legal process ensues. The following diagram summarises the relevant regime applicable under Australian legislation to the possible scenarios:
The application of this process then leads to the necessity of determining an arm’s length price.
Whether the multinational bank operates through branches or subsidiaries, and whether or not a double tax agreement applies, once it is established that related party transactions have occurred at a non-arm’s length price there may be a substitution of that price for one that represents an arm’s length price. Whether it is the domestic legislation that applies, or a double tax agreement, and whether the related party transactions are inter-entity or intra-entity, to determine that an arm’s length substitution is warranted, all provisions lead back to the matter of determining an appropriate arm’s length price. This determination is the crux of the traditional transfer pricing regime and has caused the most debate.
The legislation and double tax agreements, while they provide for the substitution of an arm’s length price where one is not paid or given, offer no further assistance in the determination of that arm’s length price. Further guidance is sought from the OECD guidelines, and domestically, Australia’s Taxation Rulings (which have followed the approach of the OECD). To establish this arm’s length price certain methodologies have been determined to ascertain the substituted price. Both sources support the use of the arm’s length method for determining the substituted price, with five methods suggested as appropriate.
The five arm’s length methodologies, recognised both internationally and by the Australian Taxation Office (“ATO”), can be categorised into two basic divisions: the traditional methods; and the profit methods. The first category consists of the comparable uncontrolled price, the resale price and the cost plus methods. It is the view of both the OECD and the Australian Commissioner that these are to be preferred over the alternative profit methods. The second category includes, but is not limited to, valuation procedures such as the profit split and the profit comparison methods.[86] While both the Commissioner and the OECD have stated that the traditional methods are preferred over the profit methods,[87]it has been recognised that highly comparable data is required in order to apply the former.[88]Where there is adequate highly comparable data the preferred traditional approach is the comparable uncontrolled price method.[89]
It can be seen from the five methods that the obvious and preferred method of determining the arm’s length price is to simply look at similar transactions between unrelated parties. This, however, is not always possible. To overcome this limitation the OECD has devised the other four methods and suggested when they should be used. Again, Australia has followed this procedure and issued rulings explaining how it understands the OECDs guidelines to work.
By providing alternative guidelines in addition to the obvious one (comparable uncontrolled price) the OECD is in effect conceding that certain multinational entities are unique. The parts of the enterprise operate together to create a synergy individual enterprises cannot replicate.[90]
As a consequence, in many cases there will be no comparable transaction between unrelated parties. It is then necessary to go through the fiction of pretending it is possible, and then use OECD proposed procedures to determine what the price would have been had there been a transaction between unrelated parties.[91] Nevertheless, the OECD argues that it is possible and the comparable uncontrolled price method can be used to allocate income between parts of a multinational entity. A closer look suggests the rules may not be optimal for taxing multinational banks and the alleged benefits are at best exaggerated and at worst wrong.
The question then is, if there is an air of artificiality about applying the arm’s length rules then why is this method used? The OECD claims that the transfer pricing regime, imposing the requirement of an arm’s length price for related party transactions, is an optimal model for determining the price of transfers between different parts of a multinational entity.[92] While the OECD maintains this argument for all types of multinational entities, this article asserts that the arm’s length model is not optimal for multinational banks.
In the next Section, the theoretical rationale for maintaining the arm’s length requirement is examined to demonstrate that while it may be reasonable to maintain this standard for traditional multinational entities, such logic does not transpose to multinational banks.
The attention given to transfer price manipulation by international tax policymakers in recent times has placed emphasis on the difficulties faced in enforcing the requirement that related-company prices be equivalent to those that would occur in arm’s length transactions between unrelated companies.[93] Very little has been done, therefore, by bodies such as the OECD to determine whether the arm’s length standard for determining transfer prices should be retained. As such, it must be assumed that the original rationale for adopting the arm’s length price is maintained.
The rationale for maintaining the arm’s length requirement for multinational banks is that the current regime is an optimal one for distributing the taxing rights of all multinational enterprises. While the 1984 Report, and the more recent discussion drafts dealing with international banking, specifically acknowledge that unique taxation issues arise in relation to multinational banking transactions, they fail to acknowledge the cause of these issues being the unique features of multinational banks as contrasted with their more traditional counterparts.
The consequence of this acceptance by the OECD that multinational banks are essentially the same as traditional multinational entities means that it is assumed that the rationale for maintaining the use of the arm’s length standard for traditional multinational entities is transposed to multinational banks. This is also the view of some authors who believe that the existing rules are adequate to deal with new means of delivering goods and services.[94] Based on the conclusion that multinational banks are unique, it is posited that the arm’s length approach fails to correlate with economic reality when applied to these entities. As a consequence, the rationale for the use of the arm’s length price for all multinational entities is considered to determine whether the same rationale is true for multinational banks.
In the current literature support for the arm’s length requirement of the transfer pricing regime is based on four perceived theoretical advantages: it is the internationally recognised standard;[95] it is an objective and determinate standard; it creates neutrality between affiliated and unaffiliated firms; and the allocation of income is based on economic reality.[96] It is also argued that, at a practical level, the arm’s length principle has been successfully applied in the majority of cases.[97] This practical application does not, however, translate into an accurate allocation of income. Each of the purported theoretical advantages is considered in turn.
The most compelling argument in favour of the arm’s length requirement is that it is one which has gained international acceptance[98] through its objective and determinate standard.[99] The arm’s length standard has both longevity and international adoption.[100]It is also argued that it has been remarkably successful over the years,[101] accurately reflecting, when first developed, the income earned by various parts of the multinational entity.[102] Further, it is undisputed that globally it is the more commonly used method.[103]
Some defenders of the arm’s length price model argue that it has the force of customary international law,[104] while others argue that treaties allow the adoption of formulary apportionment.[105] While the fact that most jurisdictions adopt the arm’s length standard can be construed as the standard gaining international acceptance,[106] it can also be interpreted as consequence of history. It is this proposition that Stanley Langbein puts forward.[107] The arm’s length standard is a product of history with its foundations laid in a report commissioned by the League of Nations and delivered in 1933.[108] The Carroll report[109] advocated the separate accounting approach as the preferred model for the allocation of profits under the model conventions.[110]
Relative to today’s multinational entities, there was very little activity undertaken by such entities, prior to the 1960s. The multinational entity, as a concept, is a relatively new phenomenon, with multinational banking only becoming prominent in the 1980s. As such, prior to the 1960s very little attention was paid to transfer pricing issues and the arm’s length pricing requirement. While the arm’s length pricing requirement existed in OECD Model Tax Conventions and domestic legislation, the methods for determining this arm’s length price had not been considered in any detail. In 1979, the OECD released its first report into transfer pricing,[111] and has continued to actively examine transfer pricing issues ever since. As a response to increased cross border-transactions and multinational entity operations, the OECD revised its international transfer pricing guidelines in 1995. The revised guidelines, issued on 13 July 1995, replace those contained in the 1979 Report on transfer pricing.
Both the United States and Australia are domestic examples of this relative lateness in the consideration of methods to determine the arm’s length price. In 1968, the US Treasury, for the first time, considered the methods for determining the arm’s length price. It has continued to refine the methods ever since.
In Australia, prior to 1981, the former Div 13, Pt III of the ITAA36 dealing with transfer pricing, consisted of one section, s 136, an original provision without amendments, of the ITAA36.[112] Australia’s reform of its international transfer pricing regime commenced in 1981 with amendments to the ITAA36, resulting in the introduction of the present Div 13 of Pt III of that Act. It has not been until the 1990s, however, that the ATO has taken an active interest in the operation of the transfer pricing provisions. Since 1992 the Commissioner has issued numerous rulings directly relating to international transfer pricing.[113] The ATO has issued these comprehensive rulings outlining not only the selection and application of transfer pricing methodologies, but also documentation requirements and penalties for non-compliance.
