University of Western Sydney Law Review
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of Western Sydney Law Review |
|
Michelle Markham[*]
Nuclear physics is much easier than tax law. It's rational and always works the same way. (Jerold Rochwald)
In the 21st century, the number one international tax issue of interest to multinational enterprises (MNEs) is undoubtedly transfer pricing.[1] The reason for this is that as global trade increases, so too does the uncertainty of the tax treatment of inter-affiliate transactions across national boundaries and the spectre of double taxation. The Australian Deputy Commissioner of Taxation has outlined the concept of transfer pricing as follows: 'Broadly, transfer pricing relates to the setting of prices by multinationals for the goods and services that they supply to related parties. It also covers the structuring of transactions and financial relationships, and how innovation happens and is rewarded.'[2]
The OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (the OECD Guidelines) make it clear that the concept of transfer pricing should not be confused with that of tax fraud, or of tax avoidance, even though transfer pricing transactions may be utilised for such purposes.[3] A number of international tax specialists have also stressed that incorrect references to transfer pricing as 'income-shifting' obfuscate a clear analysis.[4]
A paper prepared by the United Nations Secretariat in 2001 has described transfer pricing as 'probably the most important tax issue in the world'.[5] Referring to the fact that over 60 percent of international trade is carried out within MNEs,[6] the paper also refers to the impact which intangible property has had on this trade.[7] Both US and Australian tax practitioners have acknowledged that some of the most difficult transfer pricing issues have always been in the area of intangible property.[8] The tax treatment of intangible assets therefore warrants particular attention in the transfer pricing context.
In the United States, for purposes of section 482 of the final regulations, the term 'intangible' refers to any item included in one of six broad categories specified in the regulations, provided the item has substantial value independent of the services of any individual.[9] These categories of intangible property include:
| • | Patents, inventions, formulas, processes, designs, patterns or know-how; |
| • | Copyrights and literary, musical, or artistic compositions; |
| • | Trademarks, trade names, or brand names; |
| • | Franchises, licenses, or contracts; |
| • | Methods, programs, systems, procedures, campaigns, surveys, studies, forecasts, estimates, customer lists, or technical data; and |
| • | Any other similar item that derives its value from its intellectual content rather than its physical attributes. |
For the purposes of Chapter VI of the OECD Guidelines, 'intangible property' includes rights to use industrial assets, such as patents, trademarks, trade names, designs or models, literary and artistic property rights, and intellectual property such as know-how and trade secrets. [10] Australian Taxation Rulings generally refer to the OECD definitions of intangible assets.[11]
The internationally accepted arm's length principle demands that MNEs charge transfer prices in their controlled transactions that are consistent with the prices that would have been charged for the same uncontrolled transaction taking place between unrelated, independent enterprises under the same circumstances. To this end, transfer pricing methodologies are utilised by MNEs in order to establish an arm's length outcome. The transfer pricing methodology adopted by an MNE consequently constitutes a pivotal component of a determination of the arm's length consideration attributable to a transaction involving the intragroup transfer of intangible property.
As identical transactions between unrelated enterprises are rare, transfer pricing methodologies tend to focus on comparable rather than identical transactions. So-called 'transactional' methodologies have been espoused by revenue authorities as the most direct way of establishing whether arm's length conditions exist between associated enterprises. These methodologies are reliant on finding either identical transactions, or, where these are not available, similar comparable transactions. There has been a growing realisation that where intangible assets are concerned, there are grave problems in determining even a comparative analysis. This paper will explore the consequent shifting focus to newer, non-traditional methodologies, especially profit-split methodologies, in the US and Australia, and to a lesser extent by the OECD. These methodologies tend to rely in whole or in part on internal data rather than on data derived from comparable uncontrolled transactions.
The US final section 482 regulations[12] , the OECD Guidelines and the Australian transfer pricing rulings all permit MNEs to select an appropriate transfer pricing methodology for their inter-affiliate transfers of intangible assets. Different methodologies may be selected under different circumstances. It is necessary to assess a number of variables in determining the correct methodology for a particular transaction. These variables may change over time, necessitating a reconsideration of the methodology to be utilised.
Although there are marked similarities in the US, OECD and Australian approaches to choosing a transfer pricing methodology for intangible property transfers, there are also some important differences. While the official position of the US is that its final transfer pricing regulations are consistent with the OECD Guidelines, some OECD member countries disagree. This has had the unfortunate result that MNEs risk antagonising certain revenue authorities if they undertake what appears to be a US transfer pricing approach. Multinational taxpayers are therefore compelled 'to account for multiple and sometime[s] disparate rules when setting, documenting, and defending cross-border transfer prices.'[13]
The ‘Best Method Rule’[14] has been adopted by the US final regulations for all intercompany transactions. In contrast to the rigid hierarchical system that had previously applied, the final regulations now require a taxpayer to select the pricing method that provides the most reliable measure of an arm’s length result, relative to the reliability of other applicable methods, bearing in mind the facts and circumstances surrounding the particular transaction under review. Taxpayers are thus given more freedom to choose the methodology most appropriate to their specific transactions.
The transfer pricing regulations allow companies to examine and test different methods to determine which method provides the most reliable result. There is no strict priority of methods under the regulations, and no method is considered invariably more reliable than another.[15] The greater flexibility of the final regulations is also indicated by the fact that the taxpayer is allowed to establish an arm’s length amount through the use of a method (properly documented) other than those specified in the regulations.
There appears to be an underlying assumption that taxpayers will test a variety of methods before selecting the one that best suits their particular transaction. The regulations provide that in choosing a particular methodology it is not necessary to establish the inapplicability of another method. However, if another method is subsequently shown to produce a more reliable measure of an arm's length result, such other method must be used.[16]
The US regulations designate certain allowable transfer pricing methods for a determination of taxable income in connection with a transfer of intangible property.[17] Unspecified methods are also allowed, as long as they provide the most reliable measure of an arm's length result under the principles of the best method rule.[18]
MNEs should select the method that provides the greatest degree of comparability, which will be determined not only by the accuracy of the underlying data, but also by the extent of the information - eg the period of time on record. A functional analysis will determine whether the economically significant activities of the controlled and uncontrolled transactions are sufficiently similar.[19]
From a practical point of view, it may be difficult to choose the best method for an interaffiliate transfer of intangible property. Greater flexibility is generally acknowledged to be a positive outcome of the final regulations, but there is a warning here that a corollary may be less certainty in reaching a satisfactory transfer price, and a greater burden on taxpayers to meet nebulous requirements. There is therefore a need for more adequate advice and direction to be available to taxpayers.
The OECD Guidelines do not refer to a 'Best Method' rule, but rather recommend that taxpayers select a transfer pricing methodology that is able to provide the best estimation of an arm’s length price, taking into account the facts and circumstances of the case, the mix of evidence available and the relative reliability of the various methods under consideration.[20] In contrast to the US approach, the OECD Guidelines state quite explicitly that the arm's length principle does not require the application of more than one method,[21] as this may give rise to a significant administrative burden for taxpayers. The OECD appears to be aware of the fact that collecting external data on a number of different methods is not only onerous for taxpayers, but also for tax administrations, and therefore requires neither party to perform analyses under more than one method.
The Guidelines also demonstrate flexibility regarding the use of transfer pricing methods, in that MNE groups are given the freedom to apply methods other than those described in the OECD Report to establish prices, provided those prices satisfy the arm’s length principle.[22] Unlike the US regulations, the Guidelines do not designate certain specific methods as applicable to transfers of intangible assets.
They even provide for the use of a combination of methodologies in difficult cases where no one approach appears conclusive.[23]
Under the Guidelines, comparisons with uncontrolled transactions can only be relevant where the economically relevant characteristics of the situations undergoing comparison are sufficiently comparable.[24] While the Guidelines make it clear that taxpayers should base their controlled transfer prices on a sound analysis and should document the basis on which such prices are set, 'the amount of effort called for by the Guidelines is markedly less than what U.S. regulations would require to avoid a tax penalty on a large transfer pricing adjustment.'[25]
The Australian Taxation Office (ATO) seeks to adopt the most appropriate or best suited transfer pricing methodology, bearing in mind all the circumstances of the particular case.[26]
In Australia, the methodology must, on an objective basis, produce the most accurate arm's length calculation. No preference for a particular methodology is prescribed - a novel methodology may be utilised, providing it achieves an arm's length result.[27] The ATO has stated that in Australia it is possible to go beyond the OECD Guidelines, by resolving transfer pricing issues using indirect methods, while still remaining true to their fundamental principle of evaluating transactions according to what truly independent enterprises acting independently would probably have done in the taxpayer's position.[28]
Division 13 of the Australian Income Tax Assessment Act 1936 (ITAA) does not refer to the methodologies to be utilised in calculating an arm's length price - these are only referred to through public taxation rulings. The Commissioner generally seeks to apply methods given international endorsement, adopting the method best suited to the circumstances of the particular case.[29] It is recommended that a taxpayer document the reasons for their choice of method at the time of its adoption.
Traditional transactional transfer pricing methodologies rely on the utilisation of data reflecting comparable uncontrolled transactions. These methods tend to be favoured by revenue authorities in OECD countries, including the United States and Australia, as they are considered to be the most reliable way of discerning an arm's length price. Their historic record of acceptability has led them to be referred to as 'traditional' pricing methods. By contrast, financial or profit-based methods tend to rely on the gross margins of comparable companies, rather than on actual transactions.
In recent years, more and more attention has been focused on profit-based methods. In practice, where inter-affiliate transactions involve intangibles[30] , MNEs experience great difficulty in finding reliable data to analyse comparability, especially where nonroutine intangibles are involved. Consequently, the US and Australian tax authorities and the OECD have all come to accept the need for profit methods, with varying degrees of enthusiasm. The US and Australia initially led the way in promoting the increased use of transactional profit methods in determining the arm's length nature of cross-border transactions between affiliates.[31]
The OECD has in the past been more reluctant to embrace profit-based methods, stating that ''so-called 'comparable profits methods'' [32] were acceptable only to the extent that they were consistent with the Guidelines. However, it should be noted that at the first International Tax Review Transfer Pricing Forum, which took place in Amsterdam in September 2001, the OECD concluded that the profit-split method is becoming more widespread.[33]
In fact the OECD is currently seeking comments on the application of transactional profit methods as part of the Working Party's monitoring of the OECD Guidelines.[34] The head of the OECD's TTP Division (tax treaties, transfer pricing and financial transactions) has indicated that the Guidelines will be revised to include a more technical application basis for profit methods, as it has become the general trend to use those methods, there being more publicly available data at a prof it-margin level.[35]
Profit-based methods can be divided into two categories of methodologies. Firstly, there are profit comparison methods, which focus on a comparison of profits at the net profit level. Secondly, there are profit split methods, which may be referred to as apportionment methods. Various profit split methods are endorsed by the United States, by the OECD and by Australia. This article will concentrate on the profit split methods, as these are currently the subject of much attention and controversy, as evidenced by the recent OECD interest mentioned above. Moreover, the OECD noted at the first International Tax Review Transfer Pricing Forum which took place in September 2001 that the profit-split method was becoming more widespread.[36]
Profit split methods may be viewed as being at the outer limits of acceptable transfer pricing methodologies, because, as mentioned above, they tend to rely in whole or in part on internal data rather than on data derived from comparable uncontrolled transactions. They are thus seen as being a less reliable measure of an arm's length result than transactional methodolgies, which rely on data reflecting comparable uncontrolled transactions. Nevertheless, they are becoming increasingly accepted in a global business environment characterised by a proliferation of intangible asset transactions. Revenue authorities are hesitant to condone methodologies that may be considered to be incorporating an element of apportionment, but the fact remains that in certain inter-affiliate transactions, especially those involving intangibles, it is impossible to identify sufficiently similar uncontrolled comparables that form the basis for the other specified methods.