Adoption of the arm’s length standard at a domestic level, along with the supposed international acceptance, appears to generally stem from the belief that the OECD Model Tax Convention requires it. This, however, can be disputed. The standard required by the Associated Enterprise Article of the OECD Model Tax Convention[114] is that the parties deal with each other as independent parties would. This article does not contain a requirement for an arm’s length price.
The OECD interpretation of this article, however, is that there is a compulsory requirement for the use of the arm’s length principle, which has become stronger in recent times. While the OECD, in the 1979 Report, indicates that the arm’s length principle is “the underlying assumption behind Article 9”, the 1995 Guidelines provide that it is “the authoritative statement of the arm’s length principle”.[115] The 1998 Draft also refers to the view expressed in the guidelines that multinational entity “groups retain the freedom to apply methods not described in this Report to establish prices provided those prices satisfy the arm’s length principle in accordance with these Guidelines”.[116] This statement again focuses on the use of the arm’s length standard as the only available method for transactional allocation.
This acceptance of the arm’s length standard, both domestically and internationally, appears to stem from the fact that there is a general belief that to determine how independent parties would deal with each other, the arm’s length standard must be applied. This assumption, however, necessarily assumes that prices must be determined for transactions. Yet, as Dale Wickham and Charles Kerester point out “[a]llocating or apportioning taxable income among countries is the task to be accomplished; that is quite different from allocating or apportioning taxable income among internal units of a business enterprise”.[117] Essentially, the purpose of art 9 is to determine how the profits of an enterprise should be split between the relevant jurisdictions. So long as that split is according to how unrelated parties would split the profits, there is no reason why a method other than the arm’s length method could be used. To this extent, commentators have started to question whether the current interpretation of art 9 of the OECD Model, requiring strict use of the comparable transactional pricing approach, is warranted.[118]
There is similar discussion in relation to internal transactions where there are branches of the multinational entity. In relation to art 7, the 2001 Draft, in the context of attributing profits to permanent establishments, comments on the wording of para 4 which states:
Insofar as it has been customary in a Contracting State to determine the profits to be attributed to a permanent establishment on the basis of an apportionment of the total profits of the enterprise to its various parts, nothing in paragraph 2 shall preclude that Contracting State from determining the profits to be taxed by such an apportionment as may be customary; the method of apportionment adopted shall, however, be such that the result shall be in accordance with the principles contained in this Article.[119]
The 2001 Draft addresses the issue of whether this paragraph allows a method other than purely transactional profit methods to be used to determine the profits to be attributed to a permanent establishment. While the 2001 Draft suggests that this paragraph does not allow the use of other methods, there are two arguments which support such use. The first argument is that the wording is wide enough to allow the use of other methods where it is customary to do so. The second argument is that the paragraph itself talks about an “apportionment of the total profits” thereby opening the door to a profit split method or global formulary apportionment. The 2001 Draft attempts to refute the possibility of the paragraph allowing the use of global formulary apportionment, stating that it is distinguishable on the basis of the last sentence which provides that the result should be in accordance with the principles contained in the Article.[120] This document does, however, recognise that the fact that an attribution under this paragraph is one applying to total profits thereby making it difficult to achieve such a result in practice.[121]
Contrary to the view expressed in the 2001 Draft, there is a suggestion that the related-party articles of the treaties do not prohibit formulary apportionment.[122] The standard required by the treaties is that the related parties must deal with each other, as independent parties would, not at an arm’s length price.[123] To this extent, it is suggested that “[a]n arm’s-length price is only one means of satisfying an arm’s-length standard. It is not the only means”.[124] The most obvious alternative means of determining how independent parties would act is to consider the joint venture relationship. “Many unrelated parties enter into joint ventures with each other under which the combined taxable income (or loss) from the joint venture’s activities with unrelated third parties must be allocated between the members of the joint venture”.[125]
It appears, therefore, that rather than being the internationally accepted standard, the arm’s length requirement is merely a product of history, with domestic jurisdictions simply following OECD interpretations. Further, the methods that determine the arm’s length price are a relatively new area of exploration. The 1995 Guidelines were seen as an important achievement heralding new international consensus.[126] It has been suggested, however, that international consensus on the arm’s length principle is superficial.[127] Evidence of the late consideration of the application of the arm’s length principles supports this claim. Rather than accepting that the traditional methods have also become the internationally accepted norm, it is more likely that jurisdictions have had very little time to consider in detail, both theoretically and practically, whether these methods are, in fact, optimal for taxing even traditional multinational entities. Because of the relative “recency” of multinational banking, there has been even less time to consider the appropriateness of the five methods to determining an arm’s length price for related party transactions.
It is not suggested that at the time when the arm’s length method was formulated it was neither sensible nor practical. Quite the contrary, in an era where transactions involved tangible goods, related parties were relatively autonomous and only a small amount of international trade occurred, rendering it both possible and practical to find genuine arm’s length prices for comparable transactions.[128] Because of the changes in recent times, however, the arm’s length formula has ceased to have any real meaning in many situations involving multinational entities,[129] particularly multinational banks.
The OECD is now recognising the unique issues facing taxing authorities in the current global economy. While they have adopted the position that the traditional transfer pricing regime can be adequately adapted to apply the modern multinational banking industry, the inference is that it is difficult to do this.
The second argument in favour of the arm’s length standard is that it has gained international acceptance through its objective and determinative nature. Proponents of this argument base their line of reasoning on the fundamental concept of “comparability” underpinning the traditional arm’s length methodologies.[130] This concept of comparability relies on the market forces of supply and demand being the best way to allocate resources and reward effort.[131] Part of the economic rationale for the arm’s length principle is that an unrelated party would consider the available alternatives to a prospective transaction and enter into the alternative which provides the most profitable outcome.[132]
The five methodologies considered acceptable generally rely on the concept of comparability. The three preferred traditional methods rely particularly heavily on this requirement. The comparable uncontrolled price method relies on a comparison between “the price for property or services transferred in a controlled transaction to the price charged for property or services transferred in a comparable uncontrolled transaction in comparable circumstances”.[133] The resale price method relies on the comparability of functions between the taxpayer and an independent dealer to determine an appropriate resale price margin.[134] The cost plus method also relies on comparable independent dealings. The cost plus method begins with the costs incurred by the supplier of property (or services) in a controlled transaction for property transferred or services provided to a related purchaser.[135] An appropriate cost plus mark up is then added to this cost, to make an appropriate profit in light of the functions performed and the market conditions.
However, while it may be possible to find comparables for transactions undertaken by traditional multinational entities, it is unlikely to be possible to do the same for multinational banks. It is, however, worth noting at this point that this is contrary to the view of the OECD. In 1984 the OECD was of the view that:
The widespread existence of markets for the borrowing and lending of money in various forms, the fact that banks frequently borrow and lend large sums to each other on inter-banks markets and the common phenomenon of recognised inter-bank lending rates indicates that it would normally be possible to derive arm’s length interest rates for transactions between the various parts of a banking enterprise from the rates charged in comparable transactions between independent parties.
There will often therefore be evidence for arm’s length prices on the comparable uncontrolled price basis and the cost-plus or resale-minus basis will be of much less importance in the context of bank interest than in other contexts. It has to be recognised, however, that usable evidence is likely to be more freely available for short-term borrowing than for longer-term borrowing. Normally the transactions to be used for comparison should be arm’s length transactions between unrelated banks where the amount lent, the term of the loan, the currency involved and the other conditions are the same or similar to those in question.[136]
Very little has changed over the period between this report and the recently updated 2001 Draft. In the 2003 Pt II Draft on the attribution of profits to permanent establishments it is still maintained that the methods in the Guidelines can be applied to branches of multinational banks, with that document stressing that comparability can be made between the “reward earned from dealings within the bank with comparable transactions between independent enterprises”.[137]
It is suggested, however, that multinational banks are involved in activities that cannot be allocated to separate parts of the enterprise with any certainty.[138] Richard Bird believes that because of the very nature of the multinational entity, an objective and determinative standard is not possible. He states:
[t]he very essence of a multinational enterprise in a sense is thus its ability to achieve higher revenues (or lower costs) from its different subsidiaries as a whole compared to the results that would be achieved under separate management on an arm’s length basis. The allocation of profits within a multinational enterprise is thus inherently and unavoidably arbitrary since such businesses are, as a rule, inevitably “unitary” in character.[139]
As demonstrated later, this is particularly true for multinational banks.
It may be possible to argue that the arm’s length standard is theoretically objective and determinate, but this does not transpose into an economically rational argument for maintaining the standard for multinational banks as a theoretically superior model.
A third argument in favour of the arm’s length standard is that it removes economic incentives to enter a foreign market through foreign direct investment, compared with transacting with an unrelated foreign based entity.[140] The OECD provides that this is a major reason why member countries have accepted the arm’s length principle, as it “provides broad parity of tax treatment for MNEs [multinational entities] and independent enterprises”.[141] This argument is based on the premise that the current tax regime provides a tax neutral environment. That is, the multinational entity’s choices are not affected by tax consequences, but rather are made based purely on commercial reasoning. This argument in favour of the arm’s length price is that the decision of whether to become a multinational entity or to transact with an unrelated party will not be distorted by perceived tax advantages, through profit shifting and tax minimisation. Rather, a decision would be based on an analysis of the economic costs associated with the two alternatives. The arm’s length standard provides similar treatment between multinational entities and independent enterprises. In doing so, there is the avoidance of the creation of tax advantages by the concentration of economic power in large multinational entity groups.[142]
The OECD believes that this is one of the main reasons why member countries and other countries have adopted the arm’s length principle.[143] It argues that the arm’s length principle provides broad parity of tax treatment for multinational entities and independent enterprises. Further, because the arm’s length principle puts associated and independent enterprises on a more equal footing for tax purposes, it avoids the creation of tax advantages or disadvantages that would otherwise distort the relative competitive positions of either type of entity. In so removing these tax considerations from economic decisions, the arm’s length principle promotes the growth of international trade and investment.[144] Berry, Bradford and Hines explain:
The appeal of the arm’s-length standard is that it purports to place, in each taxing jurisdiction, the commonly controlled firm on the same footing as an equivalent uncontrolled firm, thereby providing no competitive advantage or disadvantage to commonly controlled firms, and hence no incentive to alter the ownership structure of corporations as a consequence of the way in which tax liability is defined. Ideally applied, the tax liability of a given firm would be unaffected were it to be totally controlled, dealing only with its corporate parent, or were it to be sold and operated totally independently.[145]
Proponents of this argument believe that any increased efficiency gained through foreign direct investment, which results in decreased costs associated with the multinational entity, should not be taken into account for taxation purposes. Savings, which allow a product or service to be transferred to another part of the entity at a price less than arm’s length, should be ignored. Any costs associated with the external market, which can be avoided through the establishment of a multinational entity should be ignored according to this argument. Such costs include: brokerage costs; costs of defining the obligations of the contracting parties; the risk of scheduling and the related input costs; and the taxes paid on the transactions.[146]
There is, however, a counter argument, that to reflect economic reality, these cost savings should be taken into account. The cost savings associated with multinational banks are caused by two factors. The first factor is the cost savings due to external market failure. The second factor relates to the synergistic advantages of being a multinational bank. Each is examined in turn.
Where intercompany transactions could be carried out with approximately equal (or greater) efficiency by unrelated firms, the neutrality rationale provides a cogent reason for applying the arm’s length standard to those multinational entities.[147] There are, however, very few multinational entities in existence where transactions can be carried out with approximately equal or greater efficiency by unrelated firms. Economic reality is that there is increased efficiency gained through foreign direct investment, resulting in decreased costs associated with the multinational entity.[148] The arm’s length principle does not recognise these efficiencies within the multinational entity.[149] Consequently, the arm’s length price fails to reflect the economic reality of a multinational entity that achieves such efficiencies. The multinational bank is one such multinational entity that exhibits these efficiencies.
Multinational banks are profit driven entities and banks are internalising a market failure – information asymmetry,[150] which means that there is greater efficiency for the banks by operating as multinationals. Internalisation – which is about the imperfections in the intermediate market – explains how the multinational bank achieves the goal of cost reduction. Banks follow their customers overseas because the knowledge advantage they possess, borne of the client-banking relationship, which becomes a public good within the firm that can be best exploited by expanding offshore. The imperfections in the open market create transaction costs, which can be minimised or avoided by operating in the international market through foreign direct investment.[151]
Stanley Langbein is one supporter of the argument that these factors should be taken into account, and maintains that at the heart of the objection to the arm’s length standard is that it is deficient in theory based on this internalising of market failure.[152] He bases this argument on the fact that “[m]ultinational enterprises are integrated for substantial economic reasons; they do not behave or govern themselves ‘as if’ their various components were separate”.[153] Relying on this internalisation of market failure argument, it can be maintained that the current approach is conceptually inconsistent with the economic purpose of a multinational enterprise,[154] which results in the arm’s length requirement being economically illogical.[155] Essentially, it is suggested that the theory of the arm’s length standard does not correspond with the theory of the multinational entity as it disregards the fact that related parties may be economically different from unrelated parties.[156]
In this context of multinational banking, this economic difference is a result of banks entering foreign markets through foreign direct investment. It is only when a bank enters a foreign market through foreign direct investment, thereby creating a multinational bank, that transfer pricing becomes an issue. There are, however, then efficiency gains through foreign direct investment which can be explained by internalisation theory.[157] “Internalisation is about imperfections in intermediate product markets. Intermediate products flow between activities within the production sector. Market imperfections generate transaction costs and these costs are often minimised for the sector as a whole by bringing interdependent activities under common ownership and control”.[158] Implicit in the internalisation theory are the assumptions that: (1) the primary goal of any capitalist entity is to maximise profit, and as part of achieving that goal, minimise expense; and (2) there is an imperfect market.[159] Internalisation explains how the multinational entity achieves this reduction in cost. Costs associated with the external market, which can be avoided through the establishment of a multinational entity are: brokerage costs; costs of defining the obligations of the contracting parties; the risk of scheduling and the related input costs; and the taxes paid on the transactions.[160] There is an argument that the reduction of these costs should be reflected in the transfer price. Referring to internalisation, Stanley Langbein maintains:
This exegesis of the origin of multinational firms, and of the “integration economies” they effect, suggests the futility of constructing a transfer pricing regime based on the identification of “inputs” to the productive process and the association of profit with particular inputs. It suggests, rather, that allocations seek to divide profits among the components of a multinational group according to the relative contributions of the components of the group profit.[161]
It is asserted that the arm’s length approach ignores the interdependence and integration of the multinational entity by treating the various parts as if they were separate unrelated entities.[162] By doing so, this traditional approach ignores the unitary nature, or the economic reality, of the multinational entity.[163] According to economic theory, business activities will be conducted according to economic and commercial imperatives, which determine the most efficient manner in which to conduct that business, including such matters as organisation and financing.[164] The recent World Tax Conference was entitled “Taxes Without Borders”, and as Scott Wilkie explains, the statement “‘Taxes Without Borders’ implicitly reflects an original conception of business enterprise that transcends – exists without regard for – artificial divisions of economic activity based upon states’ particular tax interests”.[165]
While the internalisation of market inefficiencies argument may apply to multinational entities besides multinational banks, it is of greater significance for these banks because of the unique services provided and they way they are provided. In particular, multinational banks are internalising information that is client specific, resulting in reduced transaction costs.