In the US, profit split methods can be applied to controlled transactions where tangible or intangible property is transferred within an MNE group. These methods involve a comparison of the relative economic contributions made by each taxpayer to the venture, and a division of the returns from the venture according to the relative value of each party's contribution calculated on an arm's length basis...
Although the 1993 regulations originally proposed four profit split methods, the 1994 final regulations reduced these to two, namely the comparable profit split and the residual profit split.[37] The inclusion of these two profit split methods has been attributed to 'the strong urging of high-technology industries'.[38]
This method depends on discovering the profit on comparable transactions between two unrelated enterprises. The regulations explain that a comparable profit split is derived by combining the operating profit of uncontrolled taxpayers whose transactions and activities are similar to those of the controlled taxpayers in the relevant business activity.[39] Each uncontrolled taxpayer's percentage of the combined operating profit or loss then serves to allocate the combined operating profit or loss of the relevant business activity. In other words, the uncontrolled split is used as a benchmark for the profit split between the controlled parties.
The Comparable Profit Split Method (CPSM) emphasises the quality of the third-party evidence, for example a similar division of assets and combined return on assets must exist between the uncontrolled independent parties and the controlled affiliates. The similarity of the contractual terms of the controlled and uncontrolled parties is particularly relevant to the CPSM, as such contractual terms tend to determine the allocation of functions and risks. The regulations provide that if the combined operating profit of the uncontrolled parties varies significantly from that of the controlled parties (as a percentage of the combined assets), this method may not be used.[40]
The reliability of the data and assumptions concerning each party to the transactions in question is also an important consideration under the CPSM. If the data and assumptions concerning one of the parties are significantly more reliable than the data and assumptions concerning the other party, the regulations suggest that it may be preferable to employ another method which focuses exclusively on the results of the first entity.[41]
Opinions appear to be divided as to whether the CPSM is readily applicable to intangible asset transactions. A number of commentators have stated their belief that in practice, the difficulties involved in obtaining the extensive data on uncontrolled taxpayers prevent most taxpayers from using this method.[42] Others, however, are of the opinion that the CPSM is ideally suited to intangible asset transactions, and that 'the CPSM is the best method for assessing the value of the intangible property embodied in a product with extraordinary profit potential'.[43]
According to the US regulations, the Residual Profit Split Method (RPSM) attempts to estimate an arm’s length return for each party in a controlled group by a comparison of the relative economic contribution of each party to the success of the business as a whole, and dividing the worldwide profit between them on the basis of the relative value of each contribution.
Here, the combined operating profit or loss from the relevant business activity is allocated between controlled taxpayers in a two-step process:[44]
| 1. | Operating income is allocated to each party in a manner that will yield a market return to them for routine contributions to the business activity. |
| 2. | Residual profit that is attributable to the controlled group’s valuable intangibles is apportioned. |
The first step involves using other methods to estimate market returns to routine functions and then to allocate them to the parties that performed them. Routine contributions may be described as contributions that are of the same or similar kind to those made by uncontrolled taxpayers taking part in similar business activities, for which it is possible to identify market returns.[45] These ordinarily include contributions of tangible property, services and intangibles that are generally owned by uncontrolled taxpayers engaged in similar business activities. Thus comparables still need to be found here.
The residual profit that remains after income has been allocated to routine contributions is then allocated between the parties on the assumption that this residual amount is attributable to nonroutine intangible property contributed to the activity by the controlled taxpayers. The second step therefore refers to the apportionment of residual profit, which remains after the income allocation where valuable nonroutine intangibles are owned by the controlled group but similar items are not owned by the uncontrolled taxpayers utilised to determine market returns in the first step. This second step does not involve the use of comparables.
Three methods of apportionment of such residual profit are provided in the regulations.[46] Firstly, the relative value of such intangible property may be measured by external market benchmarks that reflect the fair market value of such intangible property. The IRS acknowledges that such fair market value may not usually be readily ascertainable, and that consequently other measures of the relative values of intangible property can be used.[47]
Secondly, the relative values of intangible contributions may be estimated by the capitalised cost of developing the property, as well as the cost of developing all related improvements and updates, less an appropriate amount of amortisation, based on the useful life of each intangible. There has been some debate as to whether the IRS should rely on the market capitalisation of a company in determining the value of its intangible property, but it is still included in the regulations as a viable valuation method.[48] This capital cost method is in fact used in the only example given on computation of a profit split.[49] Finally, if the research and development expenditures of the parties are relatively constant and the useful life of the intangible asset of all parties is approximately the same, actual expenses may be used to determine the relative value of each item of intangible contributions.