It is further arguable that, also from a theoretical perspective, the arm’s length pricing model fails to take into account the synergies arguably inherent in a multinational enterprise.[166] This failing has been described as follows:
The major theoretical criticism of the arm’s length principle is that it is inherently flawed because it adopts the separate entity approach. Consequently, the arm’s length principle cannot account – at least comprehensively – for the economies of scale or other benefits of integration that accrue to integrated businesses.[167]
Multinational banks generally function as one economic unit, unlike unrelated parties. As such, there are synergistic benefits to the integrated entity as a whole.[168] It may be argued, however, that the arm’s length approach fails to take these benefits into account. This is because the arm’s length approach ignores the answer to the question of why multinational banks come into existence and structure themselves in a particular fashion. The simple answer to this question is that usually multinational entities come into existence because it is more cost effective than contracting with arm’s length parties.[169] Further, when they do come into existence, they structure themselves in a manner that maximises profits.[170] As noted by Langbein, “internal organization economizes on uncertainty and opportunism by allowing for adaptive, sequential decision making; by facilitating the development of internal codes concerning idiosyncratic knowledge; and by promoting convergent expectations, which attenuate uncertainties generated when interdependent parties make independent decisions”.[171] Multinational banks operating on a global basis will benefit from these economics of scale as they will have the ability to minimise their risk, cost effectively obtain technical expertise and information, and preserve their value. [172]
The 1998 Draft on the taxation of global trading of financial instruments acknowledges “the co-operation between the marketers and traders which is essential to produce the profits from global trading”.[173] The 1998 Draft also states that two problematic issues arise because the current tax model may not fully reflect this co-operation. The first issue is the question of how to allocate the benefits of integration. The response to this is found in the guidelines which provide “[t]here are, however, no widely accepted objective criteria for allocating the economies of scale or the benefits of integration between associated enterprises”.[174]
The second issue is the question of how to evaluate the level of integration of functions in respect of a particular transaction or transactions.[175] The 1998 Draft suggests that, in many cases it is still possible to undertake a comparability analysis to determine a comparable uncontrolled price. It is recognised, however, that a number of tax authorities believe that in some global trading cases there is such a high level of integration between locations and functions that traditional transaction methods could not reliably be applied.[176] That is, it is not possible to make reasonably accurate adjustments.
Where it is not possible to use a traditional transactional method, the OECD refers to its guidelines that state: “[u]nder similar circumstances, independent enterprises might decide to set up a form of partnership and agree to a form of profit split”.[177] This would appear to be an admission by the OECD that where there are synergies in global trading, it is unlikely to be possible to use traditional methods. Further, it is often because of the integration that there is an absence of arm’s length markets to be used as comparables.[178] Even where there are such comparisons, which is highly unlikely, the use of the traditional methods will not take account of the integration economies, as non-associated parties have not undertaken such integration.[179]
It has already been established that multinational banks are highly integrated. It is this highly integrated nature that sets apart multinational banks from their traditional counterparts and provides larger profits. As Jinyan Li explains, “[a]n integrated business offers the advantage of increased efficiency, through economies of scale, the internalisation of transaction costs, or synergies that earn profits for the MNE that are greater than the aggregate profits earned by separate enterprises”.[180] The separate entity approach requires the taxpayer to act as if there were no synergies, treating parts of the multinational as if they were distinct.
It is not disputed that the neutrality rationale provides a coherent reason to apply the arm’s length standard to multinational entities where intercompany transactions can be undertaken just as efficiently by an unrelated entity.[181] Many multinational entities, however, enter the market through foreign direct investment because those market transactions are not economically viable[182] – multinational banks are one such example.
It is argued that the discussions above clearly demonstrate that multinational banks are one such group of multinational entities that enter a foreign market because of the advantages of internalising the transactions undertaken. It is generally understood that the income allocated to each jurisdiction should reflect the economic activities carried on in that jurisdiction.[183] It is also becoming generally accepted, however, that “... the manner in which parties express their transactions and account for them based upon principles of separate entity accounting may not, in fact, be a reliable reflection of the extent to which economic activity will have taken place in a jurisdiction”.[184] In the case of multinational banks, an accurate allocation may not be achieved through an application of the traditional arm’s length regime. As such, the neutrality rationale is outweighed by the failure to take into account economic reality.
The criticism that the arm’s length standard fails to take into account economic reality is qualified by the fact that it is possible, if the arm’s length standard were replaced by a different methodology to allocate income, inefficient multinationals would come into existence for tax-driven reasons.[185] This article, however, suggests that this is not a valid reason for maintaining a less than optimal regime.
The attainment of an appropriate replacement which would provide a tax neutral environment (a difficulty this author believes can be overcome), has its potential difficulties. It may be argued, however, that the neutrality argument is clearly flawed in circumstances where multinational banks are born due to market inefficiencies. If this were correct, it would appear that multinational banks are then essentially punished by their election of foreign direct investment. It is suggested that this certainly is the case with most multinational banks, as their decision to become multinational is usually based on their compulsion to follow clients, both to take advantage of existing knowledge and for fear of competitors. This decision to become multinational means that there are synergistic benefits to the multinational bank, which the traditional transfer pricing regime also fails to take into account.
The fourth argument in support of the arm’s length requirement is that it accurately allocates income among affiliates of a multinational entity in proportion to each affiliate’s economic activity.[186] The theoretical justification[187] of this rationale is that it is the most accurate solution because of its “bottom up” approach.[188] The separate accounting approach attempts to determine the origin of the income,[189] and evaluates individual transactions rather than assessing them as a whole.[190] The 1995 Guidelines also list this as an advantage of the separate entity approach.[191] Jinyan Li explains that this perceived advantage is that “the comparable transactional pricing approach provides the most accurate measurement of the fair market value of the true economic contribution of members of a [multinational entity] group”.[192]
There are two substantive objections to this rationale. The first objection is that it ignores the industrial organisation theory of the multinational enterprise[193] as previously discussed. Because this theory argues that the sum is greater than the parts, and the consequential income cannot be allocated to the parts in any principled manner,[194] the result does not accurately allocate income. There are efficiencies which “generate value for the whole enterprise that is greater than the sum of the parts, and it is this value that cannot be meaningfully divided among separate activities conducted by separate legal entities as the arm’s length system attempts to do”.[195] This is particularly relevant to multinational banks due to the integrated nature of global trading as it is difficult to allocate profits among trading jurisdictions.[196] “Global trading involves large amounts of highly mobile capital which would suggest it is relatively easy to transfer expected profit or loss from one jurisdiction to another. The complex and innovative nature of the financial products being dealt with makes the tracing of such transfers more difficult”.[197]
The second objection to this rationale that the arm’s length standard is desirable because it accurately allocates income to source is that this argument rests on the underlying premise that the current source regime is justified.[198]
There are sound reasons to support a source based regime over one based on residency, as despite its imperfections, there are advantages to such a regime.[199] This does not mean the current source regime must be maintained as there are other source models, such as unitary taxation, available.
The current source based regime is not optimal for taxing multinational banks as it does not accurately allocate income among affiliates of the banks in proportion to each bank subsidiary or branch’s economic activity. Given the current transfer pricing regime’s reliance on source, it is naive to suggest that the current arm’s length standard achieves a result which does accurately allocate income among the relevant jurisdictions.