The reliance, at least in part, on internal data rather than comparables means that this method is not regarded as being as reliable a measure of the arm's length result as are other methods. The reliability of the results may be affected by the fact that the costs of developing the intangible may not be related to its market value. The allocation of indirect costs between the relevant business activity and the controlled taxpayer's other activities, which may be required for the calculation of the capitalised costs of development, may further affect reliability. The calculation of costs may depend on assumptions being made regarding the useful life of the intangible property, introducing a further element of instability to the equation.[50]
The regulations point out that since the second step is not usually based on a market benchmark, the reliability of this method will tend to be reduced as the amount of the residual profit allocated pursuant to this step increases.[51] However, the regulations also indicate that the reliability of the analysis under this method may be enhanced by the fact that all parties to the controlled transaction are evaluated under the profit split. Other factors have also been cited as favouring the application of a residual profit split method for transactions involving high value intangibles. These include the use of a ‘contribution’ basis avoiding double taxation by deterring the excessive allocation of income to one jurisdiction, the administrative simplicity of applying this method from the taxpayer's point of view, and the ease of evaluation by the revenue authorities.[52]
The OECD sees what it refers to as 'transactional' profit methods as methods of last resort. Such methods are only to be applied in what are supposedly 'exceptional' cases where there are difficulties in applying the traditional transaction methods, and only where the safeguards set out by the OECD are observed.[53] Such methods need to comply with Article 9 of the OECD Model Tax Convention, especially regarding compatibility. To achieve an arm's length analysis using a profit method, profits arising out of a particular controlled transaction need to be compared with the profits arising out of comparable transactions between uncontrolled taxpayers.
The OECD Guidelines therefore outline a profit split methodology similar to that adopted in the US, involving the determination of the division of profits from controlled transactions in accordance with how profits would have been divided between independent enterprises.[54]
This method first identifies the profit to be split from the controlled transactions under scrutiny. Those profits are then split between the associated enterprises on an arm's length basis. The combined profit may be composed of the total profits arising out of the transactions, or, alternatively, a residual profit which cannot be clearly assigned to one enterprise, for example the profit arising out of 'high-value, sometimes unique, intangibles.'[55] The contribution of each associated enterprise is assessed according to a functional analysis. Thus, as in the US version, the OECD utilises a contribution and a residual analysis. The difference lies in the fact that while the division of profits between comparable independent parties is only one factor to be evaluated under the Guidelines’ contribution analysis, it is the essential factor under the US regulations.
The Guidelines list the strengths of this method, including the fact that it generally does not rely on closely comparable transactions. This offers flexibility by taking into account specific facts and circumstances of the associated enterprises that are not present in independent enterprises.[56] Another strength is that it is less likely that one party will be allocated an extreme and improbable profit result, as both parties to the transaction are evaluated. The particular importance of this factor when analysing the contributions by the parties in respect of intangible property is stressed.[57]
Weaknesses of using a profit split method are stated as including the tenuous connection of external market data with the controlled transactions under consideration, resulting in a certain amount of subjectivity.[58] Furthermore, associated enterprises and tax administrations may have difficulty accessing information from foreign affiliates. It may be difficult to measure combined revenue and make adjustments in accounting practices and currencies, and to identify appropriate operating expenses and make a correct allocation of them.[59] As far as comparable uncontrolled transactions are concerned, a disadvantage of this method is that, as a general rule, independent enterprises do not use the profit split method to determine their transfer pricing (an exception here might be joint venture undertakings).
According to the OECD, a profit split method would seek to mirror the division of profits that an independent enterprise would expect to allocate in a joint venture relationship. Such expectations would be have to be based on projected profits rather than actual profits, as an independent enterprise would not be able to assess in advance what the profits of the business activity are going to be at the time the conditions are established.[60]
Two approaches for estimating the division of profits are discussed in the Guidelines, although they are not necessarily exhaustive or mutually exclusive. They are the contribution analysis approach and the residual analysis approach.
Here the profits from the controlled transactions would be divided between the associated enterprises according to the relative value of the functions performed by each associated enterprise. If possible, external data is used to provide an independent comparability measure.[61] The operating profit is generally used to determine the profit split under the contribution analysis, thus taking into account both the income and expenses of the MNE with respect to the relevant associated enterprise. In certain circumstances a split of gross profit may be utilised, followed by a deduction of the expenses incurred in or attributable to each relevant enterprise.[62]
There may be difficulties in accurately determining such a contribution analysis, which may vary from case to case. A percentage of the profits may be allocated to an associated enterprise on the basis of a relative comparison of the nature and degree of each party's contribution, supplemented by relevant external data.[63]
The OECD suggests a two-step approach, similar to that adopted by the US. The first step allocates each associated enterprise with a basic return relevant to the transactions it has engaged in, referring to market returns received by independent enterprises for similar transactions. This first step would generally not account for any unique or high value intangible assets held by the participants. In the second step, any residual profit or loss would be allocated between the associated enterprises, taking into account any contributions of intangible property as well as the relative bargaining positions of the enterprises. This would involve an analysis of the facts and circumstances that might indicate how this residual would have been divided between independent enterprises.[64] It is here that practical difficulties arise in discovering suitable comparables involving similar unique, high-value intangible assets. While the US residual profit split method may be deemed less reliable at this second stage, as it relies on internal rather than external data, it is at the same time of more practical application. The Guidelines place strong emphasis on the relative bargaining positions of the associated enterprises in this second step, along with a comparative analysis of such positions. Yet they also acknowledge the difficulties involved in finding sufficiently comparable transactions and the fact that even where such comparable transactions exist, adequate information on these transactions may not be readily available either to MNEs or to tax administrations.[65]
Although the OECD has shown great reluctance to accept all variations of the profit split method, the Guidelines now explicitly state that the profit split method may be acceptable in the case of highly valuable intangibles for which no comparable uncontrolled transactions can be found.[66] (It is in fact generally acknowledged that transactions between unrelated entities involving a high-profit intangible are extremely rare - indeed, the Tax Executives Institute has described them as 'almost non-existent', with the result that profit split methods may often be the only reliable method in such cases.[67] ) The Guidelines do, however, still anticipate practical problems in applying a profit split method, possibly because of their insistence on using comparable data in its application.
The ATO has accepted the need to use profit methods in situations where it might not be possible or practicable to use traditional methods, i.e. where comparable transactions are not readily available. Such situations tend to arise where high value, unique or 'out-of-theordinary' intangibles are involved. The ATO accepts that 'Global industries are based on highly sophisticated technology, involve valuable production, distribution or marketing intangibles and are generally vertically and horizontally integrated.'[68] Unique structures and products are the norm in such complex global networks, making it difficult to provide sufficient reliable data to analyse comparability. In such situations, profit methods may prove to be a more pragmatic way of ascertaining the arm's length nature of the cross-border dealings between associated enterprises.
In outlining the profit split methodologies applicable in Australia, frequent reference is made to the OECD Guidelines. In Australia, as with the OECD, the profit in question may involve the total profit from the transactions or the residual profit that cannot be readily assigned to one of the enterprises, such as the profit arising out of high value or even unique intangibles.[69]
The profit split method is applied to all the relevant associated enterprises, and there may, in certain situations, be a need for an aggregation of dealings to be used. It is necessary to determine whether the profit split is to be undertaken on a particular product line, an aggregation of products or on a whole of entity basis.[70] A profit split also requires a consolidation of accounts, i.e. the accounts of the associated enterprises need to be placed on an equal footing in relation to accounting practice and currency, and then consolidated. After the split has been determined, separate accounts can be issued taking into account national accounting requirements and formalities.
The ATO accepts that profits may be divided according to a contribution analysis or according to a residual analysis, along the lines accepted by the OECD Guidelines. Although the ATO does not specifically favour a particular method, practitioners have suggested that the residual profit split method is more readily accepted.[71] This is because this latter methodology ensures that even where profits are low, the Australian entity will still be allocated a stable return based on its routine functions. The allocation of profit to the Australian entity under a contribution analysis is more volatile, and hence is likely to be less attractive to the ATO.[72]
However, the ATO has also stated that it would not rule out any profit split methodology which results in an answer that approximates an arm's length outcome.[73] There is thus a recognition of the fact that it may be necessary to develop an approach that is flexible enough to recognise the different contributions by the various associated enterprises over economic and product life cycles. It is even willing to consider the use of a formula to accurately split profits between the associated enterprises.[74] Ideally, the weightings used in the formula should be based on some form of external market data, looking at what type of formulaic allocation would have been made by an independent enterprise in the same situation. Such a profit split should take into account the differences in functions, assets and risks occurring in different cases.
A specific data-related problem encountered with profit splits is that many foreign-owned subsidiaries find it difficult to persuade the parent company to release the necessary information. Looking at the Australian situation, there may be a number of reasons for this:
| • | Many of the large MNEs which have subsidiaries in Australia generate only a very small proportion of their worldwide revenues in Australia (perhaps 1-2 percent) and the work required to satisfy the ATO is disproportionately high. Anecdotal evidence suggests that these companies would prefer to pay an adjustment to the ATO (within reasonable limits) rather than devote resources to developing a transfer pricing policy purely for Australia. |
| • | There is a likelihood that any information that the MNE discloses to the ATO might be shared with other tax authorities (under exchange of information agreements). This might be especially true between tax authorities in various Pacific countries and could lead to unwanted attention from these other tax authorities.[75] |
The administrative cost of applying a profit split methodology and the possible exposure to 'unwanted attention' through what is seen by MNEs to be a breach of taxpayer confidentiality are valid objections and need to be dealt with by OECD countries such as the US and Australia.
Attempting to use traditional transaction methods to establish a price for intangible property transfers always leads back to the same problem: a lack of comparability.
As comparables are almost impossible to find where high value, nonroutine intangibles are involved, in practice it is the profit-based methodologies that are applied, especially methodologies which rely on a profit split. A major advantage of profit split methods is that their focus lies on allocating actual profits earned by MNEs in cross-border transactions, rather than concentrating on a determination of what the correct or arm's length price should be. In other words, the focus is on profits actually earned, rather than hypothetical profits. This is a particularly valid focus where nonroutine intangibles are involved, and where there are no comparable uncontrolled transactions, as:
Even aside from the problem of determining the 'right' price for each transaction, the focus on the propriety of individual transactions, as opposed to the proper amount of taxable income, multiplies substantially the number of questions that must be answered and, correspondingly, the resources required to determine the ultimate result.[76]
With comparables becoming increasingly difficult to find, the focus will be placed on profit-based methods, and particularly on profit split methods. The only alternative would appear to be the development of a global formulary apportionment method, or methods, to which there has long been resistance. Practitioners in the US have therefore come to the conclusion that 'The use, and types, of profit splits will likely increase.'[77]
The OECD Guidelines favour the traditional transaction methods, but also acknowledge that 'the complexities of real life business situations' may, in practice, inhibit their application.[78] After adopting this seemingly pragmatic approach and referring to the practical difficulties of applying such methods, the Guidelines then state that other methods may be used in those 'exceptional' situations where data is either unavailable or of insufficient quality to rely solely or at all on these traditional transaction methods.[79] Any investigation of the practical application of these traditional methods to intangible property transactions reveals that a lack of comparable transactions is the rule, rather than the exception.