Finally, there is an argument that the arm’s length approach does not in fact resolve the issue of allocating income among countries of source.[200] Dale Wickham and Charles Kerester argue that “it focuses only on allocating revenues and costs among the legally separate but commonly controlled legal entities”.[201] Further, it will “increase the revenues of the country requiring the reallocation to an entity taxed on the basis of the taxpayer’s nationality or residence, rather than the source of the income”.[202]
Also used as the basis for arguing that the arm’s length price results in an accurate allocation of income is that, assuming there is an arm’s length price for a product or service, it should be used as the transfer price, not because there is one for that product or service, but that self-interest dictates.[203] Charles Berry et al argue that the “self-interest of any independent firm would dictate that such a firm would never pay more for any product than the price at which the same product could be obtained from an alternative seller, and that no independent firm would ever accept less, in selling that product, than the price it could obtain from an alternative buyer”.[204] What this argument fails to take into account (besides the obvious one of the need for comparables to exist), is that separate parts of a multinational entity are not motivated by self-interest; rather they are motivated by the economic interest of the group as a whole. A multinational entity will arrange its affairs to maximise the overall profits.[205]
Tax considerations aside, the structure adopted will, in turn, affect the choice of transfer pricing methodology adopted by the multinational entity; whether this is established by reference to market based transfer prices, cost based transfer prices or negotiated transfer prices. The transfer pricing methodology chosen should lead management to make optimal decisions for the organisation as a whole.[206] An organisation chooses an appropriate method according to the criteria of promotion of goal congruence and promotion of a sustained high level of management motivation.[207] The promotion of goal congruence is achieved where divisional managers make decisions consistent with goals set by group management. The promotion of a sustained high level of management effort is achieved where management is motivated to attain set goals.
Setting an appropriate transfer price by reference to market-based (traditional transactional) methodologies are most appropriate where there is a competitive market to provide a price. In these circumstances there is little advantage in dealing with an associated enterprise as compared with an independent one and as such the market price will meet an entity’s goals. However problems may arise, for example, where there is excess capacity which results in a price drop that may be either short or long term. Essentially this highlights two problems for multinational banks. The first problem is that there must be a competitive market to provide the price (transactional comparability). The second problem is one of excess capacity, as it is a very real possibility in that a multinational bank may have excess funds which it can loan at a rate lower than an independent third party simply because it is excess.
While it is not possible to establish a general rule to determine the appropriate transfer price applicable to all multinational enterprises it is possible to determine a general guide. As a general guide the following formula is useful in the context of a traditional multinational entity:
In its application to multinational entities generally, the additional outlay costs per unit incurred to the point of transfer are the variable costs associated with the production of goods, while the opportunity costs per unit to the supply division will be the negotiated portion of fixed costs. This formula can be adapted for multinational banks and the service they provide of operating as an intermediary in the borrowing and lending of money.
This determination of the transfer price, appropriate from a managerial accounting perspective, is very different to that of the arm’s length requirement. Rather than examining opportunity costs, the question that a firm is required to answer for tax purposes, or in other words to determine the arm’s length price, is “what would have happened if the ownership link had been severed and the enterprise was motivated by its own economic interest?”[209] Relying on industrial organisation theory, this question does not accord with the general purpose behind the establishment of a multinational enterprise. As already pointed out, this theory essentially postulates that multinational entities are established so that the group performs better than they would as non-associated enterprises.
To answer the question required by the taxing authorities, by adopting the arm’s length requirement, it would appear that these taxing authorities generally conclude that an enterprise would have sold the goods to a non-associated party at market value and therefore the arm’s length consideration should also be market value. If this is the case then the purpose of establishing a multinational entity is defeated. That is, if you can buy or sell goods from a non-associated party for the same consideration to that of an associated party, the incentive to have related enterprises is severely undermined.
The fundamental theoretical failing of the traditional transfer pricing approach, when applied to multinational banks, is that it does not take into account the unique features of these banks when allocating profits. Yet, it is these unique features, particularly entering the market through foreign direct investment to avoid market externalities, which make the multinational bank is so successful. The separate entity paradigm adopted by the traditional transfer pricing regime is incongruous with the economic theory of multinational enterprises.
Consequently, it has been demonstrated in this article that the traditional OECD approach to the problem of transfer price manipulation of substituting an arm’s length price may not be optimal for multinational banks. Further, it was demonstrated that the rationale for the use of the arm’s length price for traditional multinational entities do not carry over to the enterprise of multinational banking. Despite this, the OECD continues to insist upon the traditional application of the transfer pricing regime.
[*] Senior Lecturer, TC Beirne School of Law, University of Queensland; and Senior Research Fellow, Taxation Law and Policy Research Institute, Deakin University.
[1] D Hay, F Homer and J Owens, “Past and Present Work in the OECD on Transfer Pricing and Selected Issues” (1994) 9 Tax Notes International 249, 254.
[2] OECD, The Taxation of Global Trading of Financial Instruments (1998) (“OECD, 1998 Draft”).
[3] OECD, Discussion Draft on the Attribution of Profits to Permanent Establishments: Part III (Enterprises Carrying on Global Trading of Financial Instruments) (2003) (“OECD, 2003 Pt III Draft”).
[4] OECD, Discussion Draft on the Attribution of Profits to Permanent Establishments (2001) (“OECD, 2001 Draft”).
[5] OECD, Discussion Draft on the Attribution of Profits to Permanent Establishments Part II (Banks) (2003) (“OECD, 2003 Pt II Draft”).
[6] OECD, Transfer Pricing and Multinational Enterprises: Three Taxation Issues (1984) (“OECD, 1984 Report”).
[7] A Plummer, International Combines in Modern History (1934); and DK Fieldhouse, “The Multi-national: A Critique of a Concept” in A Teichova, M Levy-Leboyer and H Nussbaum (eds), Multi-National Enterprise in Historical Perspective (1986) 4, 13.
[8] See, eg, TS Newlon, “Transfer Pricing and Income Shifting in Integrated Economies” in S Cnossen (ed), Taxing Income in the European Union – Issues and Options for Reform (2000) 214.
[9] P Musgrave, “Interjurisdictional Equity in Company Taxation: Principles and Applications to the European Union” in Cnossen, above n 8, 46, 55.
[10] SI Langbein, “Transaction Cost, Production Cost, and Transfer Pricing” (1989) 44 Tax Notes 1391, 1392.
[11] CT Plambeck, “Transfer Pricing Analysis of Global Trading Operations and Procedural Alternatives” (1996) 74 Taxes 1129, 1130.
[12] V Tanzi, “The Nature and Effects of Globalization in International Tax Policy Principle Papers Globalization, Technological Developments, and the Work of the Fiscal Termites” (2001) 26 Brooklyn Journal of International Law 1261, 1266.
[13] Ibid. Although Vito Tanzi, relying on a 1999 Ernst and Young survey states that multinationals continue to identify maximisation of operating performance, not optimising tax arrangements, as the most important factor shaping their behaviour.
[14] See, eg, L Eden, Taxing Multinationals: Transfer Pricing and Corporate Income Taxation in North America (1998) 35; and GN Carlson and H Galper, “Water’s Edge Versus Worldwide Unitary Combination” in CE McLure Jr (ed), The State Corporation Income Tax (1984) 1.
[15] Charles Plambeck refers to this method as the “realization approach”: CT Plambeck, “The Taxation Implications of Global Trading” (1990) 48 Tax Notes 1143, 1155.
[16] RM Bird, “The Interjurisdictional Allocation of Income” (1986) 3 Australian Tax Forum 333, 337.
[17] See, eg, RS Avi-Yonah, “Globalisation, Tax Competition, and the Fiscal Crisis of the Welfare State” (2000) 113 Harvard Law Review 1573, 1591. Reuven Avi-Yonah relies on two previous studies as evidence of tax minimisation by multinational entities through profit shifting: D Harris, R Morck, J Slemrod and B Yeung, “Income Shifting in US Multinational Corporations” in A Giovannini, RG Hubbard and J Slemrod (eds), Studies in International Taxation (1993) 72, 89, 92, 93 table 3.3; and JR Hines Jr, “Tax Policy and the Activities of Multinational Corporations” in AJ Auerbach (ed), Fiscal Policy: Lessons from Economic Research (1997) 401, 414-5. See also, RA Green, “The Future of Source-Based Taxation of the Income of Multinational Enterprises” (1993) 79 Cornell Law Review 18, 19; Newlon, above n 8; and M Mazerov, “Why Arm’s Length Falls Short” [February 1994] International Tax Review 28.
[18] A Demirguc-Kunt and H Huizinga, “The Taxation of Domestic and Foreign Banking” (2001) 79 Journal of Public Economics 429, 430.
[19] Ibid 429. Their estimates suggested that the relationship between the taxes paid and the statutory tax rate was positive for domestic banks, but negative for foreign banks. This was interpreted as profit shifting by foreign banks.
[20] Ibid 429 and 449.
[21] EE Lester, “International Transfer Pricing Rules: Unconventional Wisdom” (1995) 2 ILSA Journal of International and Comparative Law 283, 297.
[22] Senator Dorgan bases his claims on the studies of Professors Simon J Pak and John S Zdanowicz, Florida International University.
[23] In response to Senator Dorgan’s claims see RA Clark, “Dorgan’s Charges of Transfer Pricing Abuse Unfounded” (1999) 18 Tax Notes International 2263. Richard Clarke claims that the majority of MNEs spend considerable resources attempting to comply with the transfer pricing rules.
[24] Lester, above n 21. See also DW Wickham and CJ Kerester, “New Directions Needed for Solution of the Transfer Pricing Tax Puzzle” (1992) 5 Tax Notes International 399 for a summary of United States findings on transfer price manipulation.
[25] RS McIntyre, “Using NAFTA to Introduce Formulary Apportionment” (1993) 6 Tax Notes International 851, 853.
[26] Lester, above n 21, 295.
[27] The arm’s length price was introduced in the 1920s by the League of Nations and formally adopted in 1933.
[28] OECD, Report of the OECD Committee on Fiscal Affairs on Transfer Pricing and Multinational Enterprises (1979) (“OECD, 1979 Report”).
[29] OECD, Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (1995) (“OECD, 1995 Guidelines”). The revised guidelines, issued on 13 July 1995, replaced those contained in the OECD’s 1979 report on transfer pricing.
[30] RS Avi-Yonah, “Commentary: Tax Issues Through Trade Regimes” (2001) 26 Brooklyn Journal of International Law 1683, 1689. Reuven Avi-Yonah suggests that many of the OECD solutions assume that there is no significant development or growth outside the OECD. The OECD is also seen to favour the “Rich countries and ‘their’ MNEs”.
[31] See, eg, OECD, 1984 Report, above n 6; OECD, 1998 Draft, above n 2; OECD, 2001 Draft, above n 4; OECD, 2003 Pt II Draft, above n 5; and OECD, 2003 Pt III Draft, above n 3.
[32] For the most recent evidence of this see the suggested treatment of multinational bank branches in OECD, 2001 Draft, above n 4, updated by OECD, 2003 Pt II Draft, above n 5; and OECD, 2003 Pt III Draft, above n 3.
[33] It is not denied that many of these traits are also associated with other modern multinational entities.
[34] OECD, 1995 Guidelines, above n 29, 1.
[35] OECD, 1984 Report, above n 6.
[36] Ibid 45(1).
[37] Ibid 47(10).
[38] Ibid 47(14). See also p 49, para 24, where the OECD recognises that in former time the main reason for a bank to extend its operations offshore was to service its existing clients.
[39] Ibid 50(28).
[40] lbid 51(32).
[41] A Snyder, “Taxation of Global Trading Operations: Use of Advance Pricing Agreements and Profit-split Methodology” (1995) 48 The Tax Lawyer 1057, 1058.
[42] OECD, 1984 Report, above n 6, 51(32).
[43] OECD, 1998 Draft, above n 2.
[44] OECD, 2003 Pt III Draft, above n 3.
[45] A broader range of financial institutions also undertakes this activity.
[46] OECD, 1998 Draft, above n 2, 30(116).
[47] lbid 7.
[48] Ibid.
[49] OECD, 2001 Draft, above n 4.
[50] OECD, 2003 Pt II Draft, above n 5; and OECD, 2003 Pt III Draft, above n 3.
[51] OECD, 2003 Pt II Draft, above n 5, 3(2). See OECD, 2001 Draft, above n 4, 42(2).
[52] MJ Graetz, “Taxing International Income: Inadequate Principles, Outdated Concepts, and Unsatisfactory Policies” (2001) 26 Brooklyn Journal of International Law 1357, 1420.
[53] Ibid.
[54] OECD, 2001 Draft, above n 4, as updated by the 2003 documents.
[55] Ibid 13(39).
[56] OECD, 1998 Draft, above n 2.
[57] Ibid 57(224).
[58] Ibid.
[59] Ibid.
[60] For a history of this approach internationally see JM Weiner, “Using the Experience in the US States to Evaluate Issues in Implementing Formula Apportionment at the International Level” (1996) 13 Tax Notes International 2113, 2115.
[61] Throughout this article, the domestic legislation of Australia is used as an example of a traditional international tax regime which adopts the OECDs views on the distribution of taxing rights. The adoption of the OECD solution to transfer pricing manipulation is no exception.
[62] OECD, Model Tax Convention on Income and on Capital (1992), art 9 (“OECD, Model Tax Convention”).
[63] Ibid art 7.
[64] R Shaddick, “Related-Party Transactions: Australia” in 2000 World Tax Conference Report 14:1, 14.2.
[65] Ibid.
[66] The present Div 13 was introduced by Act No 29 of 1982, s 19(1). Section 19(3) provides that the amendments apply in respect of income of the year of income in which 28 May 1981 occurred and all subsequent years. Australia’s reform of its international transfer pricing regime commenced in 1981 with amendments to the Income Tax Assessment Act 1936 (Cth) (“ITAA36”) resulting in the introduction of the present Div 13 of Pt III of that Act. Prior to 1981, the former Div 13, Pt III of the ITAA36 dealing with transfer pricing, consisted of one section, s 136, an original provision without amendments, of the ITAA36. This section was based on s 28 of the Income Tax Assessment Act 1922 (Cth), which was based on s 23 of the Income Tax Assessment Act 1915-1921 (Cth), which, in turn, was based on s 31 of the Income (No 2) Act 1915 (UK).
[67] For the history of Div 13 see generally: M D’Ascenzo, “Developments in Transfer Pricing Enforcement and Complex Audit Strategy in the Australian Taxation Office” (1988) 5 Australian Tax Forum 471, 477.
[68] Public and private rulings are binding on the Commissioner provided they are favourable to a taxpayer and the arrangement to which they relate began or begins to be carried out on or after 1 July 1992.
[69] P Anderson, “Transfer Pricing Draft Rulings: Helpful Guidance or a Compliance Nightmare?” (1996) 8 CCH Journal of Australian Taxation 30.
[70] ITAA36, s 136AE is the pivotal section in the context of single entity transactions.
[71] The fact that the Australian Commissioner of Taxation must determine that the section should apply means that the transfer pricing regime is discretionary rather than self executing.
[72] Taxation Ruling TR 94/14, para 18: In considering the application of Div 13, the terms of any relevant double taxation agreement must be considered. The Commissioner may apply the provisions of Div 13 and/or the treaty provisions. In the event of any inconsistency, the treaty provisions will prevail unless the treaty itself gives precedence to the domestic law.