An Australian Taxation Ruling has stated that the most appropriate method will be the one that produces the highest practicable degree of comparability. However, it is recognised that there will be unique situations and cases involving valuable intangibles where it is simply not practicable to apply methods based on a high degree of direct comparability.[80]
The specialised nature of intangible property necessarily gives rise to comparability problems. A lack of comparable data means that greater emphasis is placed on profit based methodologies. The ATO outlines four reasons for this dearth of information. Firstly, comparable data may be scarce due to the unique character of the intangible, resulting in difficult valuation questions. Secondly, in order for the profitability of highly valuable intangibles to be maximised, there is generally a need for them to remain within the control of the MNE's group. Thirdly, certain intangibles can only be protected by keeping their attributes secret within the MNE's group. Fourthly, the intangible may have been developed solely by the efforts of an MNE and for its own purposes, as is the case with some marketing intangibles.[81]
In addition to the global dearth of information regarding international transactions dealing with intangible assets, Australia faces the added problem of a much more limited database for assessing comparability than, for example, the United States. The ATO has recognised that, as far as Australia is concerned, this means that indirect measures to assist in achieving results that accord with arm's length outcomes may need used on a more frequent basis.[82] It has gone so far as to state that as dealings in intangibles are not well documented, taxpayers may find only very limited pricing data in this area. This in turn may affect the level at which comparisons can be made, limiting taxpayers to measures of profit performance.[83]
Conversely, the main disadvantage of the profit split method has been argued to be that it can be viewed as potentially arbitrary, as it does not rely to a great extent on independent, comparable data. The issue of the absence of external market criteria is recognised by the ATO, but it has concluded that: ' This view could be overstated in cases where an economic functional analysis is used and due regard is had to comparable rates of return on assets, functions and risks in comparable situations in comparable markets.'[84]
Ultimately, in Australia, in situations where there may be no comparables for a combination of transactions (such as transfers of tangible and intangible goods and services), 'profit methods may be a more reliable way to set or review the transfer pricing used in the dealings between the associated enterprises, or to check findings made using traditional methods if there is doubt about the reliability of the data used or the outcome produced.'[85]
Practitioners have confirmed the various barriers to finding suitable comparables for the application of traditional transactional methodologies.[86] A primary obstacle is that the structure of the world economy currently features an increasing level of globalisation. A corollary of this situation is that there are fewer unrelated party transactions or functionally comparable independent companies. MNEs are achieving efficiencies by integrating management and control mechanisms across international and fiscal borders, resulting in group transactions which are distinctively different to those negotiated between unrelated parties. This factor alone makes it almost impossible to identify good comparables. National levels of disclosure requirements vary, and even in jurisdictions requiring high levels of disclosure the information available is insufficient to undertake a full comparability analysis. Finally, differences in accounting policies have complicated cross-border comparisons.
It therefore comes as no surprise to discover that the US courts and regulations are increasingly using the 'profit split' method to allocate income between members of a controlled group using internal data to divide the income.[87]
In practice, a profit split method is particularly applicable to taxpayers owning valuable or high-profit non-routine intangibles, due to the difficulty in finding the comparable transactions necessary to an application of a transactional method.
The Guidelines recognise the difficulties in finding comparable uncontrolled transactions where highly valuable intangible property is involved, and conclude that in such cases the profit split method may be used, even though there may be practical problems in its application.[88]
The ATO has adopted a common sense approach with regard to the non-transaction-based methods that have had a mixed reception in the international community. Such methodologies (including profit comparisons, profit splits and even predetermined formula methods) are regarded as 'less direct' ways of applying the arm's length principle, as they look at the rate of return and the process by which profits and expenses are allocated. Nevertheless:
they are also accepted by the ATO as being consistent with the arm's length principle and most appropriate for cases where a more direct comparability on price or profit margin is not possible or practicable. In that sense they are methods of last resort. That is not to say that there needs to be an exhaustive search for direct comparables before these methods can be applied.[89]
In light of the fact that independent comparable transactions are very difficult to find in respect of intangible asset transactions, greater flexibility in choosing a transfer pricing methodology for such transactions is recommended as a practical way of establishing an arm's length range of prices.
From the above, it is evident that the use of profit split methods for inter-affiliate transfers of intangible assets will increase in the years ahead. The proliferation of intragroup intangibles in the global economy, and especially high value and/or unique intangibles, means that only in rare circumstances will it be possible to obtain data on similar technology used by independent parties at a transactional level.
Foreign tax authorities, including Japan's National Tax Administration (NTA) and other OECD countries are now indicating a marked preference for profit splits as the preferred second-best alternative to traditional transactional methods where there is a lack of suitable comparables. The OECD itself appears to be moving towards a greater acceptance of profit methods, and there are significant indications that the current chapter of the Guidelines which deals with the use of profit methods will soon be updated and revised.[90]
In Australia, the ATO have discovered that profit methods are becoming more acceptable in competent authority processes, and that 'Generally, ... the profit methods used in Australia are able to provide some certainty, produce fair results, require less effort in terms of documentation and are reasonably easy to administer for taxpayers and administrators.'[91]
In conclusion, despite historical resistance to the use of profit-split methodologies by revenue authorities in the US and Australia, and by the OECD, there is now an increasing openness to viewing such transfer pricing methodologies as at least one of the ways forward, especially where intangible asset transactions are concerned.