[73] OECD, Model Tax Convention, above n 62.
[74] Paragraph 2 of art 9 allows an adjustment to be made where there is economic double taxation by virtue of an assessment under para 1. The right to an adjustment is not automatic and the other Contracting State need only make an adjustment where it considers that the adjustment is justified both in principle and regards the amount.
[75] For a discussion on the introduction of s 136AE see, Shaddick, above n 64, 14:6.
[76] A statutory exception has also been provided for offshore banking units to the extent that they are treated similarly, that is, they are treated as separate legal units for the purposes of the transfer pricing provisions: ITAA36, Div 9A.
[77] See generally, M Kobetsky, “Foreign Banks: Defining the Personality” (1994) 48 Bulletin for International Fiscal Documentation 243.
[78] ITAA36, s 160ZZVA(2).
[79] ITAA36, s 160ZZV defines a “derivative transaction” as a transaction entered into for the purpose of eliminating, reducing or altering the risk of adverse financial consequences that might result from changes in rates of interest or changes in rates of exchange between currencies, or for the purpose of making a profit from such changes, but does not include a transaction for the provision of finance or a foreign exchange transaction.
[80] ITAA36, s 160ZZV defines a “foreign exchange transaction” as a transaction by which different currencies are exchanged.
[81] Australian Treasury Department, Taxation of Financial Arrangements – An Issues Paper (1996) 250 (“Australian Treasury Department, Financial Arrangements”).
[82] Ibid.
[83] Ibid.
[84] Regardless of whether the income of the foreign branch is determined under Pt IIIB of the ITAA36 or under the double tax agreements, a foreign bank branch will not be able to utilise the Australian foreign tax credit system for any foreign tax paid. The reason for this is twofold. First, the foreign tax credit system is available only to residents whereas by virtue of s 160ZZW, the foreign bank branch is deemed a non-resident. Second, even if the foreign bank branch were a resident for tax purposes, the income would be considered Australian sourced because of s 160ZZY. The inability of a foreign bank branch to avail itself of the foreign tax credit system may potentially lead to double taxation. In order to ensure there is no double taxation, the ITAA36 provides that foreign tax paid during a year of income by a foreign bank on interest received by its Australian branch from a place outside Australia is an allowable deduction for that year of income.
[85] OECD, 2001 Draft, above n 4, 12(32).
[86] It should be noted that the transactional profit methods are different to unitary taxation. Taxing authorities generally believe that an allocation based on profit split is consistent with the arm’s length principle on the basis that it “takes into account the facts and circumstances of each particular taxpayer’s situation and seeks to achieve the best estimation of what independent parties might have reasonably expected to have agreed to in a joint venture relationship”. See, eg, Australian Treasury Department, Financial Arrangements, above n 81, 16.52.
[87] Ibid 2.3.
[88] Ibid 2.27.
[89] Ibid 3.15.
[90] See, eg, DK Dolan, “Intercompany Transfer Pricing for the Layman” (1990) 49 Tax Notes 211, 214; Lester, above n 21, 295; Plambeck, above n 11, 1130; and Langbein, above n 10, 1402.
[91] Graetz, above n 52.
[92] OECD, 1995 Guidelines, above n 29.
[93] Graetz, above n 52, 1419.
[94] See, eg, A Yamanouchi, “International Tax Issues Affecting Electronic Commerce and Banking” (1997) 14 Tax Notes International 1619.
[95] Although this could be viewed as a practical argument for the maintenance of the arm’s length standard. See, eg, H Hamaekers, “Arm’s Length – How Long?” in P Kirchhof, M Lechner, R Arndt and M Rodi (eds), International and Comparative Taxation: Essays in Honour of Klaus Vogel (English version edited by Kees van Raad) (2002) 29, 39.
[96] See, eg, Green, above n 17, 37; SI Langbein, “The Unitary Method and the Myth of Arm’s Length” (1986) 30 Tax Notes 625; and Hamaekers, above n 95.
[97] Hamaekers, above n 95, 38; and Hay et al, above n 1, 253.
[98] RH Cuttler, Formulary Apportionment – Is this Alternative to the Arm’s Length Standard Possible and Practical under the United States’ Current Tax Treaties? (Working Paper, 1996)7.
[99] Green, above n 17.
[100] J Li, “Global Profit Split: An Evolutionary Approach to International Income Allocation” (2002) 50 Canadian Tax Journal 823, 831.
[101] Hay et al, above n 1.
[102] C Thomas, “Customary International Law and State Taxation of Corporate Income: The Case for the Separate Accounting Method” (1996) 14 Berkeley Journal of International Law 99, 102.
[103] D Sandler, “Slicing the Shadow – The Continuing Debate Over Unitary Taxation and Worldwide Combined Reporting” [1994] British Tax Review 572, 573.
[104] See eg, Thomas, above n 102.
[105] See eg, LM Kauder, “The Unspecific Federal Tax Policy of Arm’s Length: A Comment on the Continuing Validity of Formulary Apportionment at the Federal Level” (1993) 60 Tax Notes 1147.
[106] Hay et at, above n 1, 253.
[107] Langbein, above n 96. For a comprehensive rebuttal of Stanley Langbein’s claim see Thomas, above n 102. Thomas states “Langbein’s arguments are ultimately aimed less at demonstrating that countries do not believe the method to be an international norm and more at laying the groundwork from which to argue that they should not adhere to that norm. That is, Langbein asserts that weaknesses in the empirical data supporting the separate accounting method’s normative status exist in order to buttress his own normative argument for the formulary apportionment method. Thus, while Langbein points to valid and indisputable problems with the separate accounting method, these problems do not speak to the actual status of the method as an international norm”: ibid 127-8.
[108]Formulary apportionment was considered and rejected by the League of Nations in 1927 and 1928: Graetz, above n 52.
[109] MB Carroll, Taxation of Foreign and National Enterprises – Methods of Allocating Taxable Income (Volume IV), League of Nations Document No C.425(b).M.217(b).1933.11A (September 1933). “Fractional” (formulary) apportionment was raised as an alternative to the separate accounting approach. However, the Committee rejected this approach for practical reasons: see, JM Weiner, “Using the Experience in the US States to Evaluate Issues in Implementing Formula Apportionment at the International Level” [1999] OTA Paper 83, 4-5.
[110] For a complete history of the arm’s length standard, see Langbein, above n 96. Langbein, however, suggests that the arm’s length standard is not the international norm and criticises the Carroll report.
[111] OECD, 1979 Report, above n 28.
[112] This section was based on s 28 of the Income Tax Assessment Act 1922 (Cth).
[113] Taxation Ruling TR 92/11 income tax: application of the Div 13 transfer pricing provisions to loan arrangements and credit balances; Taxation Ruling TR 95/23 income tax: transfer pricing – procedures for bilateral and unilateral advance pricing arrangements; Taxation Ruling TR 94/14 income tax: application of Div 13 of Pt III (international profit shifting) – some basic concepts underlying the operation of Div 13 and some circumstances in which s 136AD will be applied; Taxation Ruling TR 97/20 income tax: arm’s length transfer pricing methodologies for international dealings; Taxation Ruling TR 98/11 income tax: documentation and practical issues associated with setting and reviewing transfer pricing in international dealings; Taxation Ruling TR 98/16 income tax: international transfer pricing – penalty tax guidelines; Taxation Ruling TR 1999/1 income tax: international transfer pricing for intra-group services; TR 1999/8 income tax: international transfer pricing: the effects of determinations made under Div 13 of Pt III, including consequential adjustments under s 136AF; Taxation Ruling TR 2000/16 income tax: international transfer pricing – transfer pricing and profit reallocation adjustments, relief from double taxation and the mutual agreement procedure; and Taxation Ruling TR 2001/11 income tax: international transfer pricing – operation of Australia’s permanent establishment attribution rules.
[114] OECD, Model Tax Convention, above n 62.
[115] OECD, 1995 Guidelines, above n 29, 1.6.
[116] OECD, 1998 Draft, above n 2, 35(138): quoting from para 168 of the 1995 Guidelines.
[117] Wickham and Kerester, above n 24, 406.
[118] Li, above n 100.
[119] OECD, Model Tax Convention, above n 62, art 7(4).
[120] OECD, 2001 Draft, above n 4, 40(179).
[121] Ibid.
[122] Kauder, above n 105, 1151.
[123] Wickham and Kerester, above n 24, 409.
[124] Ibid.
[125] Ibid.
[126] FM Horner, “International Cooperation and Understanding: What’s New about the OECDs Transfer Pricing Guidelines” (1996) 50 University of Miami Law Review 577.
[127] BJ Arnold and TE McDonnell, “The Allocation of Income and Expenses Among Countries: Report on the Invitational Conference on Transfer Pricing” (1993) 10 Australian Tax Forum 545, 553.
[128] RJ Vann, “A Model Tax Treaty for the Asian-Pacific Region? (Part I)” (1991) 45 Bulletin for International Fiscal Documentation 99, 105.
[129] Ibid.
[130] Shaddick, above n 64.
[131] Hay et al, above n 1, 253.
[132] Weiner, above n 109, 3.
[133] OECD, 1995 Guidelines, above n 29, II-2 (2.6); examples of the method appear in paras 2.10-2.13; also Taxation Ruling TR 94/14, paras 88-93 and 353-358. This definition is adopted by Taxation Ruling TR 97/20, para 3.10.
[134] Taxation Ruling TR 97/20, para 3.20 (1995 OECD Report Glossary). Examples of the method are found in the 1995 OECD Report, paras 2.29-2.31 (also Taxation Ruling TR 94/14, paras 94, 95 and 359-362).
[135] Taxation Ruling TR 97/20, para 3.21 (1995 OECD Report Glossary).
[136] OECD, 1984 Report, above n 6, 53(36-37).
[137] OECD, 2001 Draft, above n 4, 62(105) updated by OECD, 2003 Pt II Draft, above n 5, 32(145).
[138] Bird, above n 16, 334.
[139] Ibid. Richard Bird uses the example of technology and management services, which may be applied to a division of a firm without detracting from their value elsewhere.
[140] Hamaekers, above n 95, 38; and Green, above n 17, 45. Robert Green describes this rationale as “neutrality between affiliated and unaffiliated firms”. See also CH Berry, DF Bradford and JR Hines, “Arm’s-Length Pricing: Some Economic Perspectives” (1992) 54 Tax Notes 731, 732.
[141] OECD, 1995 Guidelines, above n 29, 1.7.
[142] Hay et al, above n 1, 253.
[143] OECD, 1995 Guidelines, above n 29, 1.7.
[144] Ibid.
[145] Berry et al, above n 140.
[146] B Williams, “Positive Theories of Multi-national Banking: Eclectic Theory Versus Internalisation Theory” (1997) 11(1) Journal of Economic Surveys 71, 74.
[147] Green, above n 17, 46.
[148]Lester, above n 21.
[149] Hamaekers, above n 95.
[150] Williams, above n 146, 71.
[151] M Casson, The Organization of International Business (1995) 22.
[152] Langbein, above n 10, 1392.
[153] Ibid.
[154] Li, above n 100, 832.
[155] BE Lebowitz, “Transfer Pricing and the End of International Taxation” (1999) 19 Tax Notes International 1201, 1202.
[156] Tanzi, above n 12, 1278.
[157] Casson, above n 151, 22; KR Cho, “Determinants of Multi-National Banks” (1986) 26(1) Management International Review 10, 13; and Williams, above n 146.
[158] Casson, above n 151, 22.
[159] Ibid.
[160] Williams, above n 146.
[161] SI Langbein, “A Modified Fractional Apportionment Proposal for Tax Transfer Pricing” (1992) 54 Tax Notes 719, 724.
[162] JR Hellerstein, “Federal Income Taxation of Multinationals: Replacement of Separate Accounting with Formulary Apportionment” (1993) 60 Tax Notes 1131, 1136; and RJ Peroni, “Back to the Future: A Path to Progressive Reform of the US International Income Tax Rules” (1997) 51 University of Miami Law Review 975, n 77.
[163] Hellerstein, above n 163.
[164] JS Wilkie, “A Canadian Perspective on Related-Party Transactions” in 2000 World Tax Conference Report 12:1, 12:4.
[165] Ibid.
[166] Dolan, above n 90.
[167] Hay et al, above n 1.
[168] Langbein, above n 96, 654.
[169] Lester, above n 21.
[170] Plambeck, above n 11.
[171] Langbein, above n 10, 1402.
[172] Mazerov, above n 17, 30.
[173] OECD, 1998 Draft, above n 2, 42(167).
[174] Ibid quoting from the 1995 Guidelines, above n 29, para 1.9.
[175] OECD, 1998 Draft, above n 2, 42(167).
[176] Ibid 42(171).
[177]Ibid referring to para 3.5 of the 1995 guidelines, above n 29.
[178] Newlon, above n 8, 216.
[179] Ibid.
[180] J Li, above n 100, 832.
[181] Green, above n 17, 46.
[182] Ibid.
[183] Li, above n 100, 825.
[184] Wilkie, above n 164, 12:3.
[185] Green, above n 17, 46.
[186] Hamaekers, above n 95; and Green, above n 17, 46.
[187] Charles E McLure Jr argues that this provides a clear theoretical foundation for the arm’s length standard. CE McLure Jr, “Replacing Separate Entity Accounting and the Arm’s Length Principle with Formulary Apportionment” (2002) 56 Bulletin for International Fiscal Documentation 586, 587.
[188] Commission of the European Communities, Commission Staff Working Paper: Company Taxation in the Internal Market (2001) 407.
[189] McLure, above n 187.
[190] Cuttler, above n 98.
[191] OECD, 1995 Guidelines, above n 29, 1.13.
[192] Li, above n 100, 831.
[193] Green, above n 17, 47.
[194] Langbein, above n 10, 1408.
[195] Mazerov, above n 17, 30.
[196] G Hoffman, “Taxation of Global Trading of Financial Instruments: A Discussion Draft by the OECD” (1997) 8 Journal of Banking and Finance Law and Practice 130, 131, referring to the 1998 Draft, above n 2.
[197] Ibid.
[198] Green, above n 17, 47.
[199] DM Ring, “Exploring the Challenges of Electronic Commerce Taxation Through the Experience of Financial Instruments” (1996) 51 New York University Tax Review 663, 663; and C Sijbren, Tax Policy in the European Union: A Review of Issues and Options (2001).
[200] Wickham and Kerester, above n 24, 403.
[201] Ibid.
[202] Ibid.
[203] Berry et al, above n 140, 733.
[204] Ibid.
[205] RM Bird and DJS Brean, “The Interjurisdictional Allocation of Income and the Unitary Taxation Debate” (1986) 34 Canadian Tax Journal 1337, 1383.
[206] CT Horngren, G Foster and SM Datar, Cost Accounting – A Managerial Emphasis (8th ed, 1994) 864.
[207] Ibid.
[208] Horngren et al, above n 206, 875.
[209] Taxation Ruling TR 97/20, para 2.5. While this question, is taken from an Australian Taxation Ruling it will be asked by all taxing authorities adopting the arm’s length approach.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/JlATax/2004/7.html