[*] BA LLB LLM (Natal), H.Dip.Tax (Witwatersrand), PhD (Bond), Advocate of the Supreme Court of South Africa. Associate Lecturer, QUT. This article is an extract/adaptation from the forthcoming Kluwer Law International publication The Transfer Pricing of Intangibles, Michelle Markham (author), due for publication in 2005, © Kluwer Law International.
[1] Ernst & Young, Transfer Pricing 2003 Global Survey: Practices, Perceptions and Trends in 22 Countries Plus Tax Authority App roaches in 44 countries 2.
[2] Killaly J, 'Transfer Pricing - Compliance Issues And Insights In the Context Of Global Profit Allocation' (Paper presented at the Transnational Crime Conference convened by the Australian Institute of Criminology in association with the Australian Federal Police and Australian Customs Service and held in Canberra, 9-10 March 2000) 1, 2.
[3] OECD Guidelines 1.2.
[4] 'The term transfer pricing is, however, sometimes used, incorrectly, in a pejorative sense, to mean the shifting of taxable income from a company, belonging to an MNE, located in a high taxing jurisdiction to a company belonging to the same group in a low taxing jurisdiction through incorrect transfer prices in order to reduce the overall tax burden of the group.' Hamaekers H, 'Taxation Vis-à-vis International Relations And The Use Of New Technologies' Topic 2: Transfer Prices At The Beginning Of The XXI Century 1 Inter-American Center of Tax Administrations - CIAT, International Bureau of Fiscal Documentation - IBFD, The Netherlands, September 20-23, 1999, 4. 'The term relates to the system of pricing the transfer of goods, services and intangibles between entities of one multinational enterprise (MNE). Often it is used, incorrectly, in a pejorative sense, to mean a pricing decision by a MNE which shifts income from one member of the group to another.' Pagan J C & Wilkie J S, Transfer Pricing Strategy In A Global Economy (1993) IBFD Publications, Amsterdam, 1.1.
[5] Ad Hoc Group of Experts on International Cooperation in Tax Matters, Transfer Pricing History, State of the Art, Perspectives Tenth meeting, Geneva, 10-14 September 2001 (The present paper was prepared by the United Nations Secretariat) 2.
[6] These MNEs are sometimes referred to as transnational corporations, or TNCs. As far as their impact on the global economy is concerned, in 1997 it was estimated that about 44,000 TNCs, with some 280,000 foreign affiliates were active. In 1995, the total foreign assets of the 100 largest TNCs amounted to $1.7 trillion, their total foreign sales to $2 trillion, and their foreign employment to 5.8 million persons. See: Easson A J, Taxation Of Foreign Direct Investment: An Introduction (1999) London Kluwer International, 3, referring to UNCTAD (1997).
[7] The Ad Hoc Group of Experts, above n 6, 2. The paper here refers to the fact that 'intangibles, developed by group entities, may be concentrated in centres operating for the whole group or specific parts; intangibles, developed by group entities, may be concentrated at certain group members;'
[8] See: Mogle J R, ‘Intercompany Transfer Pricing for Intangible Property' (May 21, 1997) 6 Tax Management Transfer Pricing Special Report No 2. 1, 2. All § 482 references are to § 482 of the US Internal Revenue Code of 1986, as amended. See also: Anderson P, ‘Australia’ (April 1997) International Tax Review 9, 12.
[9] Treasury Regs. § 1.482-4(b).
[10] OECD Guidelines 6.2.
[11] For example, see TR 98/11 'Income tax: documentation and practical issues associated with setting and reviewing transfer pricing in international dealings' 5.39.
[12] Treasury Regs. § § 1.482-1 through –6. s1.482-8 (the final regulations, T.D.8552. Fed. Reg. 34971, 7/8/94). The final regulations are generally effective for tax years beginning after Oct.6, 1994.
[13] Chip W W, 'Organization for Economic Cooperation and Development Guidelines' in Feinschreiber R (ed) Transfer Pricing Handbook (3rd ed, 2001) John Wiley & Sons Inc New York, p. 33-3.
[14] Treasury Regs. § 1.482-1(c).
[15] Treasury Regs. § 1.482-1(c) (1).
[16] Ibid.
[17] In the US the Comparable Uncontrolled Transaction Method, the Comparable Profits Method, the Comparable Profit Split Method and the Residual Profit Split Method can all be applied to transactions involving intangible assets.
[18] Treasury Regs. § 1.482-4(d).
[19] A functional analysis involves the identification and evaluation of the functions performed, assets used and risks and responsibilities assumed by the controlled and uncontrolled parties involved in the transactions under review.
[20] OECD Guidelines 1.69.
[21] Ibid.
[22] Ibid 1.68.
[23] Ibid.
[24] Ibid 1.15.
[25] Chip, above n 14, p. 33-2.
[26] TR 97/20 'Income tax: arm's length transfer pricing methodologies for international dealings’ 3.5.
[27] ATO publication International Transfer Pricing - Advance Pricing Arrangements (APAs) (1998) Canberra, 3.
[28] TR 97/20, above n 27, 1.23.
[29] TR 94/14 'Income tax: application of Division 13 of Part III (international profit shifting) - some basic concepts underlying the operation of Division 13 and some circumstances in which section 136 AD will be applied' 86.
[30] For example the US Treasury Regulations give the example of a US corporation developing bulletproof material for use in protective clothing and headgear (Nulon) and licensing its European subsidiary to manufacture and market Nulon: see Treasury Regulations Section 1.482-6(c) (3)(D)(iii).
[31] Anderson P & Heath M, ' Comparative Survey: Australia: Practical Application of Transactional Profit Methods' Sept/Oct 2000) 7 International Transfer Pricing Journal 176, 176.
[32] OECD Guidelines 3.1.
[33] Stanley G, 'Transfer pricing takes centre stage' Oct 2001 12 International Tax Review London 25, 26.
[34] The OECD issued a notice entitled 'Transfer Pricing: The OECD launches an invitation to comment on comparability issues', with a deadline for comments being set at 5 September 2003. An invitation to comment on issues related to profit issues was to have been issued later that year.
[35] Setchell M, PricewaterhouseCoopers UK 'OECD Transfer Pricing Guidelines Likely to Be Modified, Official Says' (April 12, 2002) Tax Notes International, Tax Analysts.
[36] Stanley, above n 34, 26.
[37] The Preamble to the 1994 Final Transfer Pricing Regulations, § 1.482-6.
[38] Nolan J, ‘U.S. Final Transfer Pricing Regulations' (April 1996) 50 University of Miami Law Review 537, 566.
[39] Treasury Regs. § 1.482-6 (c) (2) (i).
[40] Ibid § 1.482-6 (c) (2) (ii) (B) (1).
[41] Treasury Regs. § 1.482-6 (c) (2) (ii) (C) and (D).
[42] See for example Sherwood S G & Larson C, 'Has the Dust Settled with the Final U.S. Transfer Pricing Rules' (October 1994) 21 Tax Planning International Review 3, 12-13, and Raby N of Coopers & Lybrand L.L.P., Washington DC, 'United States' (April 1997) International Tax Review Supplement London 93, 100.
[43] Rozek R P & Korenko G G, 'Transfer Prices for the Intangible Property Embodied in Products With Extraordinary Profit Potentials' (18 October 1999) 19 Tax Notes International 1553, 1556.
[44] Treasury Regs. § 1.482-6(c) (3) (i).
[45] Ibid § 1.482-6(c) (3) (i) (A).
[46] Ibid § 1.482-6(c) (3) (i) (B).
[47] The Preamble to the 1994 Final Transfer Pricing Regulations, § 1.482-6.
[48] Barrett R, Blum K & O'Connor S, 'US: Intangible property & APAs' 2002 International Tax Review London 27, 32.
[49] Treasury Regs. § 1.482-6(c) (3) (iii).
[50] Ibid § 1.482-6(c) (3) (ii) (C).
[51] Ibid § 1.482-6(c) (3) (ii) (D).
[52] See: Finan W F ‘Comments on the OECD Draft Report: Intangible Property' (October 1995) 2 International Transfer Pricing Journal 5-12.
[53] OECD Guidelines 3.2.
[54] Ibid. 3.5.
[55] Ibid.
[56] Ibid 3.6.
[57] Ibid 3.7.
[58] Ibid 3.8.
[59] Ibid 3.9.
[60] Ibid 3.11.
[61] Ibid 3.16.
[62] Ibid 3.17.
[63] Ibid 3.18.
[64] Ibid 3.19.
[65] Ibid 3.25.
[66] Ibid 6.26.
[67] 'OECD draft transfer pricing guidelines' (Jul/Aug 1995) 47 Tax Executive 316, 317.
[68] TR 97/20, above n 27, 3.52.
[69] Ibid 3.61.
[70] Ibid.
[71] Anderson & Heath, above n 32, 177.
[72] Ibid 180.
[73] TR 97/20, above n 27, 3.71.
[74] Ibid.
[75] Elliott J, 'Transfer Pricing: Lessons from Australia', Discussion Papers in the Department of Management Number 97-135 University of Southampton, 1997 8.
[76] Lebowitz B E, 'Transfer Pricing and the End of International Taxation' (September 27,
[1999] ) Tax Notes International 1201, 1205.
[77] Hammer R M & Feinschreiber R 'Profit Split Methodologies' in Feinschreiber R, above n 14 p.47-28.
[78] OECD Guidelines 2.49.
[79] Ibid.
[80] TR 94/14, above n 22, 87 (e).
[81] TR 97/20, above n 27, 2.24.
[82] Ibid 1.24.
[83] TR 98/11 'Income tax: documentation and practical issues associated with setting and reviewing transfer pricing in international dealings' 10.10.
[84] Killaly J, First Assistant Commissioner International Tax Division, ATO 'Profit Shifting After the Initial Rulings - Myths and Formulae for Allocation Profits and Expenses of Transnational Companies' ATAX Intensive Weekend Workshop Bowral, NSW, 27 to 29 May 1994, 28.
[85] TR 97/20, above n 27, 3.52.
[86] Comments received by the OECD Steering Group by the 5 September 2003 deadline: Contribution received from Ernst & Young, 1-2.
[87] Hammer R M & Feinschreiber R, above n 78, p. 47-4.
[88] OECD Guidelines 6.26.
[89] TR 94/14, above n 30, 349.
[90] Setchell, above n 36.
[91] Killaly, above n 85, 24.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UWSLawRw/2004/3.html